Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Feb 1.
Published in final edited form as: Stroke. 2009 Dec 17;41(2):350–356. doi: 10.1161/STROKEAHA.109.564492

Brain-Derived Neurotrophic Factor Stimulates Production of Prostacyclin in Cerebral Arteries

Anantha Vijay R Santhanam 1,2, Leslie A Smith 1,2, Zvonimir S Katusic 1,2,*
PMCID: PMC2811757  NIHMSID: NIHMS167009  PMID: 20019327

Abstract

Background

The role of Brain Derived Neurotrophic Factor (BDNF) and its receptor, tropomyosin receptor kinase B (TrkB), in control of cerebral circulation is poorly understood. The present study was designed to investigate the cerebral vascular effects of BDNF in vivo.

Methods

Replication incompetent adenovirus encoding either rat BDNF (AdBDNF) or green fluorescent protein (AdGFP) was injected intracisternally into rabbits. Forty eight hours later, animals were euthanized. Plasma and cerebrospinal fluid (CSF) levels of BDNF were measured by ELISA, vasomotor function of isolated basilar arteries was studied in organ chambers, protein expression in the basilar arteries was studied by Western blotting, prostanoid levels measured by ELISA and cyclic adenosine 3′,5′-monophosphate (cyclic AMP) levels were measured by radioimmunoassay.

Results

The levels of BDNF in the CSF were significantly elevated in AdBDNF-treated rabbits as compared to AdGFP-treated rabbits (37 ± 5 ng/ml vs. 0.006 ± 0.003 ng/ml, respectively, P<0.05, n=14). Western blotting studies revealed that in basilar arteries AdBDNF increased protein expression of prostacyclin (PGI2) synthase, while expression of endothelial nitric oxide synthase (eNOS) and phosphorylated (Ser 1177) eNOS remained unchanged. During incubation with arachidonic acid (1 μmol/L), PGI2 production and levels of cyclic AMP were significantly elevated only in AdBDNF-treated rabbit basilar arteries (P<0.05, n=6). Relaxations to acetylcholine (10−9 to 10−5 mol/L) and arachidonic acid (10−9 to 10−5 mol/L) were significantly potentiated in basilar arteries from rabbits injected with AdBDNF. Potentiation of relaxations to acetylcholine in AdBDNF-treated basilar arteries was inhibited by the non-selective cyclooxygenase inhibitor, indomethacin (10−5 mol/l, P<0.05, n=6) and constitutive phospholipase A2 inhibitor, AACOCF3 (2 × 10−5 mol/L, P<0.05, n=5).

Conclusion

Our results demonstrate that in cerebral arteries, BDNF-induced activation of TrkB receptor signaling in vivo promotes PGI2 biosynthesis. These findings provide novel mechanistic insight into vascular protective effect of BDNF in cerebral circulation.

Keywords: neurotrophins, vasodilatation, basilar arteries, vasomotor function, neurovascular unit

Introduction

Brain Derived Neurotrophic Factor (BDNF) is a member of the family of neurotrophic factors, and it participates in nervous system development.1,2 The effects of BDNF are mediated by activation of neurotrophin receptors which belong to the tropomyosin receptor kinase (Trk) family of tyrosine kinases. BDNF specifically binds to TrkB receptors. In addition, all neurotrophins bind to p75 neurotrophin receptor (p75NTR).3 In the cardiovascular system, expression of BDNF and TrkB have been reported in aortic wall, as well as in the smooth muscle of other large conduit arteries.4,5 Existing evidence suggests that, in the central nervous system, BDNF regulates the homeostatic interaction between neurons, glial cells and the vasculature, collectively referred to as the “neurovascular unit”. Impairment of cerebral vascular BDNF signaling may cause disruption of the neurovascular unit thus leading to progressive neuronal dysfunction.6 It is also important to point out that the effects of BDNF on cerebral circulation have not been studied.

Our previous study identified BDNF as a major cytokine produced and released by endothelial progenitor cells (EPCs).7 More recently, we reported that in cerebral circulation, increased production of PGI2 stimulated by EPCs is caused by their paracrine effect.8 Based on these observations, we hypothesized that, in cerebral arterial wall, BDNF may activate metabolism of arachidonic acid resulting in elevated production of PGI2. In this regard, it is important to notice that PGI2 is primary mediator of endothelium-dependent relaxations in cerebral circulation of infants.9,10 During aging, contribution of PGI2 to endothelial control of vasomotor function decreases so that in adult arteries, nitric oxide becomes dominant endothelium-derived vasodilator.9,10 Relevant to our study, it has been suggested that the inhibitory effect of aging on production of PGI2 increases vulnerability of adult cerebral arteries to vascular injury9. Therefore, reactivation of PGI2 production in adult arteries may help to explain vascular protective effects of EPCs and BDNF.

Methods

Animals

Male New Zealand white rabbits (2–3 kg) were used for experiments. Rabbits were anesthetized with an intramuscular injection composed of ketamine (35 mg/kg), xylazine (5 mg/kg) and acepromazine (2.3 mg/kg) prior to intracisternal injections on day 0. Animals were anesthetized and euthanized on day 2 with intravenous Sleepaway (sodium pentobarbital, 260 mg/kg, Fort Dodge Animal Health), and basilar arteries were isolated and studied. All procedures were approved by the Institutional Animal Care and Use Committee of Mayo Clinic.

Adenovirus and gene transfer

Replication-incompetent adenoviruses constructed using vectors bearing either green fluorescent protein alone (AdGFP; vector-control in this study) and rat BDNF and GFP (AdBDNF) or were generously gifted by Dr. Steve Goldman, University of Rochester Medical Center, Rochester, NY. The detailed methodology for adenoviral construction has been published elsewhere.11 300 μL of CSF was aspirated and mixed with 50 μL vector (109 plaque-forming units [PFU]) or vehicle and injected aseptically using a 25-gauge needle into the cisterna magna.8,12,13 The transduction titer of 109 pfu/rabbit was chosen based on previous in vivo gene transfer studies.12,13 After injection, animals were maintained in a head-down position for 30 minutes before transfer to post-anesthesia recovery.

Measurement of BDNF levels

BDNF levels in the cerebrospinal fluid (CSF) and plasma were measured by an enzyme-linked immuno sorbent assay (ELISA) kit (R & D Biosciences, Catalog DBD00, Minneapolis, MN, USA) according to the manufacturer’s instructions.

Western blotting

Soluble proteins were extracted by mincing and homogenizing basilar arteries in lysis buffer, as described earlier. 8,12,13 Blots were incubated with monoclonal antibodies (1:500 dilution) against phospho Ser1177-endothelial nitric oxide synthase (NOS), inducible NOS (iNOS) (BD Biosciences), cyclooxygenase (COX)-1 (Cayman), COX-2, phospho Trk (BD Biosciences), and polyclonal antibodies to BDNF (Santa Cruz), eNOS (BD Biosciences), TrkB (Neuromics).

Confocal microscopy

Basilar arteries were embedded in paraffin and 5 μm sections were cut. Staining was performed on deparaffinized sections. Nonspecific binding was blocked by incubation of tissue with 10% normal goat serum for 20 minutes. Sections were then incubated with polyclonal antibodies (1:100 dilution) against PGI2 synthase (Santa Cruz Biotech) for 2 hours at room temperature. Texas Red conjugated secondary antibody (Invitrogen) was added to sections for 1 hour at room temperature and slides were incubated for 5 minutes with 10 μg/mL of Hoechst 33258 (Sigma) to stain for nuclei. Cover slips were mounted using Prolong Gold mounting medium (Invitrogen), and the arterial sections were visualized using a Zeiss LSM 510 laser scanning confocal microscope.

Measurement of PGI2 and TxA2

Basilar arteries isolated from rabbits injected either with AdGFP or AdBDNF were incubated in Krebs solution in a CO2 incubator at 37°C for 30 minutes, followed by incubation with arachidonic acid (1 μmol/L) at 37°C for 30 minutes. PGI2 and TxA2 were measured as their stable metabolites, 6-ketoProstaglandin F1α and TxB2, respectively, as described earlier.8

Measurement of cGMP and cAMP

Rabbit basilar arteries were incubated in MEM in a CO2 incubator at 37°C for 30 minutes in 3-isobutyl-1-methylxanthine (IBMX, 10−3 mol/L; Sigma) to inhibit the degradation of cyclic nucleotides by phosphodiesterases, followed by incubation with arachidonic acid (1 μmol/L) at 37°C for 30 minutes. Following incubation, cGMP and cAMP levels were measured by cGMP and cAMP radioimmunoassay kits respectively (Amersham).8

Analysis of vascular reactivity

Isolated arteries were connected to a force transducer for recording of isometric force and placed in organ baths filled with 25mL Krebs solution (37°C; 94% O2/6% CO2; pH 7.4). Concentration-dependent response curves to acetylcholine, diethylammonium (Z)-1-(N,N-diethylamino)diazen-1-IM1,2-diolate(DEA-NONOate) and arachidonic acid (10−9 mol/L to 10−5 mol/L) were cumulatively obtained during submaximal contractions to histamine (3 × 10−7 to 10−6 mol/L).8,12,13 In some experiments, either endothelium was mechanically removed, or endothelium intact arteries were treated with a constitutive phospholipase A2 (cPLA2) inhibitor, arachidonyl trifluoromethyl ketone (AACOCF3, 20 μmol/L, 30 minutes) prior to contracting with histamine, and responses to acetylcholine and arachidonic acid were recorded.

Drugs

DEA-NONOate was obtained from Cayman Chemical and AACOCF3 was obtained from Calbiochem. All other drugs used in the study were obtained from Sigma. The concentration of all drugs is expressed as the final mol/L in the organ chambers. AACOCF3 was dissolved in dimethylsulfoxide and appropriate control experiments were performed with vehicle.

Statistical analysis

Results of the study are expressed as means ± SEM for ‘n’ (n = the number of rabbits in each group) animals used in each experimental group. Relaxations are expressed as percentage of maximal relaxations induced by 3 × 10−4 mol/L papaverine on histamine (3 × 10−7 mol/L to 10−6 mol/L) contracted arteries. cAMP and cGMP values were analyzed by unpaired Students t-test. Densitometric values are expressed as the ratio of studied protein versus actin and comparisons between AdGFP and AdBDNF were assessed by unpaired Students t-test. Concentration-response curves were analyzed by non-linear curve fitting using GraphPad Prism 4.03 software (GraphPad Software Inc, San Diego, CA) and statistical differences among relaxation values across concentration-response curves were analyzed by two-way ANOVA followed by Bonferroni post tests using SigmaStat 3.1 software (Systat Software Inc, San Jose, USA). A P value less than 0.05 was considered statistically significant.

Results

In vivo gene delivery of AdGFP or AdBDNF did not affect blood pressure (Supplemental data, Table 1) and hematological parameters (Supplemental data, Table 2). Subsequent to gene transfer, levels of BDNF in cerebrospinal fluid was increased from 0.006 ± 0.003 ng/ml in AdGFP-transduced rabbits to 36.95 ± 4.8 ng/ml in AdBDNF-transduced rabbits (P<0.001, n=14), while plasma levels of BDNF was increased from 0.010 ± 0.001 ng/ml in AdGFP-transduced rabbits to 0.026 ± 0.003 ng/ml in AdBDNF-transduced rabbits (P<0.05, n=14).

Western blot analysis demonstrated increased expression of BDNF in AdBDNF-transduced arteries in comparison to AdGFP-transduced rabbit basilar arteries (P<0.01, n=8) (Figure 1a). Intracisternal administration of AdBDNF activated TrkB receptors, as reflected in phosphorylation of TrkB (Figure 1b), while the expression of non-specific neurotrophin receptor p75-NTR remained unchanged (data not shown).

Figure 1.

Figure 1

Open in a new tab

a) Representative Western blots and densitometric analysis demonstrating protein expression of BDNF in basilar arteries subsequent to intracisternal AdBDNF or AdGFP gene transfer. Fifty μg of protein obtained from a single basilar artery was loaded in each lane. b) Over-expression of BDNF was associated with phosphorylation of TrkB receptors in AdBDNF-trandsuced basilar arteries (* P<0.05, ** P<0.01, n=6–8).

Over-expression of BDNF in basilar arteries did not alter the expression of endothelial nitric oxide synthase (eNOS), inducible NOS or phosphorylated Ser-1177 eNOS (Figure 2a). However, the protein expression of PGI2 synthase was significantly increased in AdBDNF-transduced arteries (Figure 2b). Immunofluorescence studies confirmed increased expression of PGI2 synthase in smooth muscle cells and adventitial layer of AdBDNF-transduced arteries (Figure 2c). Further characterization of basilar arteries from AdBDNF-treated rabbits demonstrated that expression of COX-1 and COX-2 remained unchanged in comparison to AdGFP-transduced basilar arteries (Figure 2b).

Figure 2.

Figure 2

Figure 2

Open in a new tab

a) Representative Western blots and densitometric analysis demonstrating expression of eNOS, phospho S1177-eNOS, iNOS in basilar arteries of rabbits injected intracisternally with either AdGFP or AdBDNF. b) Representative Western blots and densitometric analysis demonstrating increased expression of PGI2 synthase in basilar arteries of rabbits treated with AdBDNF, while protein expressions of COX-1, COX-2 remained unchanged. Expressions of proteins were normalized to actin and represented as arbitrary units (*P<0.05, n=6–8). c) Representative confocal microscopic images demonstrating expression of PGI2 synthase in AdGFP and AdBDNF-transduced arteries. Please note increased PGI2 synthase expression in the smooth muscle and adventitial layers of AdBDNF-transduced arteries (Bar represents 10 μm, L: lumen, A: adventitia, Magnification, ×40).

On arachidonic acid stimulation, production of 6-ketoprostaglandin F1α, stable metabolite of prostacyclin (Figure 3a) was selectively elevated in AdBDNF-transduced basilar arteries, while levels of thromboxane B2 (Figure 3b) remained unchanged, suggestive of increased activation of PGI2 synthase on BDNF treatment. The levels of cyclic AMP, a second messenger of PGI2, were also significantly elevated in arachidonic acid stimulated basilar arteries from AdBDNF-transduced rabbits in comparison to AdGFP-transduced rabbit basilar arteries (Figure 3c). However, the levels of cyclic GMP, indicative of NO production remained unchanged (Figure 3d).

Figure 3.

Figure 3

Open in a new tab

Arachidonic acid (1 μmol/L) stimulation of AdBDNF-treated arteries demonstrated increased production of 6-keto PGF1α (a), stable metabolite of PGI2. without affecting TxB2 production (b) and elevated the cerebral arterial content of cyclic AMP (c). AdBDNF-treated arteries did not affect the arterial content of cyclic GMP (d). (*P<0.01, ** P<0.001, n=6–8).

In histamine-contracted arteries, relaxations to acetylcholine (10−9 to 10−5 mol/L) were significantly potentiated in AdBDNF-treated arteries (Figure 4a), while relaxations to a NO-donor, DEA-NONOate (10−9 to 10−5 mol/L) remained unaltered (Figure 4b). Relaxations to either acetylcholine or DEA-NONOate were not different between non-transduced control arteries and AdGFP-transduced control basilar arteries (data not shown). Removal of endothelium abolished acetylcholine-induced relaxations in both AdGFP and AdBDNF treated arteries (Figure 4a).

Figure 4.

Figure 4

Open in a new tab

a) In histamine (3 × 10−7 to 10−6 mol/L)-contracted arteries, relaxations to acetylcholine were selectively potentiated in rabbits injected intracisternally with AdBDNF (***P<0.001, n=8–13). Removal of endothelium abolished relaxations (### P<0.001, n=6–13) to acetylcholine in arteries of rabbits transduced with either AdGFP or AdBDNF. 100% = 1.28 ± 0.17 g (AdGFP, endothelium intact [E+]) and 1.42 ± 0.08 g (AdBDNF, E+), 1.24 ± 0.07 g (AdGFP, endothelium denuded [E−]) and 1.01 ± 0.18 g (AdBDNF), respectively. b) Relaxations to DEA-NONOate in histamine-contracted arteries remained unchanged. 100% = 1.20 ± 0.14 g (AdGFP) and 1.43 ± 0.11 g (AdBDNF), respectively (P=ns, n=5–6).

In AdGFP-transduced arteries, relaxations to acetylcholine were reduced in the presence of eNOS inhibitor, L-NMMA (100 μmol/L), but not in the presence of cyclooxygenase inhibitor indomethacin (Figure 5a). However, in AdBDNF-transduced arteries, relaxations to acetylcholine were more sensitive to the inhibitory effect of indomethacin as compared to the effect of L-NMMA, suggestive of involvement of PGI2 in mediating potentiation of relaxations in AdBDNF-transduced arteries (Figure 5b). In the presence of both L-NMMA and indomethacin, relaxations to acetylcholine were almost abolished in basilar arteries from both AdGFP and AdBDNF-treated rabbits (Figures 5a, 5b). To further examine the role of arachidonic acid metabolism in potentiation of acetylcholine-induced relaxations in AdBDNF-treated rabbit basilar arteries, experiments were performed in the presence of a cPLA2 inhibitor, AACOCF3. Relaxations to acetylcholine remained unaffected by cPLA2 inhibitor in AdGFP-treated basilar arteries (Figure 5c), while the relaxation responses to acetylcholine were inhibited by AACOCF3 (P<0.05, n=6) in AdBDNF-treated arteries (Figure 5d).

Figure 5.

Figure 5

Figure 5

Open in a new tab

a) Relaxations to acetylcholine in histamine-contracted basilar arteries from rabbits injected intracisternally with AdGFP were inhibited by L-NMMA (100 μmol/L). Data are expressed as percentage of maximal relaxation induced by 3 × 10−4 mol/L papaverine; 100 % = 1.28 ± 0.17g (AdGFP), 1.45 ± 0.30g (AdGFP + L-NMMA), 1.32 ± 0.17g (AdGFP + Indomethacin), and 1.96 ± 0.29g (AdGFP + L-NMMA + Indomethacin), respectively (*P<0.05, ** P<0.01, *** P<0.001, n=6). b) Relaxations to acetylcholine in histamine-contracted basilar arteries from rabbits injected intracisternally with AdBDNF demonstrated sensitivity to L-NMMA as well as to indomethacin (10−5 mol/L,). 100 % = 1.42 ± 0.08g (AdBDNF), 1.62 ± 0.14g (AdBDNF + L-NMMA), 1.51 ± 0.19g (AdBDNF + Indomethacin), and 2.28 ± 0.38g (AdBDNF + L-NMMA + Indomethacin), respectively (*P<0.05, **P<0.01, *** P<0.001, n=6). c) Relaxations to acetylcholine remained unchanged in basilar arteries from AdGFP-treated rabbits on incubation with cPLA2 inhibitor, AACOCF3 (20 μmol/L, 30 minutes). 100% = 1.44 ± 0.20 g (AdGFP) and 1.51 ± 0.19 g (AdGFP + AACOCF3), respectively (P=ns, n=5). d) Potentiation of relaxations in AdBDNF-treated arteries were abolished in the presence of cPLA2 inhibitor, AACOCF3. 100% = 1.60 ± 0.22 g (AdBDNF) and 1.36 ± 0.14 g (AdBDNF + AACOCF3), respectively (*P<0.05, **P<0.01, n=5).

In agreement with enhanced metabolism of arachidonic acid favoring PGI2 production in AdBDNF-transduced arteries, concentration-dependent relaxations to arachidonic acid (10−9 to 10−5 mol/L) were selectively potentiated in AdBDNF-treated arteries (Figure 6a). In basilar arteries of rabbits transduced with AdGFP, removal of endothelium significantly attenuated arachidonic acid-induced relaxations (Figure 6b). In comparison, arachidonic acid-induced relaxations remained potentiated in endothelium-denuded AdBDNF-treated basilar arteries (Figure 6b).

Figure 6.

Figure 6

Open in a new tab

a) Concentration-dependent relaxations to arachidonic acid in histamine-contracted arteries were potentiated in endothelium intact basilar arteries of rabbits transduced with AdBDNF. 100% = 1.64 ± 0.15 g (AdGFP, endothelium intact [E+]) and 1.78 ± 0.16 g (AdBDNF, E+), respectively (*P<0.05, **P<0.01, n=5). b) Relaxations to arachidonic acid remained potentiated in endothelium denuded basilar arterial rings obtained from rabbits injected intracisternally with AdBDNF. 100% = 1.01 ± 0.18 g (AdGFP, endothelium denuded [E−]) and 1.11 ± 0.09 g (AdBDNF, E−), respectively (*P<0.05, **P<0.01, n=5–6).

Discussion

To the best of our knowledge, this is the first study to investigate the effects of BDNF on cerebrovascular function. Our study presents several novel findings. First, intracisternal administration of AdBDNF increased BDNF levels in the perivascular space and protein expression of BDNF in cerebral arterial wall. Second, in arteries transduced with AdBDNF, selectively increased expression of PGI2 synthase was observed. Third, stimulation with arachidonic acid favored production of prostacyclin over thromboxane A2 and increased the levels of cyclic AMP consistent with BDNF-induced activation of PGI2 synthase. Fourth, relaxations to acetylcholine and arachidonic acid were selectively potentiated in AdBDNF-treated arteries, while relaxations to a NO donor, DEA-NONOate remained unchanged. Fifth, exposure to BDNF amplified the indomethacin-sensitive component of acetylcholine-induced relaxation in rabbit cerebral arteries. Finally, potentiation of relaxations to acetylcholine in basilar arteries exposed to BDNF was sensitive to cPLA2 inhibition.

Over the last two decades, research on BDNF had been focused on neuronal differentiation and neuroprotection. Despite the identification of BDNF receptor TrkB on vascular endothelial and smooth muscle cells, knowledge of the functional role of BDNF had been primarily restricted to cell survival and angiogenesis. More importantly, in vivo studies on cerebrovascular function of recombinant BDNF were hindered by its short half life and inability to cross blood-brain barrier. Our laboratory has extensive experience with adenovirus-mediated gene transfer in the cisterna magna, and we have reported that adventitial gene delivery results in persistent expression of recombinant proteins in adventitial fibroblasts in the cerebral arterial wall.14,15 In the present in vivo study, using adenovirus-mediated gene transfer, we achieved high perivascular concentrations of recombinant BDNF (~35 ng/mL). Consistent with our findings, in vitro studies have reported BDNF-mediated neuroprotection or anti-apoptotic effect on cultured endothelial cells in concentration range of 10 to 100 ng/mL.16,17 We also observed that intracisternal injections of AdBDNF resulted in increased cerebral arterial expression of BDNF and subsequent activation of TrkB receptors.

By screening vascular protective proteins expressed in the cerebral arterial wall exposed to BDNF, we identified selective up-regulation of PGI2 synthase. Increased activity of PGI2 synthase was confirmed by selective increase in production of PGI2 as well as elevation of cAMP levels and enhanced vasodilatation mediated by PGI2. In the rabbit basilar artery, endothelium-dependent relaxation to acetylcholine is primarily mediated by nitric oxide. Cyclooxygenase blockade by indomethacin has minimal effect on vasodilatation induced by acetylcholine.18,19 Our results with AdGFP-transduced control basilar arteries are in agreement with these reports as indomethacin did not significantly inhibit the maximal relaxations to acetylcholine. Lack of inhibition with cPLA2 inhibitor on acetylcholine-induced relaxations further confirmed nitric oxide playing a predominant vasodilator role in basilar arteries of rabbits exposed to AdGFP. It is likely that, under physiological conditions, activation of cholinergic receptors on endothelium does not couple to the production and release of vasodilator prostanoids. However, on BDNF treatment, acetylcholine-induced relaxations were sensitive to indomethacin as well as to cPLA2 inhibition, suggestive of involvement of arachidonic acid – derived metabolites in the augmentation of relaxations. Inhibition of acetylcholine-induced relaxations in endothelium-denuded arteries from BDNF-treated rabbits suggests role of PGI2 derived by activation of muscarinic receptors on the endothelium. Nevertheless, it is possible that up-regulation of PGI2 synthase is not limited to the endothelium but occurs in smooth muscle cells as well. Morphological studies by confocal microscopy indeed demonstrated that up-regulation of PGI2 synthase was observed in the endothelial, medial and adventitial layers of cerebral arteries obtained from AdBDNF-treated rabbits.

Arachidonic acid induces endothelium-dependent dilation of cerebral blood vessels,20,21 through endothelial lipoxygenase, endothelium-dependent hyperpolarization and by stimulating production of the vasodilator prostanoid, prostacyclin. In agreement, relaxations to arachidonic acid in AdGFP-transduced arteries were significantly reduced on endothelium denudation in the present study. However, in arteries exposed to BDNF, relaxations to arachidonic acid remained potentiated despite endothelium removal. It is likely that, in agreement with our morphological studies, upregulation of PGI2 synthase in medial and adventitial layers of arteries exposed to BDNF contributed to the observed potentiation. Production of 6-keto PGF1α and cAMP levels were increased on arachidonic acid treatment consistent with increased activation of PGI2 synthase and production of prostacyclin. The fact that relaxation to NO donor, DEA-NONOate, was not affected in AdBDNF-treated arteries suggests the effect of BDNF is selective toward PGI2-adenylate cyclase signaling and resulting elevation of cyclic AMP.

Our results significantly expand understanding of the mechanisms underlying cerebrovascular protective effects of BDNF, as PGI2 may also prevent platelet aggregation and thrombus formation, inhibit smooth muscle proliferation, and stimulate therapeutic angiogenesis.22,23 Upon binding to TrkB receptors, BDNF could activate one or more of these three pathways: Ras/Rap-MAPK, PI3K-Akt, and the PLCγ-PKC cascades.24 The molecular targets stimulated by BDNF, mediating production of PGI2 and cerebrovascular protection requires further investigation.

We would also like to point out that in cerebral circulation, high production of PGI2 has been detected in arteries obtained from young children and newborn pigs.9,10 Moreover, with maturation, production of PGI2 in cerebral arterial wall is decreasing thereby increasing reactivity to vasoconstrictors. 9,10 Results of the present study suggest that in adult cerebral arteries, BDNF favors production of PGI2 thus rejuvenating cerebral arterial wall by enhancing vasodilator capacity and protecting against vasoconstrictor stimuli. In addition, increased local concentration of PGI2 in arterial wall is known to activate pro-survival signaling by activation of peroxisome proliferator-activated receptor delta (PPARδ).25 This in turn, may increase resistance of cerebral circulation against injury.

Findings of the present study are the first to demonstrate that, in cerebral arteries, BDNF promotes vasodilatation. PGI2 biosynthesis appears to be the most-likely mediator promoting vasodilatation. In addition to its existing trophic role in the ‘neurovascular unit’, our study adds substantial evidence to the concept that, in cerebral circulation, BDNF exerts beneficial effects by favoring arachidonic acid metabolism via PGI2 synthase.

Supplementary Material

Supp1

Acknowledgments

This work was supported in part by National Heart, Lung, and Blood Institute Grants HL-53524 and HL-91867 (ZSK), the American Heart Association Scientist Development Grant #0835436N (AVRS), and the Mayo Foundation.

References

  • 1.Barde YA. Trophic factors and neuronal survival. Neuron. 1989;2:1525–1534. doi: 10.1016/0896-6273(89)90040-8. [DOI] [PubMed] [Google Scholar]
  • 2.Lewin GR, Barde YA. Physiology of the neurotrophins. Ann Rev Neurosci. 1996;19:289–317. doi: 10.1146/annurev.ne.19.030196.001445. [DOI] [PubMed] [Google Scholar]
  • 3.Caporali A, Emanueli C. Cardiovascular actions of neurotrophins. Physiol Rev. 2009;89:279–308. doi: 10.1152/physrev.00007.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Donovan MJ, Lin M, Wiegn P, Ringsted T, Kraemer R, Hahn R, Wang S, Ibanez C, Rafii S, Hempstead BL. Brain derived neurotrophic factor is an endothelial cell survival factor required for intramyocardial vessel stabilization. Development. 2000;277:2353–2359. doi: 10.1242/dev.127.21.4531. [DOI] [PubMed] [Google Scholar]
  • 5.Scarisbrick IA, Jones EG, Isackson PJ. Coexpression of mRNAs for NGF, BDNF, and NT-3 in the cardiovascular system of the pre- and postnatal rat. J Neurosci. 1993;13:875–893. doi: 10.1523/JNEUROSCI.13-03-00875.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Guo S, Kim WJ, Lok J, Lee SR, Besancon E, Luo BH, Stins MF, Wang X, Dedhar S, Lo EH. Neuroprotection via matrix-trophic coupling between cerebral endothelial cells and neurons. Proc Natl Acad Sci USA. 2008;105:7582–7587. doi: 10.1073/pnas.0801105105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.He T, Smith LA, Harrington S, Nath KA, Caplice NM, Katusic ZS. Transplantation of circulating endothelial progenitor cells restores endothelial function of denuded rabbit carotid arteries. Stroke. 2004;35:2378–2384. doi: 10.1161/01.STR.0000141893.33677.5d. [DOI] [PubMed] [Google Scholar]
  • 8.Santhanam AV, Smith LA, He T, Nath KA, Katusic ZS. Endothelial progenitor cells stimulate cerebrovascular production of prostacyclin by paracrine activation of cyclooxygenase-2. Circ Res. 2007;100:1379–1388. doi: 10.1161/01.RES.0000265848.55035.5d. [DOI] [PubMed] [Google Scholar]
  • 9.Charpie JR, Schreur KD, Papadopoulos SM, Webb RC. Endothelium dependency of contractile activity differs in infant and adult vertebral arteries. J Clin Invest. 1994;93:1339–1343. doi: 10.1172/JCI117093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Willis AP, Leffler CW. Endothelial NO and prostanoid involvement in newborn and juvenile pig pial arteriolar vasomotor responses. Am J Physiol Heart Circ Physiol. 2001;281:H2366–H2377. doi: 10.1152/ajpheart.2001.281.6.H2366. [DOI] [PubMed] [Google Scholar]
  • 11.Benraiss A, Chmielnicki E, Lerner K, Roh D, Goldman SA. Adenoviral brain-derived neurotrophic factor induces both neostriatal and olfactory neuronal recruitment from endogenous progenitor cells in the adult forebrain. J Neurosci. 2001;21:6718–6731. doi: 10.1523/JNEUROSCI.21-17-06718.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Santhanam AV, Smith LA, Nath KA, Katusic ZS. In vivo stimulatory effect of erythropoietin on endothelial nitric oxide synthase in cerebral arteries. Am J Physiol Heart Circ Physiol. 2006;291:H781–H786. doi: 10.1152/ajpheart.00045.2006. [DOI] [PubMed] [Google Scholar]
  • 13.Santhanam AV, Smith LA, Akiyama M, Rosales AG, Bailey KR, Katusic ZS. Role of endothelial NO synthase phosphorylation in cerebrovascular protective effect of recombinant erythropoietin during subarachnoid hemorrhage-induced cerebral vasospasm. Stroke. 2005;36:2731–2737. doi: 10.1161/01.STR.0000190021.85035.5b. [DOI] [PubMed] [Google Scholar]
  • 14.Chen AFY, Jiang SW, Crotty TB, Tsutsui M, Smith LA, O’Brien T, Katusic ZS. Effects of in vivo adventitial expression of recombinant endothelial nitric oxide synthase gene in cerebral arteries. Proc Natl Acad Sci USA. 1997;94:12568–12573. doi: 10.1073/pnas.94.23.12568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Tsutsui M, Chen AFY, O’Brien T, Crotty TB, Katusic ZS. Adventitial expression of recombinant endothelial nitric oxide synthase gene restores nitric oxide production in cerebral arteries without endothelium. Arterioscler Thromb Vasc Biol. 1999;18:1231–1241. doi: 10.1161/01.atv.18.8.1231. [DOI] [PubMed] [Google Scholar]
  • 16.Kim H, Li Q, Hempstead BL, Madri JA. Paracrine and autocrine functions of brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF) in brain-derived endothelial cells. J Biol Chem. 2004;279:33538–33546. doi: 10.1074/jbc.M404115200. [DOI] [PubMed] [Google Scholar]
  • 17.Wang YD, Hu Y, Sun CY, He WJ, Zhang XP. Involvement of AKT/eNOS in brain derived neurotrophic factor-induced angiogenesis. Zhonghua Xue Ye Xue Za Zhi. 2006;27:529–533. [PubMed] [Google Scholar]
  • 18.Sercombe R, Sercombe C, Oudart N, Seylaz J. Critical role of endothelial nitric oxide synthase and cyclooxygenase in response of rabbit basilar artery to serotonin. Jpn J Pharmacol. 2002;90:67–76. doi: 10.1254/jjp.90.67. [DOI] [PubMed] [Google Scholar]
  • 19.Mackert JL, Parsons AA, Wahl M, Schilling L. Mediation of endothelium-dependent relaxation: Different response patterns in rat and rabbit basilar artery. Neurol Res. 1997;19:521–526. doi: 10.1080/01616412.1997.11740851. [DOI] [PubMed] [Google Scholar]
  • 20.Faraci FM. Reactive oxygen species: influence on cerebral vascular tone. J Appl Physiol. 2006;100:739–743. doi: 10.1152/japplphysiol.01044.2005. [DOI] [PubMed] [Google Scholar]
  • 21.Ospina JA, Duckles SP, Krause DN. 17β-estradiol decreases vascular tone in cerebral arteries by shifting COX-dependent vasoconstriction to vasodilation. Am J Physiol Heart Circ Physiol. 2003;285:241–250. doi: 10.1152/ajpheart.00018.2003. [DOI] [PubMed] [Google Scholar]
  • 22.Brian JE, Jr, Moore SA, Faraci FM. Expression and vascular effects of cyclooxygenase-2 in brain. Stroke. 1998;29:2600–2606. doi: 10.1161/01.str.29.12.2600. [DOI] [PubMed] [Google Scholar]
  • 23.Mitchell JA, Warner TD. COX isoforms in the cardiovascular system: understanding the activities of non-steroidal anti-inflammatory drugs. Nat Rev Drug Disc. 2006;5:75–86. doi: 10.1038/nrd1929. [DOI] [PubMed] [Google Scholar]
  • 24.Arevalo JC, Wu SH. Neurotrophin signaling: many exciting surprises. Cell Mol Life Sci. 2006;63:1523–1537. doi: 10.1007/s00018-006-6010-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.He T, Lu T, d’Uscio LV, Lam CF, Lee HC, Katusic ZS. Angiogenic function of prostacyclin biosynthesis in human endothelial progenitor cells. Circ Res. 2008;103:80–88. doi: 10.1161/CIRCRESAHA.108.176057. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supp1
NIHMS167009-supplement-Supp1.pdf (30.1KB, pdf)

RESOURCES