Cell Culture

Pure cultures of mouse brain endothelial cells (BEC) were prepared as previously described [19]. Briefly, brains were removed from 8 week-old C57Bl/6 mice, minced, dissociated for one hour in papain, centrifuged through 22% BSA to remove myelin, and then endothelial cells cultured in endothelial cell growth media (ECGM) consisting of Hams F12, supplemented with 10% FBS, Heparin, ascorbic acid, L-glutamine, penicillin/streptomycin (all from Sigma, St. Louis, MO) and endothelial cell growth supplement (ECGS) (Upstate Cell Signaling Solutions, Lake Placid, NY), on type I collagen (Sigma)-coated 6-well plates. Puromycin (4 μg/ml, Alexis GmbH, Grunberg, Germany) was included in culture media between days 1-3 to remove contaminating cell types. Endothelial cell purity was >99% as determined by CD31 in flow cytometry. For all experiments, BEC were used only for the first passage.

Mixed glial cultures were prepared from 0-2 day old C57Bl/6 mouse pups, as previously described [19], and maintained in poly-D-lysine coated T75 flasks in DMEM containing 10% fetal bovine serum (FBS) (all from Sigma). After 7-10 days, flasks were mechanically shaken to yield microglia, which were plated into uncoated 6 well plates. Microglial purity was >99% as determined by Mac-1 in flow cytometry. Pure astrocyte cultures were prepared as previously described [20], by plating neurospheres into poly-D-lysine coated 6-well plates and maintained in DMEM containing 10% FBS. Astrocyte purity of these cultures was >99% as determined by GFAP immunocytochemistry.

Microglia-conditioned media (MG-CM)

Microglia were shaken off mixed glial cultures and plated in 6-well plates overnight. Media was then changed to serum-free DMEM containing N1-supplement, L-glutamine and penicillin/streptomycin (all from Sigma). Microglia were left unstimulated or activated with 1 μg/ml lipopolysaccharide (LPS, Sigma) to produce resting and activated MG-CM respectively. After 3 days, media was collected and filtered through a 0.22 μm filter before use in BEC studies. Astrocyte conditioned media (A-CM) was prepared in a similar manner.

Brain endothelial cell proliferation assay

Glass coverslips were coated with collagen I (10 μg/ml, Sigma) for 2 hours, then washed, and BEC plated and cultured until cells reached ~50% confluence. For investigation of the effect of MG-CM, BEC were cultured in 67% ECGM and 33% MG-CM. Control cultures were maintained in 67% ECGM and 33% N1-supplemented media. In the function-blocking experiments, the anti-TNF and anti-TGF-β1 blocking antibodies and control antibodies (R&D, Minneapolis, MN) were used at 2 μg/ml. For the investigation of the influence of cytokines on BEC proliferation, BEC were cultured in ECGM with factors added across a concentration range, with the maximum indicated: IFN-α (10³ U/ml, PBL Biomedical Labs), IFN-γ (10 U/ml, R&D), IL-6 (20 ng/ml, R&D), TNF (20 ng/ml, Genentech, San Francisco, CA), and TGF-β1 (20 ng/ml, R&D). BEC were cultured for 16 hours in the presence of BrdU (Invitrogen, Carlsbad, CA), then fixed in acid/alcohol and processed for BrdU immunocytochemistry according to the manufacturer's instructions. BEC proliferation was assessed by quantifying the number of BrdU-positive cells as a percentage of the total number of cells (Hoechst staining), and the results expressed as the mean ± SEM of four experiments. Statistical significance was assessed by using Student's t test, in which p < 0.05 was defined as statistically significant.

TNF and TGF-β1 have antagonistic effects on BEC proliferation

Microglia produce a large number of cytokines, whose production is heavily-dependent on the state of microglial activation [21]. To identify which of these factors might be responsible for the microglial influence on BEC proliferation, we screened a panel of different cytokines for their ability to regulate BEC proliferation. This showed that TNF and TGF-β1 had the strongest, though opposing influences on BEC proliferation (Figure ​(Figure2).2). TNF promoted BEC proliferation in a dose-dependent manner, which plateaued at a concentration of 10 ng/ml. Compared to control conditions (21.3 ± 1.4%), 10 ng/ml TNF increased the rate of BEC proliferation to 32.6 ± 4.5% (p < 0.01). In contrast, TGF-β1 reduced the rate of BEC proliferation, an effect which plateaued at a concentration of 10 ng/ml. Compared to control conditions, this concentration of TGF-β1 reduced BEC proliferation from 20.6 ± 1.9% to 13.8 ± 2.1% (p < 0.05). IFN-α, IFN-γ and IL-6 had no significant effect on BEC proliferation at any concentration tested.

Open in a separate window

Figure 2

Quantification of BEC proliferation in response to cytokines. BEC proliferation was examined over 16 hours in the presence of different concentrations of the cytokines, IFN-α (A), IFN-γ (B), IL-6 (C), TGF-β1 (D), or TNF (E). BEC proliferation is expressed as the percentage of BEC that incorporated BrdU; all points represent the mean ± SEM of 4 experiments. Note that BEC proliferation was significantly stimulated by TNF, and inhibited by TGF-β1. p < 0.05.

Microglia secrete TNF and TGF-β1, with the balance determined by microglial activation state

We have demonstrated that BEC proliferation rate is regulated by MG-CM in a biphasic manner; inhibited by resting MG-CM, and promoted by activated MG-CM. Furthermore, we have shown that BEC proliferation is inhibited by TGF-β1, but promoted by TNF. While it is known from previous studies that microglia can produce these two cytokines [22,23], we next used ELISA to investigate in our own cultures, whether MG-CM contained TNF or TGF-β1, and how this expression is regulated by LPS. As shown in Figure ​Figure3A,3A, TNF was present in resting MG-CM (27.7 ± 8.6 pg/ml), but strongly increased (50-fold) in activated MG-CM (1381.2 ± 82.7 pg/ml). In contrast, TGF-β1 levels were not significantly different in resting (196.5 ± 42.3 pg/ml) or activated MG-CM (177.6 ± 43.4 pg/ml) (Figure ​(Figure3B).3B). A-CM contained no TNF, either before or after activation (Figure ​(Figure3A).3A). A-CM did contain TGF-β1, though this expression was not regulated by LPS treatment, and was at lower levels than MG-CM (39.2 ± 23.6 pg/ml and 75 ± 20.5 pg/ml in resting and activated A-CM, respectively). Thus, microglial activation switches the TNF/TGF-β1 balance heavily in favor of TNF.

Open in a separate window

Figure 3

ELISA quantification of TNF (A) and TGF-β1 (B) production by microglia or astrocytes, under resting or LPS-activated conditions. All points are expressed in pg/ml concentration of cytokine and represent the mean ± SEM of 4 experiments. Microglial activation resulted in a massive increase in TNF production (50-fold), while astrocytes produced no TNF, even after LPS stimulation. Activation had no significant influence on TGF-β1 production by either cell type.

Microglial activation and angiogenesis

Microglia, as the primary immune effector cells in the CNS, play important roles in the surveillance and response to pathological insults. As the gatekeeper of the BBB, microglia are well positioned to mount a rapid aggressive response to noxious stimuli that enter the CNS. Under normal conditions, microglia occupy a resting "on surveillance" phenotype, but after stimulation, switch to an activated highly migratory, mitogenic phenotype, with upregulated production of inflammatory cytokines such as TNF and IFN-γ [21,24]. Within days of an ischemic insult, angiogenic remodeling can be observed at the ischemic penumbra [15,16], and interestingly, angiogenic vessels are often surrounded by inflammatory microglia and macrophages [17,18], suggesting that activated microglia and/or macrophages may be instrumental in promoting the angiogenic response to cerebral ischemia. While this idea has not yet been directly tested in the ischemic CNS, traumatic CNS injury leads to activation of microglia and macrophages, and drugs that block activation and proliferation of these cells, also inhibit neovascularization [25]. In addition, studies from outside the CNS support the concept that activated cells of the macrophage lineage directly promote angiogenic events. In tumor models, depletion of monocytes, the blood precursors of tissue macrophages, leads to marked reduction in tumor-associated angiogenesis [26], and conversely, addition of tissue macrophages strongly promotes neovascularization in corneal and rabbit ear chamber models [27,28]. Consistent with this, and in agreement with our own finding, conditioned media taken from activated macrophages directly promotes endothelial cell proliferation in vitro [29].