Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Sep 22;170(9):e70553. doi: 10.1111/jnc.70553

Multifaceted Immune Effects of the Scavenger Receptor CD36 on Long‐Term Outcomes After Neonatal Arterial Focal Stroke

Elena Di Martino 1, Aditya Rayasam 1, Amin Mottahedin 2, Joel Faustino 1, Shivani Mahuvakar 1, Joakim Ek 2, Carina Mallard 2, Zinaida S Vexler 1,✉
PMCID: PMC13595489  PMID: 42770491

ABSTRACT

The scavenger receptor CD36 is a versatile receptor that regulates various inflammatory processes by interacting with different ligands in concert with other receptors in a context‐ and cell‐type‐specific manner. While CD36 is injurious in acute adult stroke, it is beneficial in acute perinatal stroke. This differing acute response may be influenced by the maturation‐dependent availability of CD36 receptor partners in the brain, such as Toll‐like receptors (TLRs), the developmental state of the blood–brain barrier, and leukocyte subtypes during injury. Here, we further explore CD36‐TLR2 interactions in neonatal brain injury by examining whether CD36 similarly affects gene expression and leukocyte trafficking via the blood‐CSF barrier/choroid plexus after transient middle cerebral artery occlusion and direct TLR2 stimulation in wild type and CD36 knockout mice. We then examine the consequences of CD36 deficiency for long‐term injury outcomes after neonatal stroke. We demonstrate that CD36 deficiency affects long‐term behavior performance under physiological conditions and a shift in the role of CD36 from beneficial during acute injury to detrimental in the chronic phase following neonatal stroke. We further show that CD36 contributes to sustained immune responses, as indicated by long‐term alterations in microglial phenotypes, and influences long‐term functional outcomes. Collectively, these findings underscore the complex time‐dependent roles of CD36 and provide insight into the mechanisms by which it modulates ischemic injury in neonatal mice.

graphic file with name JNC-170-e70553-g004.webp

Keywords: choroid plexus, leukocytes, MCA occlusion, neuroinflammation, perinatal stroke, TLR2

In neonatal mice, both stroke and direct TLR2 stimulation elicit pronounced but partially distinct CD36‐dependent gene expression profiles in the choroid plexus. CD36 mediates leukocyte trafficking through the choroid plexus and immune cell presence in the brain parenchyma after direct TLR2 stimulation and stroke in a context‐dependent manner. While CD36 protects during acute stroke, it is detrimental during the chronic stroke phase. CD36 deficiency affects long‐term functional outcomes. CD36 deficiency mediates sustained immune responses.

graphic file with name JNC-170-e70553-g005.webp

Abbreviations

BBB

blood–brain barrier

BCSFB

blood–cerebrospinal fluid barrier

BrdU

5‐bromo‐2′‐deoxyuridine

CC

corpus callosum

CC1

anti‐APC clone CC‐1

CD36/45/11b/68

cluster of differentiation 36/45/11b/68

CP

choroid plexus

CSF

cerebrospinal fluid

DCX

doublecortin

DG

dentate gyrus

ECM

extracellular matrix

GCL

granular cell layer

Glut1

glucose transporter protein type 1

Iba1

ionized calcium‐binding adapter molecule 1

KO

knock out

MBP

myelin basic protein

NeuN

Neuronal Nuclei

NLRP3

NLR family pyrin domain containing 3

Olig2

oligodendrocyte transcription factor 2

PAM

Pam3CSK4

PPARγ

peroxisome proliferator‐activated receptor gamma

ROS

reactive oxygen species

SVZ

subventricular zone

TLR2/4

toll‐like receptor 2/4

tMCAO

transient middle cerebral artery occlusion (note that in the text sometimes it is only written MCAO)

WT

wild‐type

1. Introduction

CD36 is a pleiotropic receptor that mediates signaling of both innate and adaptive immune cells (Chen et al. 2022). It plays a role in the pathogenesis of several diseases including atherosclerosis, cancer, neurovascular dysfunction, and cerebral ischemia. Consequently, it is regarded as a target for intervention in neurodegenerative diseases, stroke, and brain trauma in adults. CD36 mediates multiple biological functions in endothelial cells, microglia/macrophages and monocytes, such as uptake of long chain fatty acids and oxLDL, phagocytosis of apoptotic debris by microglia/macrophages, cell chemotaxis and regulation of metabolic pathways (Chen et al. 2022; Febbraio et al. 2001; Primo et al. 2005; van Berkel et al. 2005). It mediates the recognition and clearance of pathogens and apoptotic cells (Griffiths et al. 2010). CD36 acts as a key regulator in lipid signaling, inflammation, and ROS production under stress by interacting with various lipid receptors, including TLR2/4/6 (Abe et al. 2010). CD36 deficiency has been found to decrease chronic blood–brain barrier (BBB) dysfunction and scar formation, as well as improve memory deficits in a preclinical model of adult ischemic stroke (Balkaya et al. 2021). Lack of CD36 also offers protection in a contusion injury model by mitigating white matter loss and lowering lipid oxidation (Myers et al. 2014). Furthermore, CD36 has been identified as having a negative impact on angiogenesis (Jimenez et al. 2000). Although its detrimental role in adult stroke is well documented (Cho et al. 2005; Woo et al. 2016), recent studies have demonstrated that its upregulation through the PPAR γ pathway can be beneficial following intracerebral hemorrhage by enhancing microglia‐mediated phagocytosis (Zhao et al. 2009). Furthermore, during the resolution phase of stroke, CD36‐expressing monocytes and macrophages facilitate debris clearance via phagocytosis (Woo et al. 2016).

We previously established that genetic deletion of CD36 contributes to, rather than attenuates, early injury after transient middle cerebral artery occlusion (tMCAO) in neonatal mice (Woo et al. 2012). In this neonatal stroke model, significant leukocyte accumulation—including neutrophils, inflammatory monocytes, and surveilling monocytes—occurs at the choroid plexus (CP) ipsilateral to tMCAO (Rayasam et al. 2020). This accumulation is linked to rapid, substantial gene expression changes in the affected CP (Rayasam et al. 2022). Lack of CD36 reverts changes in genes involved in inflammatory, metabolic, and extracellular matrix (ECM) signaling compared to those in wild‐type (WT) neonatal mice after tMCAO and modulates the magnitude of neuroinflammation (Rayasam et al. 2022). Independently, using in vivo imaging of TLR2, one of the co‐receptors of CD36, combined with measures of neuroinflammation, we demonstrated the pathogenic role of TLR2 in neonatal stroke (Lalancette‐Hebert et al. 2017) and TLR2‐mediated leukocyte trafficking via the CP (Mottahedin et al. 2019).

In this study, we further investigated the interactions between CD36 and TLR2 in neonatal brain injury by assessing whether CD36 influences gene expression and leukocyte trafficking through the blood–cerebrospinal fluid barrier (BCSFB)/CP following tMCAO compared to direct TLR2 stimulation in neonatal WT and CD36 knockout (KO) mice. We then evaluated the impact of CD36 deficiency on long‐term outcomes after neonatal stroke. Our findings indicate a shift in the role of CD36 from being beneficial during the acute phase following neonatal stroke to becoming detrimental during the chronic injury period, as demonstrated by both histological and functional outcomes. Furthermore, we show that CD36 is involved in persistent immune modulation, as evidenced by sustained changes in microglial phenotypes observed after stroke during the neonatal period. Overall, this study underscores the various roles of CD36 at different stages following ischemic brain injury in neonatal mice.

2. Materials and Methods

2.1. Animals

All research conducted on animals was approved by the Institutional Animal Care and Use Committee of the University of California San Francisco (Protocol # AN183374), in accordance with the Guide for the Care and Use of Laboratory Animals (U.S. Department of Health and Human Services). The data follow the ARRIVE guidelines (Animal Research: Reporting in Vivo Experiments). Animals were housed with nesting material and shelters and kept in rooms with controlled temperature and light/dark cycles, and received ad libitum access to food and water. C57BL/6 wild type (WT) mice were originally purchased from Charles River and CD36 null mice (CD36KO; C57BL/6 background) were originally received from a colony established by Drs. Iadecola and Zhou. Animals were bred in house, and pups belonging to the same litters were randomized in different groups. Blinded data analysis was used where possible including litter block design and arbitrary assignment of mice within the same litter to experimental/control groups. Sample sizes were chosen based on previously published studies in tMCAO in neonatal mice (Fernandez‐Lopez et al. 2016; Fukuzaki et al. 2023; Li et al. 2015; Rayasam et al. 2020; Rayasam et al. 2022; Woo et al. 2012). Pups were excluded from the study if body weight was lower than 2 g prior to surgery, pups lost more than 15% of body weight within 48 h after tMCAO or had a blood clot by the occluder site. Mice were sacrificed with a lethal dose (> 200 mg/kg) of Na Pentobarbital and checked for pedal withdrawal reflex before proceeding with tissue collection, as outlined in the IACUC protocol. Figure 1 provides a detailed graphical timeline of the study.

FIGURE 1.

FIGURE 1

Open in a new tab

CD36‐TLR2 interaction induces similar and distinct patterns of gene expression in the CP in neonatal mice upon direct TLR2 stimulation and tMCAO in neonatal mice. (A, B) PCA plots of genes regulated in the CP at 6 h after PAM injection (A) or 3 h after reperfusion following a 3 h tMCAO (B). (C,D) Heatmaps of top 1000 significantly (ANOVA, q < 0.005) regulated genes by PAM (C) or tMCAO (D), replicated from reference 15 for comparison, and hierarchical clustering of genes and GO terms attributed to the clusters. (E) Heatmap of genes with opposite expression behavior in the CP of WT PAM and ipsilateral tMCAO in comparison to control and KO CPs. GO terms related to the upregulated (top) and downregulated (bottom) clusters are stated. (F) Example genes in sterol biosynthesis pathway (Cyp51) and DNA replication pathway (Msh6), upregulated or downregulated in the CP of WT PAM and ipsilateral tMCAO in comparison to control and KO CPs. Data shown as mean ± SD. Statistical significance was determined using one‐way ANOVA followed by Tukey's multiple comparisons test.

2.2. Transient Middle Cerebral Artery Occlusion (tMCAO)

Surgery was performed on mice of both sexes on postnatal day 9 (P9)–P10 as previously described (Fernandez‐Lopez et al. 2016; Woo et al. 2012). Briefly, pups were anesthetized with 3.5% isoflurane and maintained at 1%–2% in a gas mixture of 70% N2O and 30% oxygen as needed. The common carotid artery and internal carotid artery (ICA) were exposed via midline neck incision; a single 7–0 silk suture thread was used to temporarily tie a knot below the ICA origin and prevent retrograde bleeding; a coated 8–0 nylon suture was advanced 4–5 mm, and the skin sutured. The occluding filament was removed 3 h later, and gel foam was used to seal the arteriotomy. Throughout the whole surgical procedure, pups' body temperature was maintained with an overhead lamp and a temperature‐controlled blanket. In sham‐operated pups, the suture was inserted but not advanced. All animals undergoing tMCAO surgery were administered 0.25% bupivicaine (locally) at the time of surgery, as well as 3 doses of buprenorphine (0.05 mg/kg; i.p.) over 24 h following surgery. 5% dextrose in 0.9% saline (10 ul/g; i.p. per day) was administered on the day of surgery, the following day, and to animals losing weight compared to the previous day starting at 24 h after surgery. Animals were allowed to fully recover from anesthesia before returning to colony. Animal weights were taken prior to tMCAO and every 18–24 h following surgery.

Pam3CSK4 (PAM; 5 mg/kg body weight; Invivogen) was administered intraperitoneally to P9‐P10 C57BL/6 WT and CD36KO mice of both sexes (Rayasam et al. 2020).

2.3. Transcriptomics Analysis of the CP

The RNA extraction and RNA sequencing was performed as we described previously (Mottahedin et al. 2019; Rayasam et al. 2022). There was no mortality in PBS/PAM‐treated groups (13/18 WT and 15/15 CD36KO, respectively) and naïve and sham‐treated groups. Mortality was 3/34 WT and 3/24 CD36KO mice subjected to tMCAO. Briefly, the CP of mice administered PAM or subjected to tMCAO (separately from ipsilateral and contralateral side) was collected, placed in RNAlater (Ambion) for 24 h at 4°C to preserve the RNA, and, after a brief centrifuge to remove RNAlater, was stored at −80°C. Total RNA was extracted using Trizol (Qiazol; Qiagen) and miRNAeasy Kit (Qiagen) following manufacturer protocols. RNA quality was assessed using Experion Automated Electrophoresis System (Bio‐Rad) and 500 ng of RNA was used for library preparation using TruSeq Stranded mRNA and Total RNA Sample Preparation Kit with Rib‐Zero Gold (Illumina). The CPs from 5 mice were pooled together before RNA extraction. In sham‐operated mice, ipsilateral and contralateral CP were mixed and samples from 3 mice were pooled. The sequencing was performed on a NovaSeq 6000 (Illumina) with a read length of 2 × 100. Mouse assembly mm10 from UCSC was used as reference genome. The full list of the transcriptional effects is shown in Tables S1 and S2.

2.4. RNA Sequencing Data Analysis

Differential expression was obtained using DEseq2 with Benjamini‐Hochberg correction for multiple comparisons. Gene cluster analysis was performed using the Qlucore Omics Analyzer v.3.6 on top 1000 genes differentially regulated between groups (q < 0.005, F‐test and ANOVA). Enrichment analysis of gene clusters was performed using g:Profiler (Raudvere et al. 2019).

2.5. Opposite‐Behavior Gene‐Set Expression Analysis

To discover similar pathways regulated by PAM and tMCAO which were affected by CD36 deletion, we employed an approach analogous to bidirectional expression signatures used to contrast phenotypes in the Connectivity Map (Lamb et al. 2006; Subramanian et al. 2017). Expression values (normalized counts) were transformed to per‐gene z‐scores across all samples, and for each canonical group, we computed the group‐mean z for every gene. Two canonical groups were then compared: Group A = PBS‐WT, PBS‐KO, PAM‐KO, WT‐Sham‐Comb, KO‐Sham‐Comb, WT‐MCAO‐Contra, KO‐MCAO‐Contra, KO‐MCAO‐Ipsi and Group B = PAM‐WT, WT‐MCAO‐Ipsi. Thresholds were applied as z ≥ 0.3, and |B–A| ≥ 1.0. One‐way ANOVA across canonical groups was performed and genes with FDR q < 0.05 (Benjamini–Hochberg) were retained. A heatmap of z‐scores was then generated and hierarchical clustering was performed using Pearson correlation. Enrichment analysis of gene clusters was performed using g:Profiler.

2.6. Flow Cytometry

Single‐cell myelin‐free suspensions from PAM/PBS–treated or contralateral and injured regions after tMCAO were plated (5 × 105/96 well), centrifuged, pellet resuspended in 100ul blocking buffer containing CD16/32 (1:70, Biolegend) and incubated in 150 ul FACs staining buffer containing 2% FBS and evaluated on BD LSRII flow cytometer (BD Biosciences)., as we previously reported (Rayasam et al. 2022). Fluorescence Minus One (FMO) samples, a commonly used strategy to prevent false positive results through overlap of fluorophores (Aghaeepour et al. 2013), was applied. The utilized combinations of antibodies listed in Table 1 were diluted 1:200 in FACS staining buffer were incubated in Fixation and Permeabilization solution (100ul, 4°C, 20 min), incubated with antibody mixture (4°C, 30 min) and resuspended in staining buffer. Gating and data analysis were performed using FlowJo software (Tree Star).

TABLE 1.

Reagents used in the study.

Name RRDI
Rabbit anti Iba1 (1:200) RRID:AB_839504
Rat anti CD68 (1:200) RRID:AB_322219
Mouse anti Glut1 (1:200) RRID:AB_2190927
Rat anti BrdU (1:200) RRID:AB_305426
Mouse anti NeuN (1:200) RRID:AB_2298772
Rabbit anti DCX (1:400) RRID:AB_561007
Rabbit anti Olig2 (1:200) RRID:AB_776906
Mouse anti CC1 (1:100) RRID:AB_2057371
Mouse anti MBP (1:200) RRID:AB_2860023
Goat anti Rat 488 (1:500) RRID:AB_2534074
Goat anti Rabbit 488 (1:500) RRID:AB_2576217
Goat anti Rat 568 (1:500) RRID:AB_141874
Goat anti Mouse 568 (1:500) RRID:AB_144696
Goat anti Rabbit 647 (1:500) RRID:AB_2535813
Goat anti Mouse 647 (1:500) RRID:AB_2535805
IB4 (1:150) Thermo Fisher; cat. no. I32450
Mouse Ig Blocking Reagent RRID:AB_2336587
ProLong Gold Antifade Mountant Thermo, cat. no. P36930
DAPI (1:500) RRID:AB_2307445
Cresyl Violet Thermo Fisher; 405 760 025
Anti‐CD45‐Pacific Blue (1:200) RRID:AB_493536 Biolegend
Anti‐CD11b‐APC‐Cy7 (1:200) RRID:AB_830641 Biolegend
Ly6g (IA8)‐AF700 (1:200) RRID:AB_10640452 Biolegend
Ly6c (Hk1.4)‐APC (1:200) RRID:AB_1732087 Biolegend
Open in a new tab

2.7. Tissue Processing, Injury Quantification and Immunofluorescence

For histology, mice were perfused with PBS and 4% PFA, brains extracted, post‐fixed with 4% PFA overnight, cryoprotected with 30% sucrose, flash‐frozen, and coronally sectioned with a Leica cryostat (12 μm thick serial sections, 360 μm apart). For the 2‐week outcome, mice received BrdU injections intraperitoneally (50 mg/kg per day for 3 days). Injury extent was examined on 6–7 Nissl‐stained sections by two independent investigators blinded to the sample identity using Image J.

Immunofluorescence was performed on adjacent sections blocked in PBST (0.1% TritonX in PBS) containing 10%–20% normal goat serum and overnight incubation with primary antibodies at 4°C, followed by appropriate secondary antibodies (listed in Table 1). Secondary Abs (Invitrogen) were diluted 1:500. Z‐stack images (8 × 1μm slice‐depth) were acquired using a Zeiss Axiovert100 microscope (25× oil objective) in 2–3 fields of view per hemisphere/region of interest per section, in 2 sections per brain. Images were analyzed with Volocity Software (Improvision, Perkin Elmer) using automated protocols for signal intensity threshold (> 2SD background in each channel) in a 1 × 106μm3 voxels and results averaged per animal. Sections that were folded in the region of interest were not included in the analysis. Insufficiently perfused brains were excluded because of the autofluorescence affecting Volocity analysis.

2.8. Behavior Testing in P38‐P50 Mice

Out of mice designated to evaluate long‐term outcomes, 4/22 WT and 4/15 CD36KO that underwent tMCAO died either during MCAO or within 2 weeks after reperfusion. All behavior studies were conducted in the morning hours except when two sessions per day were required (Beam test and Rotarod test). Tests were performed blinded, and the experimenter was provided only animal numbers and the grouping, but not group identity. Mice were transferred to testing room 1 h prior to behavior testing to acclimate. Apparatuses were cleaned with Vimoba before and after testing and with 70% alcohol between trials.

2.8.1. Open Field (OF)

Mice were placed in the center of a clear acrylic chamber (41 × 41 × 30 cm) with two 16 × 16 photobeam arrays that automatically detect horizontal and vertical movements (Flex‐Field/Open Field Photobeam Activity System, San Diego Instruments, San Diego, CA) and were allowed to freely explore for 15 min. General spontaneous locomotor activity level, exploratory behaviors, and anxiety‐like responses were measured. Total movements in the outer periphery and center of the chamber, and their proportion were determined as the degree of baseline anxiety. Ambulatory (3 or more consecutive photobeams) and fine movements (breaking the same 2 beams repeatedly) were recorded. Rearing behavior was used as an indicator of exploratory activity directed toward the space outside the OF.

2.8.2. Gait Analysis

Mice were placed inside the DigiGait enclosure, the treadmill started at the initial speed of 200 (20.0 cm/s) and gait analysis was recorded during continuous running for up to 5 min. If a mouse was unable to keep up with the treadmill and was falling behind, it was gently guided to the front end of the walking compartment. Mice unable to continuously run for 5–10s were scored as incomplete. Symmetry, Stride Length (cm) for forepaw and hind paw and its frequency (steps/s), Stance width (cm) and duration (s) were analyzed.

2.8.3. Balance Beam Test

The test was completed over 3 days, 5 min/mouse per day in the Balance Beam apparatus that consists of an open platform on one side, a 19 × 19 × 19 cm square dark box with a 3.8 cm entrance hole on the opposite side, and connecting a 41 cm long beam of three widths, Large Beam (1.9 cm), Medium Beam (1.1 cm), and Small Beam (0.5 cm). On Day 1, as pre‐training, two guided trials using Large Beam were performed with 10–15 min intervals in between. In the 1st trial, mice were placed on the middle of the beam facing the dark box and were guided to the dark enclosure. In the 2nd trial, mice were placed on the end of the beam facing the dark box and were guided to the dark enclosure. Test Session 1 was conducted 10–15 min after pre‐training by placing mice on the end of the beam facing the dark box and traversing the beam, and leaving them unguided on the beam for a maximum of 60 s. On Day 2 (Test Session 2), mice were subjected to unguided trials with a maximum latency of 60 s using Medium Beam, with 10–15 min between trials. On Day 3 (Test Session 3), three unguided trials with 10‐15 min between trials using Small Beam were conducted in mice placed on the end of the beam facing the dark box and traverse the beam for a maximum of 60 s.

2.8.4. Inverted Grid Hang Wire Test

Mouse was placed with all four paws on a piece of 10 × 14 cm wire grid, the grid gently rocked back and forth to encourage the mouse to grasp the wire with all paws and then the grid was carefully inverted so that the mouse was hanging over a diaper pad or stack of paper towels. If a mouse fell off within 10 s, it was put back on the grid immediately to redo the trial. If the mouse fell off in less than 10 s again, the time was recorded. The maximum latency was measured for a maximum of 180 s in a total of 3 trials, ~1 h between each trial.

2.8.5. Rotarod

Up to five mice of the same sex were simultaneously placed on the rotarod apparatus with the rod rotating at an accelerated speed, from 4 rpm to 40 rpm. The rotation speed increases by 4 rpm every 30 s. The trial ends when the mouse falls off the rod or when 5 min has elapsed. The mice are given 2 sessions of 3 trials each, with a 15–20 min inter‐trial interval.

2.9. Statistical Analysis

Randomization for tMCAO/sham was used including litter block design and arbitrary assignment of mice within same litter to experimental/control groups and unsexed data were analyzed with GraphPad Prism 10 software. Data are presented as mean ± SD. Outliers were checked with ROUT (Q = 1%). Comparisons between two groups were evaluated by Student t‐test or Mann–Whitney based on normality distribution (Shapiro–Wilk test). 2‐way ANOVA with Tukey's multiple comparisons test was used to compare four groups. 2‐way ANOVA with Šídák's multiple comparisons test was used for repeated measurements. Differences were considered significant at p < 0.05. Additional details on assessment of the normality of data and statistical test/p‐value are provided in Tables S3 and S4.

3. Results

3.1. Both Direct TLR2 Stimulation and tMCAO Elicit Pronounced but Partially Distinct Gene Expression Profiles in the CP of Neonatal WT and CD36 KO Mice

To uncover common mechanisms of TLR2‐CD36 control of leukocyte trafficking via CP following tMCAO and TLR2 agonist Pam3CSK4 (PAM) administration in neonatal mice, we first evaluated the transcriptional effects of these stressors (Table S1). PCA plots demonstrate that both direct TLR2 stimulation (Figure 1A) and tMCAO (Figure 1B) led to rapid and significant gene upregulation in WT mice compared to their respective controls 6 h after the stressor. The lack of CD36 mitigated changes in the CP ipsilateral to injury and after PAM administration. Analyses of the top 1000 regulated genes following PAM administration and tMCAO detail classes of significantly modulated genes, as evident from the heat maps. Gene clusters associated with translation, mitochondrial function and response to peptide hormone were mainly influenced by the presence of CD36 following PAM (Figure 1C). In the tMCAO setting, CD36 also impacted DNA repair, inflammatory pathways, sterol biosynthesis and ECM pathways, as we reported previously (Rayasam et al. 2022) (Figure 1D).

Employing an approach analogous to Connectivity Map approach (Lamb et al. 2006; Subramanian et al. 2017) revealed that one set of genes was significantly upregulated in the CP by PAM and ipsilateral to MCAO in WT groups, but remained constant among control and KO pups (Figure 1E). GO term analysis revealed that these genes are involved in sterol and cholesterol biosynthesis, among which is sterol 14α‐demethylase Cyp51 (Figure 1F and Table S2). Another set of genes downregulated in WT PAM and ipsilateral MCAO groups was related to DNA replication, among which is the DNA mismatch repair gene Msh6 (Figure 1F and Table S2). These findings indicate that gene expression related to CD36 in the CP is complex and context‐dependent, and that CD36 plays a key role in facilitating leukocyte migration through the CP, possibly by regulating sterol or cholesterol homeostasis.

3.2. TLR2‐Induced CD36‐Mediated Leukocyte Trafficking Through the CP and Immune Cell Presence in the Brain Parenchyma Are Context‐Dependent in Neonatal PAM and tMCAO Models

We previously reported that CD36 influences both the extent and type of leukocyte migration through the CP after tMCAO by demonstrating that CD36 deficiency reduces myeloid cell infiltration in the CP post‐MCAO at both early (3 h) and later (13 h) reperfusion stages (Rayasam et al. 2022). Considering that the spatio‐temporal pattern of leukocyte appearance in injured regions after stroke is very dynamic (Garcia‐Bonilla et al. 2016) and can depend not only on stroke severity but on the state of brain maturation at the time of insult as well, we characterized cells of the monocyte lineage in neonatal mice after tMCAO and after TLR2 activation, including the microglial and monocyte populations, by non‐invasive imaging (Lalancette‐Hebert et al. 2017), immunofluorescence (Rayasam et al. 2020), and flow cytometry (Mottahedin et al. 2019; Rayasam et al. 2020; Rayasam et al. 2022). We demonstrated accumulation of toxic CD45+CD11b+Ly6ChighCX3CR1intCCR2highLy6Gneg and patrolling CD45+CD11b+Ly6CintCX3CR1highCCR2negLy6Gneg monocytes upon injury (Rayasam et al. 2020). In this study, we compared effects of previously reported CD36 deletion in a neonatal stroke model (Rayasam et al. 2022) with responses triggered by direct TLR2 stimulation. Figure 2 provides side‐by‐side comparisons of the magnitude and the dynamics of immune cell accumulation 6 h and 16 h after tMCAO (Rayasam et al. 2022) and PAM administration (current study). Flow cytometry in isolated CPs showed that in both WT and KO mice, the percentage of CD45+CD11b+ myeloid cells in the CP following PAM was significantly increased compared to their respective PBS controls, but significantly fewer CD45+CD11b+ myeloid cells were apparent in the PAM CP of KO compared to WT mice 6 h after stimulation (Figure 2B). The overall number of CD45+CD11b+ cells was reduced by 16 h following PAM in both WT and KO mice (Figure 2B). Here we used Ly6Chigh and Ly6Cint to distinguish toxic and patrolling monocytes based on previous characterization of toxic monocytes as CD45+CD11b+Ly6ChighCX3CR1intCCR2highLy6Gneg and patrolling cells as CD45+CD11b+Ly6CintCX3CR1highCCR2negLy6Gneg (Rayasam et al. 2020). PAM increased numbers of CD45+CD11b+Ly6Chi inflammatory monocytes (Figure 2C) and CD45+11b+Ly6cintLy6G+ neutrophils (Figure 2D) in the CPs of both WT and KO mice, yet by 16 h, in both WT and KO mice, the influence of genotype and PAM on inflammatory monocyte or neutrophil accumulation was not as evident (Figure 2C,D). Interestingly, the lack of CD36 increased the numbers of CD45+11b+Ly6CintLy6G− patrolling monocytes at 6 h and 16 h after PAM (Figure 2E), indicating a delayed dynamic of accumulation of these beneficial cells in the KO. Cumulatively, these data demonstrate that genetic deletion of CD36 restricts and delays the overall presence of both toxic and beneficial myeloid cells in the CP early after PAM administration.

FIGURE 2.

FIGURE 2

Open in a new tab

The patterns of TLR2‐mediated leukocyte trafficking in the CP and accumulation in the cortex depend on the CD36 after direct TLR2 stimulation and tMCAO. (A) Representative gating of immune cells in the CP of PBS and PAM‐injected mice 6 h after administration. (B–E) Quantification of CD45+CD11b+ myeloid cells (B), CD45+CD11b+Ly6Chi inflammatory monocytes (C), CD45+11b+Ly6CintLy6G+ neutrophils (D), and CD45+11b+Ly6CintLy6G+ patrolling monocytes (E) in WT and CD36 KO CPs 6 h and 16 h after PAM injection or 3 h tMCAO followed by 3 h or 13 h reperfusion. (F–I) Quantification of CD45+CD11b+ cells in the cortex of WT and CD36 KO PBS/PAM‐injected mice and mice that underwent tMCAO (F), CD45+CD11b+Ly6Chi cells (G), CD45+11b+Ly6CintLy6G+ neutrophils (H), and CD45+11b+Ly6CintLy6G− patrolling monocytes (I). Two‐way ANOVA with post hoc Tukey's Multiple Comparison test was performed to compare groups with multiple independent variables. Data shown as mean ± SD. Each dot represents data from an individual mouse. The number of mice in individual group are: MCAO 3 h/3 h: WT n = 6, KO: N = 9; MCAO 3 h/13 h WT n = 13, KO n = 9; PAM 6 h WT n = 12, PAM 6 h KO n = 6; PAM 16 h WT n = 11, PAM 16 h KO n = 4, PBS 6 h WT n = 12, PBS 6 h KO n = 6, PBS 16 h WT n = 6, PBS 16 h KO n = 4. To more comprehensively show the relative magnitude of changes the figure shows effects of both previously reported CD36 deletion in a neonatal stroke model (Rayasam et al. 2022) and effects of PAM administration. Statistical values are indicated on individual panels.

In the cortex, our previous results demonstrated that KO neonatal mice have blunted and delayed accumulation of immune cells after tMCAO. Here, PAM administration significantly increased accumulation of CD45hiCD11b+ cells in WT mice compared to KO mice at 6 and 16 h (Figure 2F). Analysis of cell phenotypes showed significantly more CD45hiCD11b+Ly6Chi inflammatory monocytes and/or activated microglia in WT than in KO mice at 6 h but not 16 h (Figure 2G). Lastly, PAM‐treated KO mice had less CD45hi11b+Ly6CintLy6G+ neutrophils but more CD45hi11b+Ly6CintLy6G− cells at 16 h in the cortex (Figure 2H,I); the latter can be microglial cells and patrolling monocytes. Overall, these results demonstrate that in neonatal mice CD36 deficiency distinctly modifies the myeloid cell accumulation profile within the cortex following PAM, as compared to tMCAO.

3.3. The Lack of CD36 Protects From Chronic Injury Phase After tMCAO in Neonatal Mice

Compared to the significantly higher incidence of severe histologic outcome in KO compared to that in WT pups 24 h after a 3 h MCAO, due in part to reduced removal of neurons dying by caspase‐3 dependent apoptosis acutely (Woo et al. 2012), at 7 days after tMCAO, the overall tissue loss was significantly lower in KO compared to WT mice (Figure 3A), including tissue loss in the corpus callosum (CC), granular cell layer (GCL), and hippocampus (Figure 3B). Unilateral ventricular enlargement was apparent in both groups but tended to be smaller in KO than in WT pups (0.095; Figure 3C). Volumes of the hemisphere contralateral to the occlusion were similar in WT and KO groups.

FIGURE 3.

FIGURE 3

Open in a new tab

Lack of CD36 attenuates injury at 1 and 2 weeks and affects neurogenesis and endothelial proliferation 2 weeks after neonatal tMCAO. (A–C) Volumetric size of injured vs. contralateral hemisphere 1 week after tMCAO (A), including regional changes in the corpus callosum (CC), caudate, granular cell layer (GCL) and hippocampus (hippo) (B), and ventricle enlargement (C) in WT and KO mice 1 week after tMCAO. (D–F) Volumetric size of injured vs. contralateral hemisphere 2 weeks after tMCAO (D), including regional changes in the corpus callosum (CC), caudate, granular cell layer (GCL) and hippocampus (hippo) (E), and ventricle enlargement (F) in WT and KO mice 2 weeks after tMCAO. (G, H) A representative image of anatomical representation of the scar region (G) and subventricular zone (SVZ) (H) for histological quantification of proliferating cells in the SVZ (I), and endothelial volume (J) and proliferation (K) in the scar region in WT and KO mice 2 weeks after tMCAO. Data shown as mean ± SD. Each dot represents data from an individual mouse. The number of mice in individual groups are: WT: N = 5–10; KO: N = 7–9. Comparisons between two groups were evaluated by Student t‐test or Mann–Whitney based on normality distribution. Statistical values are indicated on individual panels.

CD36 deficiency attenuated chronic tissue loss 14 days after injury, including the overall tissue loss (Figure 3D) and loss in individual brain regions (Figure 3E). Unilateral ventricular enlargement was similar in both groups at this point (Figure 3F). No differences were observed in injury volume between male and female pups (Figure S2A).

Knowing that CD36 can be anti‐angiogenic, we next evaluated vascular coverage and endothelial proliferation in the same brains at 14 days. Representative examples of images are shown in Figure 3G,H. Evaluation of angiogenesis in the scar cortical regions, which was prominent, showed increased volume of Glut‐1+ vessels and endothelial proliferation (BrdU+/Glut‐1+ cells) ipsilateral to tMCAO, yet similar magnitude in WT and KO pups (Figure 3I–K).

To define the effects of genetic deletion of CD36 on myelination and white matter, we analyzed cell proliferation and immature/mature oligodendrocytes in the CC in the same brains (Figure S2B–G). Compared to the contralateral hemisphere, the number of BrdU+ cells was significantly increased in the ipsilateral CC of WT mice and, to a similar extent, in KO mice (Figure S2C). The number of newly generated BrdU+/Olig2+ oligodendrocytes was also increased in the ipsilateral CC of WT mice compared to the contralateral hemisphere, as were the numbers of Olig2+, CC1+, and Olig2+/CC1+ cells, while no difference was observed in KO mice (Figure S2E–G). Together, these results demonstrate that CD36 can in part be protective during the chronic phase of ischemic injury in neonatal mice by affecting oligodendrocytes.

Unilateral increase in cell proliferation was observed in the striatum, DG/GCL, and SVZ ipsilateral to tMCAO in WT pups (Figure S2H). Compared to the contralateral hemisphere, the presence of doublecortin (DCX+) neurons was reduced in ipsilateral SVZ but increased in the ipsilateral striatum and GCL of WT mice (Figure S2I). Furthermore, a higher DCX+ number was observed in the ipsilateral DG/GCL of WT compared to KO mice. To better define the timing of neurogenesis after tMCAO, we determined the number of BrdU+/DCX+ cells. Neurogenesis was enhanced in the ipsilateral striatum and hippocampus of both WT and KO mice compared to the contralateral hemisphere (Figure S2J).

3.4. Long‐Term Effects of Genetic CD36 Deletion on Functional and Histological Outcomes of tMCAO in Neonatal Mice

Multiple behavior tests to comprehensively examine the long‐term effects of genetic deletion of CD36 on outcomes of stroke during the neonatal period revealed test‐specific effects in P38‐P50 mice under both physiological conditions (i.e., naïve mice) and after stroke. There was no significant difference between naïve and sham‐operated mice; these subgroups were combined. In the Rotarod test, consisting of six sessions with accelerating speed, compared to naïve/sham‐operated mice, no significant differences in performance were observed between naïve KO and WT mice. Both WT and KO subjected to tMCAO exhibited a significant increase in latency to fall between trials 1 and 3, a reduction in latency between trials 3 and 4, and a subsequent increase from trials 4 to 6 (Figure 4A,B).

FIGURE 4.

FIGURE 4

Open in a new tab

CD36 deficiency affects behavioral performance in naïve/sham mice and is associated with greater tissue preservation after neonatal tMCAO. Behavior outcomes 6 weeks after tMCAO. (A, B) Rotarod test: Latency to fall on accelerating platform in WT (A) and KO (B) mice. (C) Open field test: Total movement (left) and rearing frequency (right). (D) Wire test: Paw strength. (E–G) Histological outcome 6 weeks after tMCAO. Volumetric size of injured vs. contralateral hemisphere (E), sizes of individual regions including the corpus callosum (CC), the caudate, and the hippocampus (F) and ventricle enlargement in WT and KO mice (G). Data shown as mean ± SD. Each dot represents data from an individual mouse. The number of mice in individual groups are: Rotarod and Open field: MCAO+WT n = 14; MCAO+KO n = 16; Sham+WT n = 24; Sham+KO n = 14; Wire MCAO+WT n = 22; MCAO+KO n = 16; Sham+WT n = 24; Sham+KO n = 14 and histology WT: N = 14; KO: N = 12. Comparisons between two groups were evaluated by Student t‐test or Mann–Whitney based on normality distribution. Comparison between four groups were evaluated by 2‐way ANOVA with Tukey's multiple comparisons test, or 2‐way ANOVA with Šídák's multiple comparisons test for repeated measurements. Statistical values are indicated on individual panels.

In the Open Field test, spontaneous locomotor activity and total movement were significantly higher in naïve/sham WT than in KO mice (Figure 4C). In WT mice, these measures increased further following tMCAO, whereas KO mice showed no significant change compared to their respective controls (Figure 4C). Time spent in the outer periphery and center of the chamber, and their proportion as the degree of baseline anxiety‐like responses were similar in naïve/sham WT and KO, and unchanged by tMCAO (Figure S3A,B). Rearing behavior as an indicator of exploratory activity showed significantly higher frequency in control WT and a further significant increase in WT pups that underwent tMCAO than respective KO (Figure 4C).

In the inverted Wire Hang test, control WT mice demonstrated significantly longer hanging time than control KO mice, but performance was similar in both groups that underwent tMCAO (Figure 4D).

Comparisons of individual gate analysis test parameters between WT and KO, including gait symmetry, stride frequency, forepaw or hindpaw width or length, showed similar outcomes and no sex differences in naïve/sham mice and in mice subjected to tMCAO (Figure S3C). WT mice subjected to tMCAO had significantly altered stride length for both fore‐ and hindpaws but there were no significant differences between groups (Figure S3C).

The beam test in naïve/sham WT mice demonstrated a longer latency during the initial trial on the small beam compared to both the large and medium beams, with performance improvement over two subsequent trials (Figure S3D). Naïve/sham KO mice, in contrast, maintained stability on large and medium beams. WT mice subjected to tMCAO exhibited significantly increased latency to reach the dark box. WT mice that underwent tMCAO showed a marked reduction in their ability to reach the dark box while crossing large and medium‐sized beams; their performance worsened further on the first attempt using the small beam, but improved over two consecutive tests (Figure S3D). Despite exhibiting more footslips during their first attempt on the small beam, stability returned in the next two trials (Figure S3D). MCAO markedly adversely affected the ability of mice to remain on beams of all sizes (Figure S3D). The numbers of footslips in naïve/sham WT and KO were similar, whereas the number of footslips was significantly increased in the injured WT group but remained at the level of controls in the KO group (Figure S3D).

Histological analysis at 6 weeks revealed significantly better‐preserved tissue in KO compared to WT mice (Figure 4E), along with significant preservation of tissue in the CC, the caudate and the hippocampus (Figure 4F). While ventricles continued to be enlarged on the side of tMCAO, there was no significant difference between groups (Figure 4G). Volume of MBP was similar in both groups (Figure 5A–C). Vessel coverage was significantly higher in the CC of WT than KO mice but was not significantly different between groups in the caudate and cortex (Figure 5D–F). Overall, these data comprehensively define the long‐term functional and histological outcomes of CD36 deficiency following neonatal stroke.

FIGURE 5.

FIGURE 5

Open in a new tab

Lack of CD36 does not affect brain myelination 6 weeks after neonatal tMCAO. Representative image of myelin basic protein (MBP) staining (A). MBP+ fiber density quantification in different brain areas (B, C). (D–F) Endothelial volume in the corpus callosum (CC), caudate and cortex of WT and KO. Dots represent individual mice. Data shown as mean ± SD. Each dot represents data from an individual mouse. The number of mice in individual groups are: WT: N = 8; KO: N = 6. Comparisons between two groups were evaluated by 2‐way ANOVA with Tukey's multiple comparisons test to compare four groups (B–F). Statistical values are indicated on individual panels.

3.5. Long‐Lasting Effects of Genetic Deletion of CD36 Receptor on Immune Mechanisms

Based on the established function of CD36 in monocyte lineage cells, we evaluated the density and activation status of Iba1+ microglia/macrophages 6 weeks after tMCAO. Absence of CD36 did not influence the density of Iba1+ microglia/macrophages in the contralateral cortex (Figure 6A,B). However, Iba1+ cell density was increased in the ipsilateral hemisphere of WT mice, an effect not observed in KO mice. The lack of CD36 also modified the phenotypes of Iba1+ cells in the contralateral cortex based on the presence of the lysosomal marker CD68 on Iba1+ cells in KO mice by (Figure 6A) relative to both contralateral and ipsilateral cortices of WT mice, as quantified in Figure 6C,D. Although the number of Iba1+ cells associated with vessels was not affected by injury or lack of CD36, KO mice show a higher number of CD68+ microglia/macrophages in vessel proximity (Figure 6E,F). These results suggest that CD36 may have long‐term effects on the function of immune cells.

FIGURE 6.

FIGURE 6

Open in a new tab

Lack of CD36 affects microglial phenotypes 6 weeks after tMCAO. (A) Representative images of microglial cells/macrophages in the injured and control cortex. Thin white arrows point to dotted/punctate appearance of CD68+ on Iba1+ cells. White arrowheads point to Iba1+ cells with extensive CD68 expression. (B–D) Quantification of Iba1+ cells (B), CD68+ cells (C), and Iba1+/CD68+ cells (D). (E, F) Iba1+ cells (E) and CD68+ cells (F) touching Glut1+ endothelial cells in WT and KO mice 6 weeks after tMCAO. Data shown as mean ± SD. The data shown as cells/mm2 per mouse. Each dot represents data from an individual mouse. The number of mice in individual groups are: WT: N = 8; KO: N = 6. Two‐way ANOVA with Tukey's multiple comparisons test. Statistical values are indicated on individual panels.

4. Discussion

This study further advances the understanding of multiple CD36‐dependent mechanisms of acute injury in neonatal stroke and, for the first time, demonstrates the impact of CD36 on long‐term outcomes. Comparative gene expression analysis in neonatal WT and KO mice after tMCAO and direct TLR2 stimulation shows robust context‐dependent changes at the CP and CD36–TLR2–dependent leukocyte trafficking through the CP. Our results show that in contrast to its previously demonstrated beneficial effects in the acute phase of neonatal stroke, CD36 contributes to tissue loss at later stages, suggesting a shift from beneficial to a detrimental role over time. Behavioral assessments indicate that CD36 deficiency itself influences motor function, paw strength, and exploratory activity during physiological development. Surprisingly, tMCAO does not further alter balance, paw strength, fine movement, or anxiety at late stage. We also highlight that CD36 affects chronic injury partly through its impact on the vasculature in a region‐specific manner, without modifying white matter coverage. Finally, CD36 mediates long‐lasting effects on the immune response after injury, as suggested by sustained microglial activation in CD36 deficient mice.

CD36 scavenger receptor signaling is complex and varies depending on its receptor and ligand partners in the lipid fraction (Abe et al. 2010; Jimenez et al. 2000), with unique biological functions across cell types including endothelial cells, microglia/macrophages, and monocytes, functions that are not necessarily synchronized in different cell types. In adult stroke, CD36 promotes early injury via inflammatory and ROS pathways (Cho et al. 2005; Woo et al. 2016), whereas in neonates, its deletion worsens acute injury, likely due to impaired phagocytic clearance of apoptotic debris and resultant increased secondary inflammation (Woo et al. 2012). The latter was evident from a similar extent of cytotoxic brain edema during MCAO in WT and KO pups on diffusion‐weighted MRI (i.e., similar “tissue at risk” before reperfusion), but significantly higher incidence of severe injury in KO neonates at 24 h after reperfusion. A number of mechanisms accounted for distinct effects of CD36 in acute neonatal and adult stroke (Wang et al. 2007). First, consistent with the notion that phagocytosis is particularly important in post‐ischemic neonatal brain due to more prominent apoptotic neuronal death compared to the adult post‐ischemic brain (Hu et al. 2000; Manabat et al. 2003), impaired CD36‐dependent engulfment and clearance of neuronal debris may exacerbate injury after tMCAO. In CD36‐deficient neonates, the accumulation of uncleared cellular debris could amplify secondary inflammatory responses, thereby worsening ischemic brain injury (Faustino et al. 2011; Woo et al. 2012). Second, immune mechanisms central to adult CD36‐mediated toxicity (e.g., NF‐κB activation, superoxide production) are largely unaffected in neonates, highlighting developmental differences (Woo et al. 2012). Echoing developmental differences, while CD36 forms complexes with TLRs to function, including TLR2 in adult (Abe et al. 2010) and neonatal (Gadagkar et al. 2023) brains under physiological conditions, TLR2 expression is more than 25‐fold times higher in neonatal than in adult brain and gradually reduces over a one‐month period, as shown by live longitudinal bioluminescence imaging in mice bearing the dual luc/gfp reporter under the TLR2 promoter, with microglia as predominate cell type expressing TLR2 (Lalancette‐Hebert et al. 2017). In adult mice, TLR2 expression raises ~20‐fold after tMCAO and expression remains elevated for weeks, with microglia also being the predominant cell type expressing TLR2 (Lalancette‐Hebert et al. 2009). However, TLR2 expression is not further increased after tMCAO in P9 mice. Of note, comparative analysis showed context‐dependent responses following IL1β or LPS administration in neonates (Gadagkar et al. 2023; Lalancette‐Hebert et al. 2017). The relative contribution of cell type‐specific CD36 responses in neonates and adults may also contribute to the distinct effects on injury. For example, in adult mice, conditional deletion of CD36 either in microglia or endothelium reduces ischemic injury, indicating that CD36 exerts cell type‐specific detrimental effects in both populations (Garcia‐Bonilla et al. 2021), whereas neonatal mice subjected to tMCAO exhibit CD36‐dependent, cell type‐specific changes in Src/Lyn‐mediated signaling without appreciable changes in IL‐1β expression or NF‐κB activation (Woo et al. 2012).

The CP is an entry route for circulating immune cells to the parenchyma under both physiological and pathological states in adult (Kunis et al. 2013) and in neonatal mice following administration of TLR2 or TLR4 ligands (Mottahedin et al. 2019; Mottahedin et al. 2017). Examination of whether CD36 mediates such effects at this interface revealed that CD36 mediates neonatal stroke by governing peripheral immune cell traffic through the CP (Rayasam et al. 2020; Rayasam et al. 2022), while BBB integrity to molecules of various size/structure remains largely intact, limiting leukocyte trafficking through the latter route after acute neonatal stroke (Fernandez‐Lopez et al. 2012; Fernandez‐Lopez et al. 2016). Direct TLR2 stimulation as an independent injury model showed that this TLR2 agonist also rapidly recruits myeloid cells to the CP, the CSF and blood, but infiltration into the cortex remains low acutely (Rayasam et al. 2022).

In the present study, both stroke and direct TLR2 stimulation triggered rapid CP gene expression changes, but these responses were attenuated in KO mice. Shared CD36‐dependent pathways include cholesterol biosynthesis and DNA replication. Stroke, but not isolated TLR2 stimulation, induced ECM remodeling, suggesting injury‐specific mechanisms. Leukocyte recruitment through the CP was CD36‐dependent, though magnitude and dynamics differed between models. Cholesterol biosynthesis and DNA replication were among the common CD36‐dependent gene clusters in the CP ipsilateral to MCAO and after PAM treatment. Cholesterol homeostasis may affect various steps of leukocyte phenotypes and their transmigration, particularly CD36‐dependent effects under hyperlipidemic conditions after stroke (Kim et al. 2012; Peng et al. 2025). Heat map analyses indicate that the activation of several inflammatory genes occurs after both types of TLR2 stimuli in the neonatal brain. In contrast, alterations in the ECM were observed following tMCAO, but not after PAM administration. Consistently, comparisons of magnitude and particulars of leukocyte subtype trafficking via CP by flow cytometry revealed that while PAM induces CD36‐mediated recruitment of myeloid cells, the magnitude and temporal dynamic are different compared to that after tMCAO, likely due to the CD36 interplay with other ligands after stroke. In the parenchyma, the magnitude of neutrophil and macrophage presence after PAM is less profound than after tMCAO, also demonstrating the complex interplay between individual CD36‐mediated mechanisms. Yet, as we alluded to previously (Rayasam et al. 2020), our data do not allow discrimination between parenchyma immune cells from monocyte‐derived macrophages and whether the latter cells route through the CSF barriers or the BBB.

Our findings demonstrate that the acute CD36‐dependent beneficial effect converts to poorer long‐term histological outcome and is associated with features of late neuroinflammatory/vascular dysfunctions. In particular, extensive long‐term behavioral testing showed that CD36 deficiency by itself affects physiological responses including motor function, paw strength, and exploratory activity by young adulthood. However, tMCAO at P9 did not significantly alter balance, strength, fine movement, or anxiety at this stage on top of changed manifested by a lack of CD36 in naïve mice. At the same time, these findings may also be seen as a potential confounding factor in the interpretation of post‐stroke behavioral outcomes. Although CD36 deficiency did not produce clear improvements in long‐term functional recovery, deficiency was associated with structural preservation. CD36 also influenced chronic injury in a region‐specific vascular manner without altering white matter coverage but mediated sustained immune effects, including prolonged microglial activation in KO mice. KO mice showed better tissue preservation despite enhanced chronic microglial activation. Typically, expression of CD68+ on Iba‐1+ cells is attributed to neurotoxic rather than neuroprotective mechanisms, yet in this model at chronic time points, this can also be associated with neuroprotective/restorative mechanisms such as debris clearance. We show that in WT mice injury during the neonatal period translates to persistent microglia activation through young adulthood, whereas in KO mice, microglial cells in the hemisphere opposite to tMCAO retain ‘memory’ of activation (CD68+/Iba‐1+ cells), suggesting the change in the phenotypes over time. At the same time, a subpopulation of Iba1+/CD68+ cells can also be border macrophages, cells known to promote vascular oxidative stress and injury in adult neurodegeneration models (Park et al. 2011; Uekawa et al. 2023). Microglial cells continue to mature during the neonatal period (Li et al. 2019), and CD36 can affect metabolism of the maturating microglial population and the phenotypes of these cells. The data that microglial activation is common in naïve mice and in contralateral (not directly affected) regions of KO mice suggests that CD36 modulates the overall phenotypes of microglial cells long after injury during the neonatal period. At the same time, further studies of other immune signaling are needed to put in perspective the relative contribution of microglia‐mediated mechanisms to the overall long‐term CD36‐mediated immune responses after neonatal stroke. While a shift in CD36‐mediated effects on injury over time is apparent, our data do not allow identification of exact mechanisms of such a temporal shift in CD36‐associated effects. Several developmental and injury‐dependent processes may explain why CD36 is beneficial during the acute phase after neonatal stroke but detrimental during the chronic phase. The neonatal brain has a high baseline rate of programmed neuronal death, reflecting elevated expression of apoptotic machinery in the developing brain (Hu et al. 2000; Zhu et al. 2005), and compared to the adult brain after focal ischemia–reperfusion, caspase‐3‐dependent apoptosis is markedly greater after hypoxia–ischemia and focal ischemia–reperfusion in neonates (Han et al. 2002; Manabat et al. 2003; Zhu et al. 2005). Because neuronal death pathways are diverse and developmentally regulated, they strongly influence both acute injury and longer‐term injury progression (Chavez‐Valdez et al. 2018; Graham et al. 2004; Mike and Ferriero 2021; West et al. 2006). In this context, efficient CD36‐dependent clearance of apoptotic cells and debris may be beneficial early after injury; conversely, impaired debris clearance in CD36‐deficient mice may exacerbate acute pathology (Woo et al. 2012). The postnatal CNS immune and neurovascular environments also undergo substantial maturation. Developmental changes in complement activity, inflammatory signaling, and TLR pathways can alter the nature and consequences of immune responses to neonatal brain injury (Doverhag et al. 2010; Lalancette‐Hebert et al. 2017; Lalancette‐Hebert et al. 2009; Mallard et al. 2024). In parallel, the ECM, BBB, and blood–CSF interface continue to mature and shape age‐dependent responses to ischemic injury (Fernandez Lopez et al. 2012). Early‐life inflammation can also produce lasting increases in BBB permeability and impaired myelination (Stolp et al. 2005). Angiogenic and reparative responses are also developmentally constrained. While angiogenesis is induced relatively rapidly after adult stroke (Hayashi et al. 2003; Marti et al. 2000), vascular remodeling is delayed after tMCAO in P7 and P10 rats, despite the appearance of endothelial tip cells and rapid neuronal followed by astrocytic VEGF induction (Dzietko et al. 2013; Fernandez‐Lopez et al. 2013; Mu et al. 2003; Shimotake et al. 2010). Because CD36 regulates apoptotic‐cell clearance, inflammatory signaling, angiogenesis, and lipid handling across multiple CNS cell types, its absence may have distinct and lasting effects on vascular and neural repair and myelination. Studies in adult stroke, in turn, demonstrate that CD36 function is temporally regulated and can have context‐dependent effects (Kim et al. 2020). Thus, future work should define how developmental and temporal changes in CD36 expression, ligand availability, BBB maturation, microglial differentiation, and vascular repair govern its divergent roles during acute and chronic phases of neonatal stroke. In vivo MRI to assess acute injury and its progression, alongside myelination, histological, and behavioral outcomes, may further improve comprehensive understanding of the long‐term effects of CD36 deficiency after neonatal stroke.

Questions also remain regarding the developmental trajectory of cell‐specific CD36–TLR2 interactions, intracellular signaling dynamics, and metabolic regulation. Redistribution between intracellular and cell surface CD36 expression and its increased surface presence may become important during the sub‐chronic injury phase, as shown in adult mice (Woo et al. 2016). Intracellular signals from CD36 into the TLR4–TLR6–CD36 complex assembly have been proposed (Shmuel‐Galia et al. 2017). Dissociation of the CD36–TLR2 oligomeric complex pharmacologically or by CD36 deficiency induces a metabolic shift in adult models via PPARγ activation and reduces retinal damage by photo‐oxidative injury by muting activation of the pro‐inflammatory NF‐κB complex and NLRP3 inflammasome (Mellal et al. 2019). Inhibition of TLR4–TLR6 assembly is associated with reduced secretion of proinflammatory mediators from microglia, ultimately rescuing neurons from death in Alzheimer's disease (Shmuel‐Galia et al. 2017). CD36 was shown to influence metabolic pathways in adult models (Chen et al. 2022; Su and Abumrad 2009), but this aspect is essentially unknown in neonatal stroke. Gene expression changes indicate that CD36 influences metabolic and mitochondrial pathways in the neonatal CP, but the magnitude and specifics of CD36 in intracellular homeostasis continue to evolve over time in the adult (Hao et al. 2020; Su and Abumrad 2009), but the trajectory of its function in this manner is largely unknown in neonates.

Overall, our data highlight the complex age and context‐dependent impact of TLR and CD36 signaling and how this can play an essential role in determining functional and long‐term responses to brain injury.

Author Contributions

Elena Di Martino: investigation, writing – original draft, formal analysis, data curation, conceptualization. Aditya Rayasam: investigation, writing – original draft, conceptualization, data curation. Amin Mottahedin: investigation, writing – original draft, data curation, formal analysis, resources, software. Joel Faustino: investigation, methodology, writing – original draft. Shivani Mahuvakar: investigation, methodology, data curation. Joakim Ek: investigation, funding acquisition, writing – original draft, formal analysis, supervision. Carina Mallard: funding acquisition, conceptualization, data curation, supervision, writing – original draft, project administration. Zinaida S. Vexler: conceptualization, funding acquisition, writing – original draft, methodology, visualization, formal analysis, data curation, supervision, validation, project administration.

Funding

The work was supported by RO1 NS44025 (Z.V.), RO1 NS103483 (Z.V., C.M., C.J.E.), RO1 NS76726 (Z.V.), VR 2021‐01872 (C.M.), ALFGBG‐1005755 (C.M.), Åhlen Foundation (243041, C.M.), Brain Foundation FO2024‐0049‐TK‐142 (C.M.).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: RNA sequencing Deseq2 normalized count data.

JNC-170-e70553-s007.xlsx (19.6MB, xlsx)

Table S2: Opposite‐behavior gene‐set expression analysis. Source data for Figure 1E.

JNC-170-e70553-s002.xlsx (595KB, xlsx)

Table S3: Full statistical 2‐group analysis (A) and ANOVAs analysis (B) of histological, immunofluorescence and behavior outcomes.

JNC-170-e70553-s003.docx (25KB, docx)

Table S4: Full ANOVA analysis of flow cytometry data in the choroid plexus (CP) (A) and the cortex (B).

JNC-170-e70553-s005.docx (22.3KB, docx)

Figure S1: Detailed experimental time‐line.

JNC-170-e70553-s006.pdf (742.6KB, pdf)

Figure S2: (A) Injury volume in males and female mice 14 days after tMCAO. (B–G) Anatomical representation of the corpus callosum (B) for quantification of proliferating cells and oligodendrocytes at different maturation stages (C–G) in WT and CD36KO mice 2 weeks after tMCAO. Data shown as mean ± SD. Each dot represents data from an individual mouse. 2‐way ANOVA with Tukey's multiple comparisons test. Statistical values are indicated on individual panels. (H–J) Cell proliferation and neurogenesis in individual brain areas. Quantification of BrdU+, DCX+ and BrdU+ /DCX+ cells in the striatum, dentate gyrus (DG), granular cell layer (GCL) and subventricular zone (SVZ) of WT and KO mice 2 weeks post tMCAO. Data shown as mean ± SD. Comparison between four groups were evaluated by 2‐way ANOVA with Tukey's multiple comparisons test. Each dot represents data from an individual mouse. The number of mice in individual groups: WT: n = 5–9; KO: n = 6–10. Statistical values are indicated on individual panels.

JNC-170-e70553-s004.pdf (1.5MB, pdf)

Figure S3: (A, B) Open Field test: Effects on total fine movements (A) and relative time spent in the center vs. periphery (B). (C) Gait test: stride length and frequency of fore and hind paws 6 weeks after tMCAO in neonatal mice. (D) Beam walk test: Latency to fall (left) and the number of slips (right). Comparison between four groups were evaluated by 2‐way ANOVA with Tukey's multiple comparisons test, or 2‐way ANOVA with Šídák's multiple comparisons test for repeated measurements. Each dot represents data from an individual mouse. Statistical values are indicated on individual panels. The number of mice in individual groups: Gait test: MCAO WT: n = 24; MCAO KO: n = 13; Sham WT n = 14; Sham KO n = 12. Beam and Open Field: MCAO WT: n = 14; MCAO KO: n = 16; Sham WT n = 24; Sham KO n = 14.

JNC-170-e70553-s001.pdf (582.8KB, pdf)

Data S1: jnc70553‐sup‐0008‐Supinfo1.docx.

JNC-170-e70553-s008.docx (23.5KB, docx)

Acknowledgments

The authors thank Jeff Simms and Iris Lo from Behavior Core, Gladstone Institute, University California San Francisco, San Francisco, CA, USA, for conducting behavior analysis.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

References

  1. Abe, T. , Shimamura M., Jackman K., et al. 2010. “Key Role of CD36 in Toll‐Like Receptor 2 Signaling in Cerebral Ischemia.” Stroke 41, no. 5: 898–904. 10.1161/STROKEAHA.109.572552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aghaeepour, N. , Finak G., Flow C. A. P. C., et al. 2013. “Critical Assessment of Automated Flow Cytometry Data Analysis Techniques.” Nature Methods 10, no. 3: 228–238. 10.1038/nmeth.2365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Balkaya, M. , Kim I. D., Shakil F., and Cho S.. 2021. “CD36 Deficiency Reduces Chronic BBB Dysfunction and Scar Formation and Improves Activity, Hedonic and Memory Deficits in Ischemic Stroke.” Journal of Cerebral Blood Flow and Metabolism 41, no. 3: 486–501. 10.1177/0271678x20924099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chavez‐Valdez, R. , Emerson P., Goffigan‐Holmes J., Kirkwood A., Martin L. J., and Northington F. J.. 2018. “Delayed Injury of Hippocampal Interneurons After Neonatal Hypoxia‐Ischemia and Therapeutic Hypothermia in a Murine Model.” Hippocampus 28, no. 8: 617–630. 10.1002/hipo.22965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chen, Y. , Zhang J., Cui W., and Silverstein R. L.. 2022. “CD36, a Signaling Receptor and Fatty Acid Transporter That Regulates Immune Cell Metabolism and Fate.” Journal of Experimental Medicine 219, no. 6: e20211314. 10.1084/jem.20211314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cho, S. , Park E. M., Febbraio M., et al. 2005. “The Class B Scavenger Receptor CD36 Mediates Free Radical Production and Tissue Injury in Cerebral Ischemia.” Journal of Neuroscience 25, no. 10: 2504–2512. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=15758158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Doverhag, C. , Hedtjarn M., Poirier F., et al. 2010. “Galectin‐3 Contributes to Neonatal Hypoxic‐Ischemic Brain Injury.” Neurobiology of Disease 38, no. 1: 36–46. 10.1016/j.nbd.2009.12.024S0969-9961(09)00379-9. [DOI] [PubMed] [Google Scholar]
  8. Dzietko, M. , Derugin N., Wendland M. F., Vexler Z. S., and Ferriero D. M.. 2013. “Delayed VEGF Treatment Enhances Angiogenesis and Recovery After Neonatal Focal Rodent Stroke.” Translational Stroke Research 4: 189–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Faustino, J. , Wang X., Jonhson C., et al. 2011. “Microglial Cells Contribute to Endogenous Brain Defenses After Acute Neonatal Focal Stroke.” Journal of Neuroscience 31, no. 36: 12992–13001. 10.1523/JNEUROSCI.2102-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Febbraio, M. , Hajjar D. P., and Silverstein R. L.. 2001. “CD36: A Class B Scavenger Receptor Involved in Angiogenesis, Atherosclerosis, Inflammation, and Lipid Metabolism.” Journal of Clinical Investigation 108, no. 6: 785–791. 10.1172/jci14006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fernandez Lopez, D. , Faustino J., Daneman R., et al. 2012. “Blood‐Brain Barrier Permeability Is Increased After Acute Adult Stroke but Not Neonatal Stroke.” Journal of Neuroscience 32, no. 28: 9588–9600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fernandez‐Lopez, D. , Faustino J., Daneman R., et al. 2012. “Blood‐Brain Barrier Permeability Is Increased After Acute Adult Stroke but Not Neonatal Stroke in the Rat.” Journal of Neuroscience 32, no. 28: 9588–9600. 10.1523/JNEUROSCI.5977-11.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fernandez‐Lopez, D. , Faustino J., Derugin N., and Vexler Z. S.. 2013. “Acute and Chronic Vascular Responses to Experimental Focal Arterial Stroke in the Neonate Rat.” Translational Stroke Research 4: 179–188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fernandez‐Lopez, D. , Faustino J., Klibanov A. L., et al. 2016. “Microglial Cells Prevent Hemorrhage in Neonatal Focal Arterial Stroke.” Journal of Neuroscience 36, no. 10: 2881–2893. 10.1523/JNEUROSCI.0140-15.2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fukuzaki, Y. , Faustino J., Lecuyer M. J., Rayasam A., and Vexler Z. S.. 2023. “Global Sphingosine‐1‐Phosphate Receptor 2 Deficiency Attenuates Neuroinflammation and Ischemic‐Reperfusion Injury After Neonatal Stroke.” iScience 26: 106340. https://www.cell.com/iscience/fulltext/S2589‐0042(23)00417‐0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gadagkar, S. G. , Lalancette‐Hébert M., Thammisetty S. S., Vexler Z. S., and Kriz J.. 2023. “CD36 Neutralisation Blunts TLR2‐IRF7 but Not IRF3 Pathway in Neonatal Mouse Brain and Immature Human Microglia Following Innate Immune Challenge.” Scientific Reports 13, no. 1: 2304. 10.1038/s41598-023-29423-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Garcia‐Bonilla, L. , Faraco G., Moore J., et al. 2016. “Spatio‐Temporal Profile, Phenotypic Diversity, and Fate of Recruited Monocytes Into the Post‐Ischemic Brain.” Journal of Neuroinflammation 13, no. 1: 285. 10.1186/s12974-016-0750-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Garcia‐Bonilla, L. , Sciortino R., Shahanoor Z., et al. 2021. “Role of Microglial and Endothelial CD36 in Post‐Ischemic Inflammasome Activation and Interleukin‐1β‐Induced Endothelial Activation.” Brain, Behavior, and Immunity 95: 489–501. 10.1016/j.bbi.2021.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Graham, E. M. , Sheldon R. A., Flock D. L., et al. 2004. “Neonatal Mice Lacking Functional Fas Death Receptors Are Resistant to Hypoxic‐Ischemic Brain Injury.” Neurobiology of Disease 17, no. 1: 89–98. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=15350969. [DOI] [PubMed] [Google Scholar]
  20. Griffiths, M. R. , Gasque P., and Neal J. W.. 2010. “The Regulation of the CNS Innate Immune Response Is Vital for the Restoration of Tissue Homeostasis (Repair) After Acute Brain Injury: A Brief Review.” International Journal of Inflammation 2010: 151097. 10.4061/2010/151097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Han, B. H. , Xu D., Choi J., et al. 2002. “Selective, Reversible Caspase‐3 Inhibitor Is Neuroprotective and Reveals Distinct Pathways of Cell Death Following Neonatal Hypoxic‐ Ischemic Brain Injury.” Journal of Biological Chemistry 277, no. 33: 30128–30136. [DOI] [PubMed] [Google Scholar]
  22. Hao, J. W. , Wang J., Guo H., et al. 2020. “CD36 Facilitates Fatty Acid Uptake by Dynamic Palmitoylation‐Regulated Endocytosis.” Nature Communications 11, no. 1: 4765. 10.1038/s41467-020-18565-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hayashi, T. , Noshita N., Sugawara T., and Chan P. H.. 2003. “Temporal Profile of Angiogenesis and Expression of Related Genes in the Brain After Ischemia.” Journal of Cerebral Blood Flow and Metabolism 23, no. 2: 166–180. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=12571448. [DOI] [PubMed] [Google Scholar]
  24. Hu, B. R. , Liu C. L., Ouyang Y., Blomgren K., and Siesjo B. K.. 2000. “Involvement of Caspase‐3 in Cell Death After Hypoxia‐Ischemia Declines During Brain Maturation.” Journal of Cerebral Blood Flow and Metabolism 20, no. 9: 1294–1300. [DOI] [PubMed] [Google Scholar]
  25. Jimenez, B. , Volpert O. V., Crawford S. E., Febbraio M., Silverstein R. L., and Bouck N.. 2000. “Signals Leading to Apoptosis‐Dependent Inhibition of Neovascularization by Thrombospondin‐1.” Nature Medicine 6, no. 1: 41–48. 10.1038/71517. [DOI] [PubMed] [Google Scholar]
  26. Kim, E. , Febbraio M., Bao Y., Tolhurst A. T., Epstein J. M., and Cho S.. 2012. “CD36 in the Periphery and Brain Synergizes in Stroke Injury in Hyperlipidemia.” Annals of Neurology 71, no. 6: 753–764. 10.1002/ana.23569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kim, E. , Yang J., Woo Park K., and Cho S.. 2020. “Preventative, but Not Post‐Stroke, Inhibition of CD36 Attenuates Brain Swelling in Hyperlipidemic Stroke.” Journal of Cerebral Blood Flow and Metabolism 40, no. 4: 885–894. 10.1177/0271678x19850004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kunis, G. , Baruch K., Rosenzweig N., et al. 2013. “IFN‐Gamma‐Dependent Activation of the Brain's Choroid Plexus for CNS Immune Surveillance and Repair.” Brain 136, no. Pt 11: 3427–3440. 10.1093/brain/awt259. [DOI] [PubMed] [Google Scholar]
  29. Lalancette‐Hebert, M. , Faustino J., Thammisetty S. S., Chip S., Vexler Z. S., and Kriz J.. 2017. “Live Imaging of Innate Immune Response in Neonates Reveals Differential TLR‐2 Dependent Activation Patterns in Sterile Inflammation and Infection.” Brain, Behavior, and Immunity 65: 312–327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lalancette‐Hebert, M. , Phaneuf D., Soucy G., Weng Y. C., and Kriz J.. 2009. “Live Imaging of Toll‐Like Receptor 2 Response in Cerebral Ischaemia Reveals a Role of Olfactory Bulb Microglia as Modulators of Inflammation.” Brain 132: 940–954. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=19153151. [DOI] [PubMed] [Google Scholar]
  31. Lamb, J. , Crawford E. D., Peck D., et al. 2006. “The Connectivity Map: Using Gene‐Expression Signatures to Connect Small Molecules, Genes, and Disease.” Science 313, no. 5795: 1929–1935. 10.1126/science.1132939. [DOI] [PubMed] [Google Scholar]
  32. Li, F. , Faustino J., Woo M., Derugin N., and Vexler Z. S.. 2015. “Lack of the Scavenger Receptor CD36 Alters Microglial Phenotypes After Neonatal Stroke.” Journal of Neurochemistry 13, no. 3: 445–452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Li, Q. , Cheng Z., Zhou L., et al. 2019. “Developmental Heterogeneity of Microglia and Brain Myeloid Cells Revealed by Deep Single‐Cell RNA Sequencing.” Neuron 101, no. 2: 207–223.e210. 10.1016/j.neuron.2018.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mallard, C. , Ferriero D. M., and Vexler Z. S.. 2024. “Immune‐Neurovascular Interactions in Experimental Perinatal and Childhood Arterial Ischemic Stroke.” Stroke 55, no. 2: 506–518. 10.1161/strokeaha.123.043399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Manabat, C. , Han B. H., Wendland M., et al. 2003. “Reperfusion Differentially Induces Caspase‐3 Activation in Ischemic Core and Penumbra After Stroke in Immature Brain.” Stroke 34, no. 1: 207–213. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=12511776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Marti, H. J. , Bernaudin M., Bellail A., et al. 2000. “Hypoxia‐Induced Vascular Endothelial Growth Factor Expression Precedes Neovascularization After Cerebral Ischemia.” American Journal of Pathology 156, no. 3: 965–976. 10.1016/S0002-9440(10)64964-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mellal, K. , Omri S., Mulumba M., et al. 2019. “Immunometabolic Modulation of Retinal Inflammation by CD36 Ligand.” Scientific Reports 9, no. 1: 12903. 10.1038/s41598-019-49472-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mike, J. K. , and Ferriero D. M.. 2021. “Efferocytosis Mediated Modulation of Injury After Neonatal Brain Hypoxia‐Ischemia.” Cells 10, no. 5: 1025. 10.3390/cells10051025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mottahedin, A. , Joakim Ek C., Truve K., Hagberg H., and Mallard C.. 2019. “Choroid Plexus Transcriptome and Ultrastructure Analysis Reveals a TLR2‐Specific Chemotaxis Signature and Cytoskeleton Remodeling in Leukocyte Trafficking.” Brain, Behavior, and Immunity 79: 216–227. 10.1016/j.bbi.2019.02.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mottahedin, A. , Smith P. L., Hagberg H., Ek C. J., and Mallard C.. 2017. “TLR2‐Mediated Leukocyte Trafficking to the Developing Brain.” Journal of Leukocyte Biology 101: 297–305. 10.1189/jlb.3A1215-568R. [DOI] [PubMed] [Google Scholar]
  41. Mu, D. , Jiang X., Sheldon R. A., et al. 2003. “Regulation of Hypoxia‐Inducible Factor 1alpha and Induction of Vascular Endothelial Growth Factor in a Rat Neonatal Stroke Model.” Neurobiology of Disease 14, no. 3: 524–534. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=14678768. [DOI] [PubMed] [Google Scholar]
  42. Myers, S. A. , Andres K. R., Hagg T., and Whittemore S. R.. 2014. “CD36 Deletion Improves Recovery From Spinal Cord Injury.” Experimental Neurology 256: 25–38. 10.1016/j.expneurol.2014.03.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Park, L. , Wang G., Zhou P., et al. 2011. “Scavenger Receptor CD36 Is Essential for the Cerebrovascular Oxidative Stress and Neurovascular Dysfunction Induced by Amyloid‐Beta.” Proceedings of the National Academy of Sciences of the United States of America 108, no. 12: 5063–5068. 10.1073/pnas.1015413108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Peng, X. , Lian Z., O'Brien V., et al. 2025. “Foamy Monocytes and Atherogenesis in Mice With Combined Hyperlipidemia and Effects of Antisense Knockdown of apoCIII.” Journal of Lipid Research 66, no. 4: 100763. 10.1016/j.jlr.2025.100763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Primo, L. , Ferrandi C., Roca C., et al. 2005. “Identification of CD36 Molecular Features Required for Its In Vitro Angiostatic Activity.” FASEB Journal 19, no. 12: 1713–1715. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=16037098. [DOI] [PubMed] [Google Scholar]
  46. Raudvere, U. , Kolberg L., Kuzmin I., et al. 2019. “G:Profiler: A Web Server for Functional Enrichment Analysis and Conversions of Gene Lists (2019 Update).” Nucleic Acids Research 47, no. W1: W191–W198. 10.1093/nar/gkz369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rayasam, A. , Faustino J., Lecuyer M., and Vexler Z. S.. 2020. “Neonatal Stroke and TLR1/2 Ligand Recruit Myeloid Cells Through the Choroid Plexus in a CX3CR1‐CCR2‐ and Context‐Specific Manner.” Journal of Neuroscience 40, no. 19: 3849–3861. 10.1523/JNEUROSCI.2149-19.2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Rayasam, A. , Mottahedin A., Faustino J., Mallard C., and Vexler Z. S.. 2022. “Scavenger Receptor CD36 Governs Recruitment of Myeloid Cells to the Blood‐CSF Barrier After Stroke in Neonatal Mice.” Journal of Neuroinflammation 19, no. 1: 47. 10.1186/s12974-022-02388-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Shimotake, J. , Derugin N., Wendland M., Vexler Z. S., and Ferriero D. M.. 2010. “Vascular Endothelial Growth Factor Receptor‐2 Inhibition Promotes Cell Death and Limits Endothelial Cell Proliferation in a Neonatal Rodent Model of Stroke.” Stroke 41, no. 2: 343–349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Shmuel‐Galia, L. , Klug Y., Porat Z., Charni M., Zarmi B., and Shai Y.. 2017. “Intramembrane Attenuation of the TLR4‐TLR6 Dimer Impairs Receptor Assembly and Reduces Microglia‐Mediated Neurodegeneration.” Journal of Biological Chemistry 292, no. 32: 13415–13427. 10.1074/jbc.M117.784983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Stolp, H. B. , Dziegielewska K. M., Ek C. J., Potter A. M., and Saunders N. R.. 2005. “Long‐Term Changes in Blood‐Brain Barrier Permeability and White Matter Following Prolonged Systemic Inflammation in Early Development in the Rat.” European Journal of Neuroscience 22, no. 11: 2805–2816. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=16324115. [DOI] [PubMed] [Google Scholar]
  52. Su, X. , and Abumrad N. A.. 2009. “Cellular Fatty Acid Uptake: A Pathway Under Construction.” Trends in Endocrinology and Metabolism 20, no. 2: 72–77. 10.1016/j.tem.2008.11.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Subramanian, A. , Narayan R., Corsello S. M., et al. 2017. “A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles.” Cell 171, no. 6: 1437–1452.e1417. 10.1016/j.cell.2017.10.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Uekawa, K. , Hattori Y., Ahn S. J., et al. 2023. “Border‐Associated Macrophages Promote Cerebral Amyloid Angiopathy and Cognitive Impairment Through Vascular Oxidative Stress.” Molecular Neurodegeneration 18, no. 1: 73. 10.1186/s13024-023-00660-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. van Berkel, T. J. , Out R., Hoekstra M., Kuiper J., Biessen E., and van Eck M.. 2005. “Scavenger Receptors: Friend or Foe in Atherosclerosis?” Current Opinion in Lipidology 16, no. 5: 525–535. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=16148537. [DOI] [PubMed] [Google Scholar]
  56. Wang, R. , Ashwal S., Tone B., et al. 2007. “Albumin Reduces Blood‐Brain Barrier Permeability but Does Not Alter Infarct Size in a Rat Model of Neonatal Stroke.” Pediatric Research 62, no. 3: 261–266. 10.1203/PDR.0b013e318123f757. [DOI] [PubMed] [Google Scholar]
  57. West, T. , Atzeva M., and Holtzman D. M.. 2006. “Caspase‐3 Deficiency During Development Increases Vulnerability to Hypoxic‐Ischemic Injury Through Caspase‐3‐Independent Pathways.” Neurobiology of Disease 22, no. 3: 523–537. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=16480886. [DOI] [PubMed] [Google Scholar]
  58. Woo, M. S. , Wang X., Faustino J., et al. 2012. “Genetic Deletion of CD36 Enhances Injury After Acute Neonatal Stroke.” Annals of Neurology 72: 961–970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Woo, M. S. , Yang J., Beltran C., and Cho S.. 2016. “Cell Surface CD36 Protein in Monocyte/Macrophage Contributes to Phagocytosis During the Resolution Phase of Ischemic Stroke in Mice.” Journal of Biological Chemistry 291, no. 45: 23654–23661. 10.1074/jbc.M116.750018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zhao, X. , Grotta J., Gonzales N., and Aronowski J.. 2009. “Hematoma Resolution as a Therapeutic Target: The Role of Microglia/Macrophages.” Stroke 40, no. 3 Suppl: S92–S94. 10.1161/STROKEAHA.108.533158. [DOI] [PubMed] [Google Scholar]
  61. Zhu, C. , Wang X., Xu F., et al. 2005. “The Influence of Age on Apoptotic and Other Mechanisms of Cell Death After Cerebral Hypoxia‐Ischemia.” Cell Death and Differentiation 12, no. 2: 162–176. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=15592434. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1: RNA sequencing Deseq2 normalized count data.

JNC-170-e70553-s007.xlsx (19.6MB, xlsx)

Table S2: Opposite‐behavior gene‐set expression analysis. Source data for Figure 1E.

JNC-170-e70553-s002.xlsx (595KB, xlsx)

Table S3: Full statistical 2‐group analysis (A) and ANOVAs analysis (B) of histological, immunofluorescence and behavior outcomes.

JNC-170-e70553-s003.docx (25KB, docx)

Table S4: Full ANOVA analysis of flow cytometry data in the choroid plexus (CP) (A) and the cortex (B).

JNC-170-e70553-s005.docx (22.3KB, docx)

Figure S1: Detailed experimental time‐line.

JNC-170-e70553-s006.pdf (742.6KB, pdf)

Figure S2: (A) Injury volume in males and female mice 14 days after tMCAO. (B–G) Anatomical representation of the corpus callosum (B) for quantification of proliferating cells and oligodendrocytes at different maturation stages (C–G) in WT and CD36KO mice 2 weeks after tMCAO. Data shown as mean ± SD. Each dot represents data from an individual mouse. 2‐way ANOVA with Tukey's multiple comparisons test. Statistical values are indicated on individual panels. (H–J) Cell proliferation and neurogenesis in individual brain areas. Quantification of BrdU+, DCX+ and BrdU+ /DCX+ cells in the striatum, dentate gyrus (DG), granular cell layer (GCL) and subventricular zone (SVZ) of WT and KO mice 2 weeks post tMCAO. Data shown as mean ± SD. Comparison between four groups were evaluated by 2‐way ANOVA with Tukey's multiple comparisons test. Each dot represents data from an individual mouse. The number of mice in individual groups: WT: n = 5–9; KO: n = 6–10. Statistical values are indicated on individual panels.

JNC-170-e70553-s004.pdf (1.5MB, pdf)

Figure S3: (A, B) Open Field test: Effects on total fine movements (A) and relative time spent in the center vs. periphery (B). (C) Gait test: stride length and frequency of fore and hind paws 6 weeks after tMCAO in neonatal mice. (D) Beam walk test: Latency to fall (left) and the number of slips (right). Comparison between four groups were evaluated by 2‐way ANOVA with Tukey's multiple comparisons test, or 2‐way ANOVA with Šídák's multiple comparisons test for repeated measurements. Each dot represents data from an individual mouse. Statistical values are indicated on individual panels. The number of mice in individual groups: Gait test: MCAO WT: n = 24; MCAO KO: n = 13; Sham WT n = 14; Sham KO n = 12. Beam and Open Field: MCAO WT: n = 14; MCAO KO: n = 16; Sham WT n = 24; Sham KO n = 14.

JNC-170-e70553-s001.pdf (582.8KB, pdf)

Data S1: jnc70553‐sup‐0008‐Supinfo1.docx.

JNC-170-e70553-s008.docx (23.5KB, docx)

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Articles from Journal of Neurochemistry are provided here courtesy of Wiley

RESOURCES