Table of Contents
Year : 2021  |  Volume : 6  |  Issue : 4  |  Page : 264-273

Re-visit the concept of M1 versus M2 phenotypes of BV2 microglia and test their effects on stroke outcome in mice

Department of Neurosurgery, Stanford University School of Medicine, CA 94305, USA

Date of Submission27-Aug-2021
Date of Acceptance06-Dec-2021
Date of Web Publication30-Dec-2021

Correspondence Address:
Heng Zhao
Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA 94305
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/2470-7511.334399

Get Permissions


Backgrounds: Whether there are distinctive macrophage functional phenotypes of M1 versus M2 has been debated. We re-examined them by studying M1/M2 gene and protein expressions in cultured BV2 microglial cells and their effects on stroke outcomes in vivo. Methods: BV2 microglia cells were cultured and polarized with lipopolysaccharide (LPS) and interleukin-4 (IL-4) to produce M (LPS) and M (IL-4) phenotypes, which were originally defined as M1 and M2 phenotypes, respectively. Typical M1 and M2 gene or protein expression patterns were analyzed in M (LPS) and M (IL-4) phenotypes and their distinctive effects on stroke outcomes were compared. Results: M (LPS) and M (IL-4) had distinctive morphologies. M (IL-4) had significantly higher gene expressions of the typical M2 markers and other anti-inflammatory genes, while M (LPS) had higher gene expression of typical M1 markers and other pro-inflammatory genes. Nevertheless, M2 gene expressions were also enhanced in M (LPS), and M1 gene expressions were increased in M (IL-4), although with relatively lower levels. Adoptive transfer of M (IL-4) reduced infarction and improved neurological scores, while M (LPS) macrophages generated the opposite effect. Fluorescence activated cell sorting (FACS) and confocal studies suggest that M (IL-4) inhibited, while M (LPS) promoted the infiltration of monocyte-derived macrophages and iNOS-positive cells. Conclusions: M (LPS) and M (IL-4) from cultured BV2 cells indeed are distinctive functional phenotypes, but it is inaccurate to simply classify them into M1 and M2 phenotypes based on a few typical gene and protein markers.

Keywords: BV2 microglia; Focal cerebral ischemia; Phenotypes; Neuroinflammation; Stroke

How to cite this article:
Yan DM, Zhang YM, Ji YH, Wang T, Xiong XX, Zhao H. Re-visit the concept of M1 versus M2 phenotypes of BV2 microglia and test their effects on stroke outcome in mice. Cardiol Plus 2021;6:264-73

How to cite this URL:
Yan DM, Zhang YM, Ji YH, Wang T, Xiong XX, Zhao H. Re-visit the concept of M1 versus M2 phenotypes of BV2 microglia and test their effects on stroke outcome in mice. Cardiol Plus [serial online] 2021 [cited 2022 Jan 19];6:264-73. Available from:

Dong-Mei Yan and Yong-Ming Zhang contributed equally to this work.

  Introduction Top

Neuroinflammation modulates brain injury induced by stroke, in which macrophages play critical roles.[1] It has been considered that macrophages are polarized into M1 and M2 subsets, which promote and inhibit inflammation, respectively.[2],[3],[4] Both M1 and M2 subtypes were identified in the ischemic brain after stroke, which are linked with brain injury.[1], [5],[6],[7],[8] Nevertheless, relatively, there are less evidence about the direct effects of M1 and M2 subtypes on brain injury.

Although the M1 and M2 concepts were introduced into the stroke field for only a few years,[1] such a classification has been questioned in the mainstream of immunology research field, and many researchers have appealed to abandon this concept.[9],[10],[11] In vitro studies suggest that M1 type is induced by interferon-gamma (IFN-γ) or lipopolysaccharide (LPS), releasing tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), IL-12, and many other pro-inflammatory factors, such as NO and ROS; M2 type is induced by IL-4/IL-13, marked with the overexpression of arginase 1, Ym-1, and Fizz-1.[11] However, there are some limitations when this M1/M2 concept is applied to in vivo studies, and several reasons are given for discarding this classification.[9],[10],[11] First, the concepts of M1 and M2 are defined in vitro with clear stimulations of LPS/IFN-γ and IL-4/IL-13, respectively, but there are mixed stimulations in an in vivo environment. Second, macrophages may not be polarized into the two extremes of M1 versus M2, but extend to a broad spectrum between the M1 and M2. Third, only a few M1 and M2 markers are usually used to define M1 and M2 phenotypes, but there are some overlaps for these protein expressions in M1 and M2 macrophages.[9]

In this study, we re-visited the concepts of M1 versus M2 phenotypes by using cultured-microglia BV2 cells and evaluated their effects on stroke outcomes. We used BV2 microglia cells because microglia are important for brain injury and recovery after stroke and BV2 is a cell line frequently used to study the function of microglia.[12],[13],[14],[15],[16] In addition, although a number of studies have studied M1 and M2 phenotypes by culturing BV2 cells,[17],[18],[19] gene expressions of M1 and M2 from BV2 cells are not well compared, and most times, only a few genes are studied.[17] Furthermore, how the adoptive transfer of BV2 cells that are activated by LPS or IL-4 affects stroke outcomes remains unknown. In the current study, BV2 cells are polarized by LPS and IL-4 into M1 and M2, which, however, are named M (LPS) and M (IL-4) to avoid any confusion in this study, as suggested.[11] We re-examined whether M (LPS) and M (IL-4) derived from BV2 cells can be defined as M1 and M2, by comparing their morphologies and gene and protein expression patterns. We hypothesize that if M1 and M2 differ, they should have distinctive effects on stroke outcomes. Thus, we assessed their effects on stroke outcomes by adoptively transferring these cells into mice in a focal cerebral ischemia model. The major purpose is to clarify if it is valid to classify M (LPS) and M (IL-4) as M1 and M2 and whether they have distinctive effects on stroke outcomes.

  Materials and Methods Top

Animal procedures were approved by the Stanford Institutional Animal Care and Use Committee and the NIH Guidelines of Care and Use of Laboratory Animals. All efforts were made to minimize the number of animals used and their suffering.

BV-2 cell culture and polarization

Murine BV2 microglial cells were cultured in RPMI1640 supplemented with 10% FBS, and 100U/ml penicillin, and 100 μg/ml streptomycin. Cells were maintained at 37°C in a 5% CO2 incubator. BV2 cells were seeded at 1 × 106 per well in a 25 cm2 plate for 24 h and then stimulated with LPS (1 μg/ml) or murine IL-4 (25 ng/ml) for 48 h, to generate M (LPS) or M (IL-4), respectively.

Determination of M (LPS) versus M (IL-4) phenotypes from cultured BV2 cells

To identify the classic M1 and M2 macrophage functional phenotypes from M (LPS) and M (IL-4) BV2 cells, we used the following three methods:

  1. Morphology detection. The cell morphologies were carefully observed under phase-contrast microscopy, pictures were taken and examined whether they showed distinctive morphology in LPS and IL-4-polarized BV2 cells.
  2. Reverse transcription-polymerase chain reaction (RT-PCR) for determining gene expression of M1 and M2 markers from M (LPS) and M (IL-4): After observing cell morphologies, we then measured gene expression levels of the typical M1 markers, iNOS and TNF-α, and the typical M2 markers, Arginase-1 (Arg-1) and transforming growth factor (TGF)-β, as a traditional way to identify M1 versus M2 phenotypes. For RT-PCR assay, BV2 cells were seeded in 25 cm2 flask and polarized with LPS (1 μg/ml) and IL-4 (25 ng/ml) to produce M (LPS) and M (IL-4), respectively. After polarization, the cells were collected and total RNA was extracted with TRIzol (Invitrogen) following the manufacturer's protocol and then reverse transcribed into cDNA using the SuperScriptR III First Strand (Invitrogen; 1 μg was used to synthesize the first strand of cDNA using the Superscript First-Strand Synthesis System for RT-PCR. The parameters for PCR were as follows: Denaturing at 94°C for 3 min, followed by 25 cycles of 94°C for 30 s and 51°C for 1 min and 72°C for 1 min, and then extension at 72°C for 2 min. The resulting cDNA was used for real-time PCR for iNOS and Tnfa, Arg1 and Tgfb) using an ABI PRISM 7300 sequence detection system (Applied Biosystems, USA). Quantitative PCR was performed using the SYBR green PCR master mix (Applied Biosystem, USA). The expression of each gene was normalized to the expression of the housekeeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primers of genes are in [Table 1].
  3. Immunofluorescent staining of the M1 and M2 markers: BV2 cells were cultured in 24-wells plate with slide. After polarized with LPS and IL-4, respectively, the slides were fixed with methyl alcohol and acetone (1:1) for 10 min and then washed three times with PBS. After washing, the slides were blocked with 5% BSA for overnight and incubated with mouse anti-iNOS Ab (1:200) and rabbit anti-Arg1 Ab (1:200) (Cambridge, MA USA) for overnight. After washing, the goat anti-mouse secondary antibody (Alexa Fluor® 594 conjugate) (1:250) and goat anti-rabbit secondary antibody (Alexa Fluor® 488 conjugate) (1:500) (Cell Signaling, Danvers, MA) were added and incubated for 2 h at room temperature. Then, 4',6-diamidino-2-phenylindole were added and stained for 5 min. The slides were sealed with Fluoromount(Sigma) and observed with Confocal laser scanning microscopy (Zeiss axiovert inverted scanning microscope; Zeiss).
Table 1: The primers of genes for reverse transcription-polymerase chain reaction

Click here to view

Microfluidic single-cell real-time polymerase chain reaction for comparative analysis of gene expression patterns of polarized BV2 cells

We then used a high throughput quantitative PCR platform to compare gene expression patterns between M (LPS) and M (IL-4) phenotypes.[20] After the polarization of BV2 cells, 100 single cells were sorted into wells containing 10 μl of reaction buffer (CellsDirect kit, Invitrogen). Reverse transcription and specific transcript amplification were performed using 1 μl of SuperScript III Reverse Transcriptase/Platinum Taq Mix (Invitrogen). A microfluidic platform was used to conduct single-cell qPCR in nanoliter reaction volumes for high throughput gene expression profiling. The amplified cDNA was loaded into Biomark 48.48 Dynamic Array chips using the Nanoflex IFC controller (Fluidigm). Using the BioMark Real-Time PCR Analysis software, we extracted the Threshold cycle (CT) as a measurement of relative fluorescence intensity. For data analyses, samples with no detectable controls genes (GAPDH and HPRT1) or with detectable genes from contaminant cells (CD3 CD19, IFNG, GZMB) are removed. For each gene, including the controls (Gapdh and Hprt1), the data with Ct Call (Failed) and Ct Quality <0.65 are considered as missing data and the values are substituted by 29. Next, low expression genes (Ct Values ≥ Ct Value Threshold) are filtered out, and the default Ct value threshold is 28.0. Finally, to make expression values more intuitive for visualization, Ct values are converted to relative copy numbers according to the following equation: Relative quantities (RQ) of cDNA molecules = 228–Ct.

Mouse stroke model, neurological functions measurement, and infarction measurement

We used a mouse stroke model to test the effects of M (LPS) versus M (IL-4) BV2 cell transplantation on stroke outcomes. Male C57BL/J with the age of 12 weeks, 20 ± 2 g, were purchased from The Jackson Laboratory (Bar Harbor, ME). Stroke was induced by transient middle cerebral artery occlusion (MCAO), as described in our previous studies.[21],[22],[23],[24] Briefly, anesthesia was induced using a face mask with 5% and maintained with 1%–2% of isoflurane in 70% N2O and 30% O2 mixture. Body core temperature was monitored and maintained at 37 ± 0.5 by a surface-heating pad by inserting a rectal probe during the entire procedure. The left common carotid artery was exposed and a small incision was made, and a 6-0 nylon monofilament suture coated with silicon on its tip was inserted into the CCA, and advanced into the ICA, to induce MCAO. The wound was sewed, and the animals were then released from the anesthesia. At 45 min after MCAO, the animals were re-anesthetized, and the nylon monofilament was removed. After the closure of the incision, the mice were allowed to wake up from anesthesia and returned to the cages.

For neurological functions measurement [Table 2], the mice were estimated at 24 and 48 h after stroke. The neurological grading score from 0 to 4 in the body symmetry, gait, climbing, circling behavior, front limb symmetry, compulsory circling.
Table 2: Neurological scores

Click here to view

Infarctions were measured 48 h after stroke. Brains were sectioned into five 2-mm coronal planes and stained with 2% 2, 3, 5-triphenyl tetrazolium chloride (TTC, Sigma, St. Louis, MO) at 37°C for 15 min before they were fixed in 10% formalin overnight. Images of the slices were digitalized, and infarct areas outlined in white were measured using Image-J software. Infarct volumes were determined by multiplying the average slice thickness (2 mm) by the sum of the infarct areas in all five-brain slices. The percentage of the total infarct area was also calculated for each of the five brain slices.

Adoptive transfer of BV2 macrophages and carboxyfluorescein succinimidyl ester labeled polarized BV2 cells

The mice that underwent MCAO were randomly divided into four groups as follows: Vehicle treatment, treatment with control BV2 cells, M (LPS), and M (IL-4) (n = 5–7 mice). Polarized BV2 in the flask were digested with trypsin and centrifuged for 5 min at 400 g. The cell mass was resuspended with PBS and regulate the concentration to 1 × 107/ml. 1 million polarized microglia in 0.1 ml PBS were administered by vein infusion after transient MCA occlusion immediately. Vehicle-treated mice were injected with an equal volume of PBS.

To track adoptively transferred cells, the macrophages cells were labeled with carboxyfluorescein succinimidyl ester (CFSE) as described.[25],[26] In brief, the cells were washed extensively (at least 2×) in buffer without FBS, and stained in the 10 μM CFSE solution. Then, the CFSE labeled cells were introduced to the ischemic mice by vein infusion. The cells with green fluorescence were observed in the ischemic brain by confocal laser scanning microscopy.

General histology and immunofluorescence staining

Animals were euthanized at 2 days after MCAO, fixed via transcardial perfusion of 4% paraformaldehyde (PFA). The mouse brains were removed and immersed in the 4% PFA in PBS for 48 h. The brain tissues were cut into slices with 50 μm thickness on a vibratome. The slides were washed three times with PBS, blocked with 5% BSA for 2 h and incubated with anti-mouse iNOS Ab (diluted 1:200; Cambridge, MA USA) and Iba1 Ab (diluted 1:200; Cambridge, MA USA) and YM1 Ab (diluted 1:200; Cambridge, MA USA) for overnight, respectively. After washing, the Goat anti-mouse secondary antibody (Cell Signaling, Danvers, MA) coupled with fluorescence were added and incubated for 2 h at room temperature.

FACS analyses

FACS was used to determine the number of infiltrated macrophages in the ischemic brain, including monocyte-derived macrophages and microglia macrophages, as we reported before.[21],[22],[23],[24] Briefly, brain tissues were collected in RPMI-1640 and homogenized on ice. The suspension was brought to 7 ml with FACS buffer and added with 3 ml 90% Percoll (GE Healthcare, Pittsburgh, PA). After thorough mixing, 1 ml 70% Percoll in PBS was laid under the solution and centrifuged at 2470 rpm for 30 min at 4°C. Leukocytes at the interphase were isolated and washed in RPMI-1640, resuspended with 100 μl of FACS buffer. The cells were immunostained with antibodies against CD45-PE/Cy5 and CD11b-percp-Cy5.5 (BioLegend Inc., San Diego, CA, USA) on ice for 30 min in dark. The stained cells were run on the BD LSR II flow cytometer, and the data were automatedly obtained using BD FACSDiva software (v6.1.2, Becton Dickinson, San Jose, CA). The results were analyzed using the FlowJo software (v7.6.2, Tree Star, Ashland, OR, USA).

Statistical analysis

Data are shown as mean ± standard error. Analysis between groups was examined by ANOVA, followed by Student's t-test. Values of P < 0.05 were considered significant difference.

  Results Top

Characterization of M (LPS) and M (IL-4) derived from BV2 cells

Stimulation with LPS and IL-4 causes dramatic changes in morphology in cultured BV2 microglial cells, as observed at 24 h after stimulation by using the images of phase-contrast microscopy. The shapes of normal BV2 cells were spherical, elliptic, or triangle. LPS stimulation resulted in flat, round, pancake-like morphology with enlarged sizes in M (LPS). IL-4 stimulation resulted in elongation in cell shapes of many M (IL-4) BV2 cells [Figure 1]A. The morphology showed changes similar to some extent in BV2 cells stimulated by LPS as reported.[18],[19]
Figure 1: The BV2 cells were polarized into M (LPS) and M (IL-4) phenotypes.
A, Representative images of control, M (LPS) and M (IL-4) macrophages by using the phase-contrast microscopy in cultured BV2 cells. M (LPS) and M (IL-4) were polarized by LPS and IL-4, respectively. B, The left panel showed representative bands of RT-PCR results for M1 and M2 gene expressions. The M1 markers include TNF-α and iNOS, and the M2 markers include Arg 1, and TGF-β. Total RNA was extracted from cultured M (LPS) and M (IL-4) BV2 cells. The right bar graph showed the statistical results of M1 and M2 gene expression levels. C, Representative imagines of confocal immunostaining for the M1 marker iNOS and M2 marker Arg. 1 in cultured M (LPS) and M (IL-4) BV2 cells.
*, P < 0.05, between the two indicated groups n = 3. LPS: Lipopolysaccharide, IL: Interleukin, TNF: Tumor necrosis factor, iNOS: Inductible nitric oxide synthase, TGF: Transforming growth factor, GAPDH: Glyceraldehyde-3-phosphate dehydrogenase

Click here to view

The results of RT-PCR showed that gene expressions of TNF-α and iNOS, the two M1 markers, were increased in M (LPS) macrophages, while gene expressions of Arginase 1 and TGF-β, the two M2 markers, were increased in M (IL-4) phenotypes [Figure 1]B. These results were consistent with typical M1 and M2 definitions.[3],[9]

The immunostaining results suggested that M (LPS) expressed higher while M (IL-4) expressed lower the M1 marker iNOS. Nevertheless, both M (LPS) and M (IL-4) expressed the M2 marker Arg1 [Figure 1]C. Thus, the results of confocal immunostaining did not provide clear evidence for defining M1 and M2 phenotypes, as both M1 and M2 markers were expressed in M (LPS) and M (IL-4) BV2 cells.

Both anti- and pro-inflammatory genes are expressed in M (LPS) and M (IL-4)

We further examined the expression patterns of more inflammatory genes for the classic M1 and M2 markers, together with other inflammatory genes by using the high-throughput single-cell RT-PCR technique. These genes were divided into 4 clusters: 1) M1 genes, iNOS, Il-1b, Il-6, Tnfa, Ifnb, Mcp-1, Cxcl2, Tspo; 2) M2 genes: Il-4, Il-10, Ym1, Arg-1, Fizz, Tgfb, Lgals3; 3) Transcription gens, Nf-κb, Stat3, Hif-1a, Cebpa, Cebpb, Pparg; and 4) Toll like receptors, Tlr1, Tlr2, Tlr4, Tlr7, Tlr9, Tlr13.

The results showed that most of the M1 genes were drastically increased in M (LPS) phenotype, but some of them, including IL-1β, IL-6, CXCL2, were also robustly increased in M (IL-4) phenotypes. Nevertheless, the M (LPS) phenotype had much higher M1 gene expression levels than M (IL-4) phenotypes. For the cluster of M2 genes, only Arg-1 was drastically increased in the M (IL-4) phenotype, but not in the M (LPS) phenotype. The rest M2 genes were not significantly altered in both M (LPS) and M (IL-4) phenotypes. Transcription factors of STAT3, HIF-1α, and CEBPb were increased in both M (LPS) and M (IL-4) macrophages. However, the expression levels of STAT3 and HIF-1α are significantly higher in M (LPS) than in M (IL-4). In addition, PPARg gene expression was increased in M (IL-4) only. In the cluster of TLR genes, the M (LPS) had higher gene expressions of TLR1, TLR2, and TLR9, while M (IL-4) had higher levels of TLR4 and TLR7 [Figure 2].
Figure 2: The effects of adoptive transfer of M (LPS) and M (IL-4) macrophages on stroke outcomes.
The polarized BV2 cells were adoptively transferred via vein injection at the onset of stroke onset. A, The right panel showed the representative TTC staining of brain infarction measured 2 days after stroke. The bar graph in the right panel showed the statistical results of infarct sizes. B, Neurological scores. The neurological grading scores were measured based on a scale from 0 to 4 in the tests of body symmetry, gait, climbing, compulsory circling, and front limb symmetry, at 24 h and 48 h after stroke onset.
*, P < 0.05, between the two indicated groups, n = 6. iNOS: Inductible nitric oxide synthase, IL: Interleukin, TNF: Tumor necrosis factor, MCP: Macrophage chemoattractant protein

Click here to view

Adoptive transfer of M (LPS) and M (IL-4) cells exacerbates and inhibits ischemic brain injuries, respectively

We then examined whether M (LPS) and M (IL-4) macrophages had distinctive effects on stroke outcomes. The infarct of the mouse brains was determined by TTC staining at day 2 after MCAO. The results showed that the adoptive transfer of the non-polarized BV2 cells did not alter infarct sizes compared with the vehicle. Nevertheless, M (LPS) adoptive transfer significantly increased while M (IL-4) robustly inhibited infarction [P < 0.05; [Figure 3]A]. In addition, we measured neurological scores at 24 h and 48 h after stroke. Inconsistent with the results of infarct sizes, the control BV2 cells did not alter neurological scores, but M (LPS) and M (IL-4) worsened and attenuated neurological scores, respectively [Figure 3]B.
Figure 3: Analyses of gene expressions in polarized M (LPS) and M (IL-4) BV2 cells by the high throughput single-cell RT-PCR technique.
The BV2 cells were sorted onto a 96-well PCR plate for analysis. The 100 cell population were loaded onto plate wells (100 cell per well). Reverse transcription was conducted and cDNA was pre-amplified with primers. The pre-amplified samples were then loaded into a chip of a 48 × 48 plate and underwent real-time PCR using the BioMark HD system. The genes were arbitrarily divided into 4 groups: M1, M2, transcription factors, and toll-like receptors.
*P < 0.05, between the two indicated groups, n = 4. LPS: Lipopolysaccharide, IL: Interleukin

Click here to view

Adoptive transfer of M (IL-4) cells reduced the infiltration of inflammatory microglia/macrophages

By using CFSE staining, we confirmed that the adoptive transfer of BV2 cells entered the ischemic brain [Figure 4]A and through vessels [Figure 4]B. There were no differences between groups. We then examined how the adoptive transfer of BV2 cells affects inflammatory responses featured by the accumulation of microglia/macrophages [Figure 5]. The anti-Iba-1 antibodies recognized macrophages derived from both microglia and monocytes. We quantified Iba-1 positive cells in the ischemic borders of the striatum and cortex by using confocal fluorescent immunostaining. The results suggest that the adoptive transfer with control BV2 cells did not alter the Iba-1-positive cells in both the striatum and cortex. However, the adoptive transfer of M (LPS) increased while that of M (IL-4) reduced the number of Iba-1-positive cells in both the striatum and cortex [Figure 5B].
Figure 4: Representative images of BV2 macrophages infiltrated into the ischemic brain after adoptive transfer.
A, The three types of BV2 cells, M (LPS) and M (IL-4), were labeled with the dye CFSE, and injection via the vein into the animals. The ischemic brain slices were prepared 2 days after stroke and examined with confocal microscopy. The results suggested that the transferred macrophages were detected in the ischemic brain. B, Partial enlarged view.
LPS: Lipopolysaccharide, IL: Interleukin

Click here to view
Figure 5: The effects of macrophage transfer on inflammatory responses.
A, A representative brain slice of infarction, on which the two sites, cortex and striatum, were marked for where the immunostaining pictures were taken. B, Immunostaining of the M1 marker iNOS. The left panel showed representative images of iNOS staining in the cortex, striatum, and contralateral side, from animals receiving vehicle treatment, adoptive transfer of control, M (LPS), and M (IL-4) BV2 cells. The brain tissues were harvested 2 days after stroke. The bar graph in the right panel showed quantitative results of iNOS positive cell numbers in the four groups receiving vehicle and 3 types of BV2 cells. C, Immunostaining of iba1 for identifying macrophages. Representative images and statistical results were shown in the left and right panels.
*, P < 0.05, between the two indicated groups n = 3. LPS: Lipopolysaccharide, IL: Interleukin

Click here to view

We further examined the effects of M (LPS) and M (IL-4) adoptive transfer on the M1 marker iNOS immunostaining, and the results were similar to that of Iba-1 immunostaining, i.e., M (LPS) transfer increased while M (IL-4) reduced iNOS staining [Figure 5]C.

We further used FACS to quantify the effect of M (IL-4) transfer on macrophage subsets. Microglia-derived macrophages (MiDMs) were identified as CD45intCD11b+ cells and monocyte-derived macrophages (MoDMs) as CD45 hiCD11b+ cells [Figure 6]A. The results showed that both MiDMs and MoDMs were significantly decreased in the M (IL-4) treatment groups compared with the control group (non-polarized BV2 cell treatment) after stroke [Figure 6]B. Moreover, BV2 stimulated by IL-4 promoted YM1 expression in the brain [Figure 6]B, which is an M2 marker.
Figure 6: BV2 stimulated by IL-4 decrease monocyte-derived macrophages (MoDMs) and microglia-derived macrophages (MiDMs) infiltration and promote YM1 expression in brain.
A, Quantification of macrophages by using flow cytometry. The left panel showed a representative gating strategy for monocyte-derived macrophages (MoDMs) and microglia-derived macrophages (MiDMs) for immune cells isolated from the ischemic brain, as the populations of CD45hiCD11b+ and CD45lowCD11b+, respectively. The middle and left panels showed the statistical results of the numbers of MoDMs and MiDMs in the ischemic hemisphere. B, YM1 expression in the ischemic brain. CSFE represent BV2 cells and Red represented YM1 positive cells.
*, P < 0.05, between the two indicated groups n = 3-5. IL: Interleukin, CFSE: Carboxyfluorescein succinimidyl ester, DAPI: 4',6-diamidino-2-phenylindole

Click here to view

  Discussion Top

In the current study, we re-visited the concept of M1 versus M2 functional macrophage phenotypes by using cultured BV2 microglia cells and studied their effects on stroke outcomes. Our major goal is to confirm if there are typical M1 and M2 phenotypes in cultured BV2 cells, but not in the ischemic brain after stroke. Our conclusion is that M (LPS) and M (IL-4) indeed show distinctive morphologies and gene expression patterns, but it might be confusing if only a few gene markers are used to define M1 and M2 phenotypes.

We provide strong evidence supporting that M (LPS) and M (IL-4) are similar functional phenotypes to M1 and M2, respectively. First, M (LPS) and M (IL-4) indeed have distinctive morphologies in cultured BV2 cells. Second, M (LPS) has high gene expression of the M1 markers, TNF-α and iNOS, and M (IL-4) has high gene expressions of the M2 markers Arginase 1 and TGF-β. Third, the adoptive transfer of M (LPS) resulted in larger infarction and worsened neurological deficits, and promoted neuroinflammation. Nevertheless, M (IL-4) transfer resulted in smaller infarction and less neurological deficits and inhibited neuroinflammation in stroke animals. Taken together, M (LPS) and M (IL-4) are distinctive macrophage phenotypes.

Nevertheless, it may cause problems and confusion if M (LPS) and M (IL-4) are simply defined as M1 and M2 by using a few typical markers, as have discussed by others.[9],[10],[11] First, we found that both the typical M1 maker iNOS and the M2 marker Arginase 1 were strongly stained in normal BV2 cells, and so did in both M (LPS) and M (IL-4). As expected, iNOS staining was increased in M (LPS) and decreased in M (IL-4). Nevertheless, although Arginase 1 gene expression was robustly increased in M (IL-4) but not in M (LPS), Arginase 1 protein levels cannot be distinguished in M (LPS) and M (IL-4). The discrepancy between gene expression and immunostaining that Arginase 1 mRNA was not translated into protein. These results suggest that it should be interpreted carefully when arginase 1 is used as a marker to define M1 and M2 in cultured BV2 cells. Indeed, macrophages can exhibit anti-inflammatory and proinflammatory properties at the same time and some markers for M2 are also expressed in M1 cells such as MRC1(CD206) and Arg1.[11],[27] Thus, the techniques of determining both gene and protein expressions and their expression at different time points should be considered.

It becomes more problematic to define M1 and M2 phenotypes when gene expressions of multiple markers are assessed. By using the high-throughput RT-PCR techniques, we studied gene expressions of typical M1 and M2 inflammatory genes, transcription factors, and Toll-like receptors in M (LPS) and M (IL-4) macrophage phenotypes. Our results suggest that M (LPS) and M (IL-4) phenotypes differ largely in M1 markers, including iNOS, Il-1b, Il-6, and TNF-α, but most M2 markers did not show significant differences in M (LPS) and M (IL-4) macrophages, except arginase 1. Transcriptional analyses suggest that M (LPS) had higher expression of Stat-1 and Hif-1a, while M2 macrophages had much higher levels of Pparg gene expression. Many other important transcriptional factors did not show difference between M1 and M2 macrophages, including Creb1, NF-κB, Cebpa, and Cebpb. We also analyzed gene expressions of toll-like receptors. The results indicate that M (LPS) had higher Tlr1 and Tlr2 gene expressions than M (IL-4) macrophages, but M (IL-4) had higher gene expression levels of Tlr4, Tlr7, and Tlr9. These results suggest that gene expressions of a few inflammatory markers, including the typical M1 and M2 markers, transcriptional factors, and TLRs, should not be used alone to define M1 and M2 functional phenotypes. Unfortunately, our study has the limitation that we were unable to provide a resolution of how to define M1 and M2 concepts, or whether these concepts should be preserved or discarded. These issues will have to be resolved with more solid and rigorous studies.

Our study suggests that microglia/macrophages can be manipulated for stroke therapy.[28] We have reported that monocyte-derived macrophages are detrimental to acute ischemic stroke but beneficial for neurological recovery after stroke.[21],[22] We also reported that the adoptive transfer of M2-like monocyte-derived macrophages attenuates brain injury after stroke.[22] Our results are consistent with a previous study suggesting that monocytes are essential for functional recovery.[29.30] Nevertheless, another recent study suggests that monocytes had no effect on stroke outcomes.[31]

How BV2 microglial cells affect brain injury remains unclear. The current study showed that the adoptive transfer of normal BV2 cells did not affect stroke outcomes compared with vehicle treatment. Nevertheless, we demonstrated that adoptive transfer of BV2-derived M2 macrophages inhibited infarction and attenuated neurological deficits, while BV2-derived M1 macrophages enlarged infarction and exacerbated neurological deficits. The underlying mechanisms of M (IL-4) treatment remain unknown. Nevertheless, we have shown that it inhibited the infiltration of total macrophages, including both monocytes and microglia-derived macrophages, as well as protein expression of iNOS. In addition, we have also found that M (IL-4) promotes the tube formation of the vascular endothelial cell through secreting exosomes in vitro.[32] At last, it may involve inflammasome-mediated inflammation, which is important in both acute stroke attack and poststroke subacute stage, as well as in vascular cognitive impairment.[33]

In conclusion, distinctive phenotypes of M1 versus M2 macrophages indeed exist in cultured BV2 macrophages with LPS and IL-4 stimulation, but they cannot be defined accurately by using a few typical inflammatory markers. Macrophages may be manipulated into M2-like macrophages for stroke therapy.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Hu X, Leak RK, Shi Y, Suenaga J, Gao Y, Zheng P, et al. Microglial and macrophage polarization-new prospects for brain repair. Nat Rev Neurol 2015;11:56-64. doi: 10.1038/nrneurol. 2014.207.  Back to cited text no. 1
Edwards JP, Zhang X, Frauwirth KA, Mosser DM. Biochemical and functional characterization of three activated macrophage populations. J Leukoc Biol 2006;80:1298-307. doi: 10.1189/jlb. 0406249.  Back to cited text no. 2
Martinez FO, Helming L, Gordon S. Alternative activation of macrophages: An immunologic functional perspective. Annu Rev Immunol 2009;27:451-83. doi: 10.1146/annurev.immunol. 021908.132532.  Back to cited text no. 3
Davies LC, Jenkins SJ, Allen JE, Taylor PR. Tissue-resident macrophages. Nat Immunol 2013;14:986-95. doi: 10.1038/ni.2705.  Back to cited text no. 4
Hu X, Li P, Guo Y, Wang H, Leak RK, Chen S, et al. Microglia/macrophage polarization dynamics reveal novel mechanism of injury expansion after focal cerebral ischemia. Stroke 2012;43:3063-70. doi: 10.1161/STROKEAHA.112.659656.  Back to cited text no. 5
Zarruk JG, Greenhalgh AD, David S. Microglia and macrophages differ in their inflammatory profile after permanent brain ischemia. Exp Neurol 2018;301:120-32. doi: 10.1016/j.expneurol.2017.08.011.  Back to cited text no. 6
Esposito E, Hayakawa K, Ahn BJ, Chan SJ, Xing C, Liang AC, et al. Effects of ischemic post-conditioning on neuronal VEGF regulation and microglial polarization in a rat model of focal cerebral ischemia. J Neurochem 2018;146:160-72. doi: 10.1111/jnc.14337.  Back to cited text no. 7
Luo Q, Fan Y, Lin L, Wei J, Li Z, Li Y, et al. Interleukin-33 protects ischemic brain injury by regulating specific microglial activities. Neuroscience 2018;385:75-89. doi: 10.1016/j.neuroscience.2018.05.047.  Back to cited text no. 8
Martinez FO, Gordon S. The M1 and M2 paradigm of macrophage activation: Time for reassessment. F1000Prime Rep 2014;6:13. doi: 10.12703/P6-13.  Back to cited text no. 9
Nahrendorf M, Swirski FK. Abandoning M1/M2 for a network model of macrophage function. Circ Res 2016;119:414-7. doi: 10.1161/CIRCRESAHA.116.309194.  Back to cited text no. 10
Murray PJ, Allen JE, Biswas SK, Fisher EA, Gilroy DW, Goerdt S, et al. Macrophage activation and polarization: Nomenclature and experimental guidelines. Immunity 2014;41:14-20. doi: 10.1016/j.immuni.2014.06.008.  Back to cited text no. 11
Kawasaki S, Abe N, Ohtake F, Islam A, Choudhury ME, Utsunomiya R, et al. Effects of hypnotic bromovalerylurea on microglial BV2 cells. J Pharmacol Sci 2017;134:116-23. doi: 10.1016/j.jphs.2017.05.007.  Back to cited text no. 12
Lawrimore CJ, Crews FT. Ethanol, TLR3, and TLR4 agonists have unique innate immune responses in neuron-like SH-SY5Y and microglia-like BV2. Alcohol Clin Exp Res 2017;41:939-54. doi: 10.1111/acer.13368.  Back to cited text no. 13
Lim JY, Sul D, Hwang BY, Hwang KW, Yoo KY, Park SY. Suppression of LPS-induced inflammatory responses by inflexanin B in BV2 microglial cells. Can J Physiol Pharmacol 2013;91:141-8. doi: 10.1139/cjpp-2012-0242.  Back to cited text no. 14
Qin Y, Sun X, Shao X, Hu MX, Feng J, Chen Z, et al. Lipopolysaccharide preconditioning induces an anti-inflammatory phenotype in BV2 microglia. Cell Mol Neurobiol 2016;36:1269-77. doi: 10.1007/s10571-015-0324-1.  Back to cited text no. 15
Shafer LL, McNulty JA, Young MR. Brain activation of monocyte lineage cells: Brain-derived soluble factors differentially regulate BV2 microglia and peripheral macrophage immune functions. Neuroimmunomodulation 2002;10:283-94. doi: 10.1159/000069972.  Back to cited text no. 16
Latta CH, Sudduth TL, Weekman EM, Brothers HM, Abner EL, Popa GJ, et al. Determining the role of IL-4 induced neuroinflammation in microglial activity and amyloid-β using BV2 microglial cells and APP/PS1 transgenic mice. J Neuroinflammation 2015;12:41. doi: 10.1186/s12974-015-0243-6.  Back to cited text no. 17
Chen J, Sun Z, Jin M, Tu Y, Wang S, Yang X, et al. Inhibition of AGEs/RAGE/Rho/ROCK pathway suppresses non-specific neuroinflammation by regulating BV2 microglial M1/M2 polarization through the NF-κB pathway. J Neuroimmunol 2017;305:108-14. doi: 10.1016/j.jneuroim.2017.02.010.  Back to cited text no. 18
Dai XJ, Li N, Yu L, Chen ZY, Hua R, Qin X, et al. Activation of BV2 microglia by lipopolysaccharide triggers an inflammatory reaction in PC12 cell apoptosis through a toll-like receptor 4-dependent pathway. Cell Stress Chaperones 2015;20:321-31. doi: 10.1007/s12192-014-0552-1.  Back to cited text no. 19
Sanchez-Freire V, Ebert AD, Kalisky T, Quake SR, Wu JC. Microfluidic single-cell real-time PCR for comparative analysis of gene expression patterns. Nat Protoc 2012;7:829-38. doi: 10.1038/nprot.2012.021.  Back to cited text no. 20
Fan Y, Xiong X, Zhang Y, Yan D, Jian Z, Xu B, et al. MKEY, a peptide inhibitor of CXCL4-CCL5 heterodimer formation, protects against stroke in mice. J Am Heart Assoc 2016;5:e003615. doi: 10.1161/JAHA.116.003615.  Back to cited text no. 21
Fang W, Zhai X, Han D, Xiong X, Wang T, Zeng X, et al. CCR2-dependent monocytes/macrophages exacerbate acute brain injury but promote functional recovery after ischemic stroke in mice. Theranostics 2018;8:3530-43. doi: 10.7150/thno.24475.  Back to cited text no. 22
Gu L, Xiong X, Zhang H, Xu B, Steinberg GK, Zhao H. Distinctive effects of T cell subsets in neuronal injury induced by cocultured splenocytes in vitro and by in vivo stroke in mice. Stroke 2012;43:1941-6. doi: 10.1161/STROKEAHA.112.656611.  Back to cited text no. 23
Joo SP, Xie W, Xiong X, Xu B, Zhao H. Ischemic postconditioning protects against focal cerebral ischemia by inhibiting brain inflammation while attenuating peripheral lymphopenia in mice. Neuroscience 2013;243:149-57. doi: 10.1016/j.neuroscience.2013.03.062.  Back to cited text no. 24
Lyons AB, Parish CR. Determination of lymphocyte division by flow cytometry. J Immunol Methods 1994;171:131-7. doi: 10.1016/0022-1759(94)90236-4.  Back to cited text no. 25
Weischenfeldt J, Porse B. Bone marrow-derived macrophages (BMM): Isolation and applications. CSH Protoc 2008;2008:pdb.prot5080. doi: 10.1101/pdb.prot5080.  Back to cited text no. 26
Cochain C, Vafadarnejad E, Arampatzi P, Pelisek J, Winkels H, Ley K, et al. Single-cell RNA-Seq reveals the transcriptional landscape and heterogeneity of aortic macrophages in murine atherosclerosis. Circ Res 2018;122:1661-74. doi: 10.1161/CIRCRESAHA.117.312509.  Back to cited text no. 27
Ajoolabady A, Wang S, Kroemer G, Penninger JM, Uversky VN, Pratico D, et al. Targeting autophagy in ischemic stroke: From molecular mechanisms to clinical therapeutics. Pharmacol Ther 2021;225:107848. doi: 10.1016/j.pharmthera.2021.107848.  Back to cited text no. 28
Wattananit S, Tornero D, Graubardt N, Memanishvili T, Monni E, Tatarishvili J, et al. Monocyte-derived macrophages contribute to spontaneous long-term functional recovery after stroke in mice. J Neurosci 2016;36:4182-95. doi: 10.1523/JNEUROSCI.4317-15.2016.  Back to cited text no. 29
ElAli A, Jean LeBlanc N. The role of monocytes in ischemic stroke pathobiology: New avenues to explore. Front Aging Neurosci 2016;8:29. doi: 10.3389/fnagi.2016.00029.  Back to cited text no. 30
Schmidt A, Strecker JK, Hucke S, Bruckmann NM, Herold M, Mack M, et al. Targeting different monocyte/macrophage subsets has no impact on outcome in experimental stroke. Stroke 2017;48:1061-9. doi: 10.1161/STROKEAHA.116.015577.  Back to cited text no. 31
Tian Y, Zhu P, Liu S, Jin Z, Li D, Zhao H, et al. IL-4-polarized BV2 microglia cells promote angiogenesis by secreting exosomes. Adv Clin Exp Med 2019;28:421-30. doi: 10.17219/acem/91826.  Back to cited text no. 32
Li YK, Chen JG, Wang F. The emerging roles of absent in melanoma 2 (AIM2) inflammasome in central nervous system disorders. Neurochem Int 2021;149:105122. doi: 10.1016/j.neuint.2021.105122.  Back to cited text no. 33


  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]

  [Table 1], [Table 2]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
Materials and Me...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded15    
    Comments [Add]    

Recommend this journal