1. Introduction
BDNF is a member of the neurotrophin family that preferentially binds to TrkB, leading to neuronal survival, long term potentiation (LTP) and synaptic plasticity. The precursor of BDNF (proBDNF) is cleaved by either intracellular or extracellular enzymes to form mature BDNF (mBDNF). proBDNF can also release and bind to p75NTR with high affinity, leading to neuronal death and long term depression (LTD) [1] [2] [3]. Researchers have analogized the opposing effects of proBDNF and mature BDNF on neurotrophic function with ancient Chinese “Yin and Yang” theory [4]. The philosophy of Yin (阴) and Yang (阳) formed in ancient China describes a complementary, interconnected, and interdependent relationship between two opposites [5]. The balance of proBDNF and mBDNF is maintained through proteolytic processing of proBDNF. Neurotrophins have been widely studied for their functions in the nervous system. However, it is becoming increasingly clear that neurotrophins also play crucial roles in the vascular system. Recent studies indicated that nerve growth factor (NGF) induced HUVEC proliferation in vitro [6]. In a Chorioallantoic Membrane (CAM) assay, NGF increased angiogenesis in a dose-dependent manner [7]. In Matrigel plug assay, NGF also promoted endothelial cell invasion and cord formation [8]. In recent work, NGF has been shown to increase cell proliferation and migration of the human endothelial cell line [9]. Similarly, BDNF displayed a direct role in angiogenesis. BDNF promoted endothelial cell survival and induced angiogenesis through the TrkB receptor [10]. BDNF increased angiogenic tube formation in HUVEC [11]. Overexpression of BDNF in a mouse endothelial cell line can promote cell proliferation, migration, and invasion [12]. NGF and BDNF also play an indirect role in angiogenesis by inducing vascular endothelial growth factors (VEGF) expression in different cell types [13] [14] [15].
Angiogenesis is the process of new blood vessels growing from the pre-existing vasculature. Angiogenesis is a critical physiological process during embryonic development and into adult life and also plays vital roles in the pathophysiological processes, such as tumour progression, proliferative retinopathy, cardiovascular disease, etc. [16]. The process of angiogenesis is regulated by the balance between pro-angiogenic and anti-angiogenic factors [17]. When the pro-angiogenic regulators exceed the effects of angiostatic regulators, angiogenesis occurs. For example, in response to pathological conditions such as low pO2, injury, tumour or cerebral ischemia, the “angiogenic switch” will be activated and endothelial proliferation increases dramatically [18]. Endogenous modulators such as VEGFs, angiotensin II, fibroblast growth factors (FGFs), or the chemokine are released, and endothelial cells start to loosen their junctions. The endothelial cell layer becomes more permeable, and plasma proteins extravasate to form the extracellular matrix (ECM) scaffold. Then endothelial cells migrate to this ECM surface, where they arrange and form capillary structures [19] [20]. Research shows that angiogenesis is the crucial step for recovery after cerebral ischemic injury to meet the oxygenation requirements of local tissues. Cerebral vasculature-based therapy with neuroprotection (neuro-restorative therapies) has been combined for a better method for future clinical ischemic stroke therapies [21].
Angiogenesis intervention is also an evolving strategy for treating cancer by working as a powerful adjunct to traditional radiotherapy and chemotherapy [22]. Unlike conventional cancer therapies, anti-angiogenic therapies target tumour-associated endothelial cells and angiogenesis, blocking the nutrition line for tumour cell growth. Tumour-related angiogenesis is initiated by a cascade of pro-angiogenic factors and involves close contact between tumour cells and host tissues. Inhibiting the pro-angiogenic process effectively and precisely has been a significant challenge for clinicians to develop optimised anti-angiogenic drugs for cancer. Recent studies in both tumour and endothelial cells have presented a wide variety of molecular targets for developing angiogenesis inhibitors against cancer and other angiogenic-related diseases [20] [23]. Several VEGF blockers have been used in the clinic for treating cancer and eye disease. However, low response rates, rapid vascular regrowth, and the abnormality of tumour vessels made anti-angiogenic therapy more difficult than expected [24] [25]. Optimization and combination of different anti-angiogenesis molecules are future directions for improving the efficacy and efficiency of cancer treatments.
The precursors of neurotrophins usually play opposite roles to mature neurotrophins in the nervous system. However, the role of pro-neurotrophin in angiogenesis remains unclear. In the present study, the effects of proBDNF on endothelial cells and blood vessel formation have been investigated to provide another molecular target for angiogenesis intervention and broaden the choices for treating angiogenesis-related diseases.
2. Material and Methods
1) HUVEC Culture
Human Umbilical Vein Endothelial Cells (HUVEC), Gibco Medium 200 and Large Vessel Endothelial Supplement (LVES) were supplied by ThermoFisher Scientific. LVES is optimized for use with Medium 200 to deliver optimal HUVEC growth and enable superior cell performance. Cryopreserved HUVEC were rapidly thawed in a 37˚C water bath and cultured in the complete media. HUVEC were seeded at a density of 2.5 × 103 cells/cm2 and incubated at 37˚C in a humidified atmosphere of 5% CO2. The cell culture medium was replaced at 24 to 36 hours after seeding and every other day until the culture reached 80% confluence (4 to 6 days).
2) Immunocytochemistry
HUVECs were seeded on sterile glass coverslips in 24-well plates and were grown until they reached confluency. The cells were fixed in situ with 4% paraformaldehyde (PFA) and rinsed with phosphate-buffered saline (PBS). The cells were then permeabilized with 0.1% Triton X-100 and incubated in 10% Normal Goat Serum for blocking. The cells were immuno-stained with different primary antibodies: sheep anti-proBDNF (made by our lab), sheep anti-mBDNF (made by our lab), goat anti-p75NTR (Santa Cruz), rabbit anti-sortilin (Abcam), goat anti-TrkB (R & D systems), and vWF (Abcam) for 45 min at room temperature. Goat IgG was used for negative immunofluorescent control. After washing, different fluorescent secondary antibodies (Alexa-cy3-conjugated donkey anti-goat antibody or Alexa-488-conjugated sheep anti-rabbit antibody, Invitrogen, Inc.) were added and incubated for 40 min in the dark at room temperature. After washing, nuclei were counterstained with 4’,6-Diamidino-2-phenylindole DAPI (Invitrogen, Inc.). Double staining of proBDNF with p75NTR and proBDNF with Sortilin were performed by using mixed primary antibodies from different species (sheep-anti-ProBDNF with rabbit-anti-p75NTR or sheep-anti-proBDNF with rabbit-anti-Sortilin). Corresponding mixed secondary antibodies were applied after washing. The samples were mounted with glycerine and imaged using confocal microscopy (Zeiss LSM 710, Oberkochen, Germany) at different magnifications.
3) Western blot analysis
Total protein from HUVEC was obtained to perform western blot analysis. Briefly, HUVEC were gently washed with chilled PBS and then lysed by suspending in ice-cold RIPA buffer with inhibitors (Roche, Castle Hill, NSW, Australia). After sonication, cells were then centrifuged at 16,000 g for 10 min at 4˚C. The supernatants were carefully collected and total protein concentrations were measured by a BCA protein assay kit (Thermo Scientific, Rockford, USA). Protein extracts from HUVEC were separated by 10% SDS-PAGE gels (Bio-Rad, USA) and transferred to a nitrocellulose membrane (GE Healthcare, Life Sciences). The membrane was then blocked with 5% skim milk for one hour at room temperature and then incubated overnight at 4˚C with respective primary antibodies: sheep anti-proBDNF (made by our lab), goat anti-TrkB (R & D system), rabbit anti-Sortilin (Abcam), goat anti-p75NTR (Santa Cruz). After washing with TBS, subsequent secondary antibodies conjugated with HRP (Sigma-Aldrich) were added and incubated at room temperature for one hour. The membrane was washed three times, and ECL reagent (GE Healthcare Life Sciences) was utilized to develop the bands. Image Quant LAS 4000 was used to take the images (GE Healthcare Australia).
4) MTT Assay
MTT (3-(4,5-Dimethylthiazol-2-yl)-2–5-diphenyltetrazolium bromide) assay is a colorimetric essay, which is commonly used for measuring cell viability, proliferation, and cytotoxicity. MTT, a yellow tetrazolium salt, can be converted to purple formazan by NAD(P)H-dependent oxidoreductase enzymes in living cells. In contrast, dead cells cannot cleave significant amounts of MTT to formazan [26]. The insoluble formazan crystals are then dissolved using a solubilization solution, and the resulting-coloured solution can be quantified by measuring absorbance at 400 to 650 nm. HUVECs were equally seeded in a 96 well plate with complete medium (density: 1 - 5 × 106 cells per mL) and incubated overnight. The complete medium was then replaced by conditioned culture medium: medium 200 (control group), medium 200 with BSA (100 ng/mL), medium 200 with a series of concentrations of proBDNF (1 ng/mL, 3 ng/mL, 10 ng/mL, 30 ng/mL, and 100 ng/mL) and medium 200 with proBDNF 100 ng/mL and anti-proBDNF (10 µg/mL). After incubating for an hour, 10 μL MTT (3 mg/mL in PBS; Sigma-Aldrich) were added to each well and incubated for 4 hours. After carefully aspirating the MTT medium, 150 µl dimethyl sulfoxide (DMSO) was added to each well. The plate was then shaken at 150 rpm for 5 min on a platform rocker to mix the formazan into the solvent thoroughly. The optical density value was recorded at 490 nm on a Microplate Reader (Bio-Rad, USA). Pentad wells were evaluated for each sample, and the test was repeated twice.
5) Endothelial cell tube formation assay (In Vitro Angiogenesis assay)
Endothelial cell tube formation assay is a well-established in vitro angiogenesis assay that utilises the ability of endothelial cells to form capillary-like structures. When reaching 70% - 80% confluence in a T25 flask, HUVECs were starved by depriving serum for 3 - 6 hours prior to performing the tube formation assay [27]. An appropriate volume of Geltrex matrix (ThermoFisher Scientific) was thawed at 4˚C one day before use. Geltrex matrix is a reduced growth factor basement membrane matrix that allows endothelial cells to form tube-like structures. Geltrex matrix was added to a pre-chilled 96-well plate using pre-chilled pipette tips (50 μL/cm2) and then kept at room temperature for 1 hour to polymerize. HUVEC were seeded (42,000 viable cells/cm2) on the matrix gel-coated plate using Medium 200 with 1% foetal bovine serum and incubated at 37˚C and 5% CO2 for 14 - 16 hours. The medium was supplemented with or without proBDNF (50 ng/mL). After incubation, the cell dye Calcein, AM (ThermoFisher Scientific) was added (2 µg/mL) and incubated for 30 minutes. The nonfluorescent Calcein AM is converted to a green fluorescent Calcein in live cells. The dye-containing media were gently removed and replaced with the medium 200. Phase-contrast images and fluorescent images were taken using a light microscope and fluorescence microscope, respectively. Triplicate wells were used for both control and proBDNF groups and repeated twice. Total mesh areas and total branch lengths of capillary networks were measured by means of the Angiogenesis Analyzer plugin for ImageJ [28].
6) Matrigel plug assay (In vivo angiogenesis assay)
The animal experiment was approved by the Animal Welfare Committee of Flinders Medical Centre and conducted in accordance with the National Health and Medical Research Council of Australia. C57Bl/6J mice were used for Matrigel plug assay. All animals were bred under a 12-hour light/dark cycle and had free access to food and water. Twelve mice were randomly divided into four groups of 3 mice in each group (PBS, proBDNF, mBDNF, and anti-proBDNF). Mice were subcutaneously injected in the flank of mice bilaterally, with 0.3 mL PBS, proBDNF, mBDNF, or anti-proBDNF (concentration, 100 µg/mL), combined with 0.3 ml Matrigel, respectively (Becton Dickinson Biosciences). After seven days, mice were humanly killed, and the Matrigel gel plug was harvested for staining.
7) Hematoxylin and eosin staining
The abdominal skin (5-mm width) around the Matrigel plug, including the skin, muscle, and peritoneum, was removed for H&E staining. The samples were fixed in 4% PFA overnight at room temperature. Graded ethanol dehydration and xylene were then applied before embedding in paraffin wax blocks. After cutting on a microtome, the sections went through a hydration process and stained in hematoxylin for 10 min, followed by eosin for 5 min. After dehydration treatment again, the slides were mounted with a coverslip and mounting medium and observed under the microscope. Scoring of tissue: Angiogenesis in Matrigel plug tissue was assessed based on a three-point scale as follows: 1) No obvious or few cells were observed in the Matrigel; 2) Significant cells were invading the Matrigel; 3) Vascular structures presented in Matrigel, with substantial cell invasion [29].
8) Immunohistochemistry
After cutting, sections from the Matrigel plug assay were processed through xylene and an ethanol series. After rehydration, heat-induced epitope recovery was performed by placing slides into 0.01 M Citrate buffer (pH 6.0) at 98˚C in a water bath for 15 min. After cooling down and washing with PBS, slides were treated with 3% H2O2 for 5 min to block endogenous peroxidase. 10% Normal Goat Serum blocking solution was then applied to sections for 30min to block non-specific antigen binding. Without washing, Primary antibody Rat Anti-Mouse CD31 (BD Pharmingen) and mouse monoclonal α-Smooth Muscle antibody (α-SMA) (Sigma-Aldrich) were added respectively and incubated at 4˚C overnight. After washing with PBS, corresponding biotinylated secondary antibodies were added for 30 min at room temperature. ABC (Vector Lab) solution was then added after washing and incubated for 1 hour at room temperature. DAB (diaminobenzidine) colour generation system was used for chromogenic visualization. The expression of CD31 and α-SMA were observed and photographed using a microscope. Positively stained area per unit area was analysed among different groups by ImageJ software (NIH, USA).
9) Statistical analyses
The results were reported as M ± SD, and the statistical significance was analysed by GraphPad Prism 7 software (USA). Parametric tests were applied after testing for normal distributions and equal variances. Student t-test was used for comparing the difference between two groups, and one-way ANOVA followed by Tukey’s multiple comparison post hoc test was utilized when comparing differences among more than two groups. p < 0.05 was used as the standard for statistical significance.
3. Results
1) Expression of BDNF, ProBDNF, TrkB, Sortilin and p75NTR in HUVEC
Von Willebrand factor (vWF) is a glycoprotein synthesized by endothelial cells and is a commonly used marker for HUVECs. Immunofluorescence staining of vWF exhibited a cobblestone-like or spindle-shaped appearance in the HUVEC cytoplasm (Figure 1(a)). ProBDNF and mBDNF are mainly expressed in the cytoplasm as homogeneous red fluorescence staining. ProBDNF has stronger staining around the nuclear membrane compared with mBDNF (Figure 1(b) and Figure 1(c)). Sortilin and p75NTR are expressed in both the cytoplasm and nucleus, shown as green fluorescence staining (Figure 1(d) and Figure 1(e)). TrkB exhibits a cytoplasmic network shaped pattern in HUVEC cytoplasm, with stronger cell membrane staining (Figure 1(f)). No fluorescence is detected in the control cells. Western blot assays further confirm that HUVEC expresses mBDNF, proBDNF, TrkB, sortilin and p75NTR (Figure 1(g)). ProBDNF staining is colocalized with p75NTR or sortilin in HUVEC cytoplasm (Figure 2).
2) Effect of proBDNF on HUVEC Viability
To investigate the effect of proBDNF on HUVEC viability, MTT assay was performed by dividing HUVEC into treatment groups: control, BSA, proBDNF 1 ng/mL, proBDNF 3 ng/mL, proBDNF 10 ng/mL, proBDNF 30 ng/mL, proBDNF 100 ng/mL and proBDNF 100 ng/mL with anti-proBDNF (10 µg/mL). Treatment groups with proBDNF concentrations of 30 ng/mL and 100 ng/mL decreased HUVEC viability significantly (p < 0.05), compared with control and BSA groups. The suppressive effect of proBDNF on HUVEC can be counteracted by introducing anti-proBDNF into the culture medium (Figure 3).
Figure 1. Immunofluorescence staining of HUVECs. (a) von Willebrand factor. (b) ProBDNF, (c) mBDNF, (d) P75NTR, (e) Sortilin, (f) TrkB. (Bar = 40 µm). (g) Western blot of ProBDNF, p75NTR, Sortilin and TrkB in HUVEC.
Figure 2. Double staining of ProBDNF with p75NTR and Sortilin (Bar = 10 µm).
3) Effects of proBDNF on Angiogenesis in Vitro
HUVEC seeded onto Geltrex matrix coated plates developed well-formed tube networks after 14 - 16 hours incubation (Figure 4(a)). HUVEC that were incubated with proBDNF (50 ng/mL) also developed tube networks. However, the total mesh areas (p < 0.01) and the total branch lengths (p < 0.05) were significantly less than the control group (Figure 4(b) and Figure 4(c)).
Figure 3. Effect of proBDNF on HUVEC viability. ProBDNF concentrations of 30 ng/mL and 100 ng/mL inhibit HUVEC viability significantly, compared with control and BSA groups. (*p < 0.05). This inhibitive effect is counteracted by anti-proBDNF (10 µg/mL).
Figure 4. ProBDNF inhibitive effects on HUVEC angiogenesis in vitro. (a) HUVEC angiogenesis in the Control group and proBDNF (50 ng/mL) group. Left: phase-contrast images; middle: Analysing images using Angiogenesis Analyzer for ImageJ (tube structures are cyan, branches are green, segments are yellow, nodes are red, isolated structures are blue); right: fluorescent image with Calcein AM; (Bar = 100 µm). (b) Total branch lengths difference between control and proBDNF group (*p < 0.05). (c) Total mesh areas difference between control and proBDNF group (**p < 0.01).
4. Matrigel Plug Angiogenesis Assay
1) Tissue scores were significantly decreased by pre-treatment with proBDNF
H & E staining of Matrigel plug tissue revealed substantial vascular structures surrounding and infiltrating to the centre of the Matrigel in the mBDNF group and anti-proBDNF group (Figure 5(a)). There are several endothelial cells in the Matrigel of PBS group; however, there are few endothelial cells around the Matrigel and rare inside the centre of the Matrigel in the proBDNF group. Tissue scores in the proBDNF group were significantly lower than the PBS group (p < 0.05), anti-proBDNF group (p < 0.001), and mBDNF group (p < 0.001, not shown) (Figure 5(b)). Tissue score in the mBDNF group is higher than the anti-proBDNF group (*p < 0.05).
2) CD31 and α-SMA expression were decreased by pre-treatment with proBDNF
CD31 and α-SMA are the common markers for angiogenesis [30]. Immunohistochemical staining showed significantly lower expression of CD31 in the
Figure 5. (a) H & E staining of Matrigel plug tissue in the different groups (Bar = 50 µm). (b) Comparison of tissue scores in different groups. Tissue scores were significantly decreased in the proBDNF group, compared with the PBS group (*p < 0.05), anti-proBDNF group (***p < 0.001) and mBDNF group (***p < 0.001, not shown). Tissue score in the mBDNF group is higher than the anti-proBDNF group (*p < 0.05) (n = 6).
proBDNF group, compared with the PBS group (p < 0.05), anti-proBDNF group (p < 0.001), and mBDNF group (p < 0.001, not shown) (Figure 6(a) and Figure 6(c)). The anti-proBDNF group also showed considerably higher CD31 expression than the PBS group (p < 0.01). The CD31 expression in mBDNF group was more elevated than anti-proBDNF group (p < 0.01) (Figure 6(a) and Figure 6(c)). The expression of α-SMA was also significantly lower in the proBDNF group, compared with the PBS group (p < 0.05), anti-proBDNF group (p < 0.001), and mBDNF group (p < 0.001, not shown) (Figure 6(b) and Figure 6(d)). The expression of α-SMA in the anti-proBDNF group was also substantially higher than in the PBS group (p < 0.001). The mBDNF group demonstrated higher expression than the anti-proBDNF group (p < 0.01) (Figure 6(b) and Figure 6(d)).
Figure 6. (a) The expression of CD31 in different treatment groups (Bar = 50 µm). (b) The expression of α-SMA in different treatment groups. (c) Comparison of CD31 expression in Matrigel plugs. CD31 was significantly decreased in the proBDNF group, compared with the PBS group (*p < 0.05), anti-proBDNF group (***p < 0.001), and mBDNF group (***p < 0.001, not shown). Anti-proBDNF showed increased CD31 expression significantly, compared with the PBS group (**p < 0.01), and the mBDNF group has more CD31 expression than the anti-proBDNF group (**p < 0.01). (d) Comparison of α-SMA expression in Matrigel plugs. α-SMA was significantly decreased in the proBDNF group, compared with the PBS group (p < 0.05), anti-proBDNF group (p < 0.001), and mBDNF group (p < 0.001, not shown) (***p < 0.001). The expression of α-SMA in the anti-proBDNF group was also substantially greater than the PBS group (p < 0.001). The mBDNF group exhibited higher expression than the anti-proBDNF group (p < 0.01) (n = 6).
5. Discussion
BDNF is a widely known neurotrophic factor, which regulates neuronal survival and differentiation, and is involved in long-term potentiation and learning memory. ProBDNF, as the precursor of BDNF, is cleaved to yield mBDNF by either intracellular or extracellular enzymes. ProBDNF and mature BDNF often manifest opposing biological effects. Recent studies show that BDNF is not only expressed in the nervous system but also in a wide range of non-neural tissues and cells, for example, heart, ovaries, bone marrow, vascular smooth cells and endothelial cells [31] [32]. In the present study, we also found that proBDNF and its receptors Sortilin and p75NTR were expressed in HUVEC.
In addition to its neuropoietic effects, BDNF is unveiled to promote endothelial cell survival and induce angiogenesis. It is shown that BDNF plays a vital role in modulating vascular development. Compared to VEGF, which is crucial for the early stages of vascular development, BDNF has indispensable effects in perinatal and adult vasculature. BDNF is detectable in endothelial cells of the heart, and skeletal muscle at late gestation and the expression persists at elevated levels into adulthood [33]. Hempstead speculated that VEGF initiates the angiogenic process, but BDNF stabilizes and maintains the vasculature [34]. In the BDNF knockout mouse (BDNF−/−) model, it was shown that embryonic vasculature could be formed in the heart and be able to transform into arteries, capillaries and veins, but the endothelial cells cannot function through the late gestational to early postnatal stages, and vascular haemorrhages occur [33]. BDNF knockout impairs endothelial cell-cell contacts and causes endothelial cell apoptosis, which results in hypo-contractile heart and perinatal mortality. These results indicate BDNF plays a crucial role in regulating cardiac vascular endothelial cells and maintaining cardiac vessel stability in the gestational and postnatal periods [33]. BDNF has recently been described as a new mediator of angiogenesis [34]. BDNF significantly stimulates the migration of HUVEC in vitro and induces angiogenesis in tube formation assay and Matrigel plug in a mouse model [35]. BDNF induces angiogenin secretion and is involved in the nuclear translocation of angiogenin in HUVEC [36]. In a conditional genetic switching of the VEGF model, VEGF induction can lead to new vessel formation, but the vessels can be leaky and disorganized, causing severe tissue oedema [37]. In comparison, BDNF promotes an angiogenic response after cerebral ischemia without inducing fragile and malformed vessels, in contrast to VEGF [38] [39]. The angiogenic effects of BDNF on ischemia are not only by stimulating existing blood vessels to sprout but also by activating and organizing stem cells to the vasculature site [10].
Based on the known effect of BDNF as an angiogenesis mediator and the importance of balancing mBDNF and proBDNF, it is intriguing to know how proBDNF affects the process of angiogenesis. Our study shows that proBDNF inhibits HUVEC viability significantly, as revealed by the MTT assay. This inhibition can be counteracted by introducing anti-proBDNF. ProBDNF can also significantly suppress angiogenesis in vivo and in vitro. Endothelial cell tube formation assay exhibits total mesh areas and total branch lengths of tube networks that are substantially less in the proBDNF group than the control group. In Matrigel plug assay, the mean score for cell infiltration is significantly less in the proBDNF group, while anti-proBDNF pre-treatment increase cell infiltration scores. The immunostaining of CD31 and α-SMA (markers for angiogenesis) in Matrigel also confirms the inhibitive effect of proBDNF on angiogenesis. ProBDNF is likely via activating p75NTR/sortilin signals by an autocrine/paracrine mechanism as we showed that HUVEC expressed proBDNF colocalized with p75NTR/sortilin in these cells. These results indicate that proBDNF has an opposite effect to BDNF, acting as an angiogenesis inhibitor, providing more evidence for the balanced function of pro-neurotrophin and mature neurotrophin. Our results are consistent with other studies that show the inhibitive effect of pro-neurotrophins on vascular structure and function. Research has shown that proNGF induced endothelial cell death via the p75NTR in an oxygen-induced retinopathy mouse model, introducing potential therapeutic targets for the treatment of proliferative retinopathies [40]. In a hypoxia/reoxygenation model, proBDNF mediated myocardial microvascular endothelial cell (MMEC) injury by reducing MMEC migration and decreasing the formation of capillary-like structures [41]. ProBDNF and proNGF are highly expressed in the infarcted rat heart, triggering damage to the pericytes of blood vessels, and p75ECD-Fc, the scavenger of proneurotrophin can protect the structure and function of the ischemic heart [42]. P75NTR, as the preferred receptor of proneurotrophin, can promote endothelial cell apoptosis and inhibit angiogenesis. P75NTR impairs postischemic neovascularization and blood flow recovery response to limb ischemia in diabetes. It suggests that inhibiting p75NTR could be a potential therapy for treating diabetes-induced microvascular diseases. [43]. Silencing p75NTR can also prevent proNGF-induced endothelial cell death and the development of acellular capillaries in rat retina, and deletion of p75NTR can protect against retinal ischemia and prevent retinal neovascularization in the ischemic retinopathy model [44] [45]. These studies support that proneurotrophins have opposite effects on endothelial cell survival and angiogenesis compared with mature neurotrophins.
Understanding the opposite functions of BDNF and proBDNF on angiogenesis is essential to broaden the insight for treating diseases like cerebral ischemia and cancer. Current research shows that neuro-restoration, including angiogenesis, neurogenesis and synaptic plasticity, are likely to have a far greater window of time for intervention than acute neuroprotective treatments [46]. The advantage of BDNF as a selective angiogenic factor is minimal secondary side effects such as the induction of vascular permeability, which could lead to the development of novel and specific angiogenic therapies for cerebral ischemia. On the other hand, angiogenesis inhibitors pose a promising therapeutic strategy for tumours [47]. Research shows the efficacy of anti-angiogenic monotherapies is compromised due to the redundancy of angiogenic signals. When treating with a single pro-angiogenic molecule, the tumours tend to evade the pathway and become refractory by upregulating alternative angiogenic factors as compensation. Therefore, more and more research focuses on combining different anti-angiogenic elements and finding accurate time windows for anti-angiogenic action on tumours [48]. Our study provides a new anti-angiogenic factor, proBDNF, which could contribute to rational anti-angiogenic drug combination therapy and personalized molecular therapy for tumour or other proliferative vascular diseases.
6. Conclusion
In contrast to mature BDNF, proBDNF acts as an inhibitor of angiogenesis as revealed by the reduction of survival in endothelial cells, and a decrease in blood vessel formation in vitro and in vivo. Therefore, suppression of endogenous proBDNF in endothelial cells can promote the survival of endothelial cells and angiogenesis. This finding could provide more therapeutic strategies for diseases, such as cerebral ischemia, tumours, and other vascular diseases.
Acknowledgements
This work was supported by the National Health and Medical Research Council, Australia (NHMRC fellowship to Xin-Fu Zhou) and the University of South Australia, Australia.
Authors’ Contributions
XFZ supervised and coordinated all aspects of the work. HL performed the experiments, analysed data, prepared figures, and wrote the paper. FFB assisted with animal experiments. AB helped with immunostaining and provided technical support. LB provided material support and revised the manuscript. All authors have read and approved the final manuscript for submission. The authors declare no competing financial interests.