| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Physiology and Biophysics, Georgetown University Medical Center, Washington, DC, USA;
Department of Biochemistry, Virginia Commonwealth University School of Medicine, Richmond, Virginia, USA; and
National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, Maryland, USA
1Correspondence: Department of Physiology and Biophysics, Georgetown University Medical Center, Washington, DC, 20057, USA, E-mail: zzukow01{at}georgetown.edu
ABSTRACT
Previously we discovered that NPY induces ischemic angiogenesis by activating Y2 and Y5 receptors. The receptors that mediate specific steps of the complex process of angiogenesis are unknown. Here, we studied in vitro NPY receptors subtypes involved in migration, proliferation, and differentiation of human endothelial cells. In cells that expressed Y1, Y2, and Y5 receptors, NPY bimodally stimulated migration and proliferation with a 2-fold increase at 10–12 M and 10–8 M (high- and low-affinity peaks, respectively). Preincubation of cells with NPY up-regulated the Y5 receptor and markedly enhanced endothelial cell migration and proliferation. NPY-induced endothelial cell migration was mimicked by agonists and fully blocked by antagonists for any specific NPY receptors (Y1, Y2, or Y5), while proliferation was blocked by any two antagonists (Y1+Y2, Y1+Y5, or Y2+Y5), and capillary tube formation on Matrigel was blocked by all three (Y1+Y2+Y5). Thus, NPY-induced angiogenesis requires participation of Y1, Y2, and Y5 receptor subtypes, with the Y5 receptor acting as an enhancer. We propose that these receptors form heteromeric complexes, and the Y1/Y2/Y5 receptor oligomer may be the uncloned Y3 receptor.—Movafagh, S., Hobson, J. P., Spiegel, S., Kleinman, H. K., Zukowska, Z. Neuropeptide Y (NPY) induces migration, proliferation, and tube formation of endothelial cells bimodally via Y1, Y2, and Y5 receptors.
Key Words: bimodal dose-response curve • chemotaxis • mitogenesis • oligomerization • GPCRs
ANGIOGENESIS IS THE highly regulated, multistep process of new blood vessel formation from preexisting vessels (1
, 2)
. Physiologically, angiogenesis occurs during embryogenesis, organ growth, reproduction, and wound healing (3)
, whereas pathological angiogenesis has been associated with such human diseases as cancer (4)
and retinopathies (5)
. Angiogenesis begins with activation of endothelial cells, followed by their migration, proliferation, and differentiation into capillary tubes (6)
. Vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) are the most potent angiogenic factors which stimulate migration, proliferation, and differentiation of endothelial cells (7
8
9)
. Neuropeptide Y (NPY) is a sympathetic neurotransmitter which our laboratory has discovered to be angiogenic with equal potency and efficacy as bFGF and VEGF (10
11
12)
.
NPY is a 36 amino acid neuropeptide present in abundance in the brain, the adrenal medulla (13)
, sympathetic nerves (14)
, and also extraneuronally in endothelial cells (12)
. This peptide belongs to a family of related peptides, such as peptide YY (PYY) and pancreatic polypeptide (PP), which have many but not all of NPY’s activities. We have shown that NPY stimulates multiple steps of angiogenesis, including endothelial cell attachment, migration, proliferation, and differentiation (12)
. In the Boyden chamber migration assay, we found that NPY induced a biphasic concentration-dependent migratory response in HUVECs (12)
.
There are five functional human NPY receptors, named Y1-Y5. Among these, Y3 is the only receptor not yet cloned, but it has been characterized pharmacologically as the "NPY-preferring" receptor; i.e., not activated by PYY or PP (15)
. Pharmacological and receptor knockout studies in mice have suggested that most of NPY’s actions in the brain and in the cardiovascular system are mediated via the Y1, Y2, and Y5 receptors (15)
. In the cardiovascular system, the Y1 receptor is involved in vasoconstriction (15
, 16)
and promotion of smooth muscle cell growth (17)
. The Y2 receptor, on the other hand, appears to be responsible for the angiogenic effects of NPY (11
, 18
19
20)
. Ghersi et al. (20)
have shown that a Y2/Y3/Y5 receptor agonist (NPY3–36) stimulates endothelial cell migration similar to NPY (NPY1–36). In the same study, the Y1/Y3/Y5 receptor agonist, [Leu31-Pro34]-NPY did not reproduce the effect. Our recent study using the Y2 receptor knockout mice showed that the knockouts have significantly diminished NPY-induced angiogenesis, both in the aortic sprouting and in the ischemic limb model compared with wild-type (WT) mice (11
, 18)
. In addition, the Y2 receptor knockout mice showed significantly less hyperoxia-induced retinopathy than the WT (21)
. Taken together, these studies suggest a major role for the Y2 receptor in NPY-mediated angiogenesis.
However, the role of the Y5 receptor in angiogenesis is yet to be defined. This receptor shares some roles with the Y1 and the Y2 receptors in the brain, such as in the stimulation of food intake and in epilepsy (22
23
24)
. The Y5 receptor gene is also colocalized with the Y1 and the Y2 receptor genes on the human chromosome 4q31–32 (15)
, and shares the same promoter with the Y1 gene (25)
. Moreover, Y1 and Y5 receptor antagonists similarly inhibit proliferation of vascular smooth muscle cells in vitro, as well as in vivo by preventing angioplasty-induced neointima formation in the rat carotid artery (26)
. Furthermore, in the rat aortic sprouting assay, Y2 and Y5 receptor antagonists equally inhibit sprout formation (10)
. An interesting feature of the Y5 receptor noted in these studies is that expression of this receptor is not detectable in normal vessels prior to injury. However, conditions such as ischemia or vascular injury appear to up-regulate its levels (26)
.
In this study, we demonstrate that NPY stimulates in vitro migration, proliferation, and differentiation of human endothelial cells by activating multiple NPY receptors that may act either as hetero-oligomeric complexes or independently of each other. In addition, we characterize the NPY receptor subtype(s) involved in these processes utilizing a wide range of NPY concentration (10–14 to 10–7M), receptor agonists and antagonists, and RT-polymerase chain reaction (RT-PCR). Knowledge of NPY receptor subtypes mediating its angiogenic effect may aid in the design of more effective NPY receptor agonists and antagonists.
MATERIALS AND METHODS
Cell culture
HUVECs were isolated from freshly delivered umbilical cords after incubation at 37°C for 20 min with collagenase type I enzyme solution in sterile deionized water and plated on gelatin-coated T75 flasks. The HUVEC media consisted of medium 199 (Sigma) supplemented with 20% FBS (Hyclone, Logan, UT, USA), 1000 U/L penicillin/streptomycin, 5 mg/L gentamicin, 2 mmol/L glutamine, 500 U/L sodium heparin, 2.5 mg/L amphotericin B (Gibco BRL, Grand Island, NY, USA), and 1 mg/L Endothelial Cell Growth Supplement (ECGS, Collaborative Biomedical Products, Becton Dickenson Labware, Bedford, MA, USA). Experiments were performed with cells between passages 3 and 7. Heparin was used to inhibit attachment of fibroblasts or smooth muscle cells, and thus, preserve the homogeneity of the endothelial cell cultures. The identity of endothelial cells was confirmed by positive immunostaining for von Willbrand factor and by negative staining for smooth muscle 
-actin.
SVEC4–10 is an endothelial cell line derived from human dermal microvascular endothelial cells that have been transformed via SV-40 large T antigen (27)
. SVEC4–10 cells at passage 8 were provided by Dr. M. Edidin (Johns Hopkins University). The cells were grown in Dulbecco’s modified Eagle medium (DMEM) with 4.5 g/l glucose (Glc) and without glutamine, supplemented with 10% FBS, 100 U/ml penicillin, 100 µg/ml streptomycin, and 2 mmol/L glutamine in a 5% CO2 humidified incubator. Cells up to passage 12 were used for the assays.
In vitro migration assay
HUVECs or SVEC4–10s in their respective serum-free media were added to the upper wells of 48-multiwell chemotaxis Boyden microchambers at 30–40,000 cells/well; the lower wells contained NPY with or without the antagonists, or NPY analogues diluted in serum-free media (medium 199, or DMEM). Each condition was assayed in triplicates and each assay was done at least two times. When indicated, cells were preincubated with the antagonists (diluted to 10–7 M concentration in serum-free media) for 15 min prior to loading into the upper wells. After 2 h at 37°C in 5% CO2, the membranes were fixed and stained, and the number of cells that migrated through to the lower surface of each membrane was counted. A negative control consisted of serum-free media (DMEM) for SVEC4–10s, and medium 199 for HUVECs), and a positive control consisted of HUVEC and SVEC4–10s growth media. The growth media were used as positive controls since they reproducibly stimulated endothelial cell migration. They contained growth factors, such as bFGF and EFGF, as part of serum and growth supplements (ECGS; see above).
Proliferation assay
DNA synthesis
SVEC4–10 cells were plated onto Nunc 96-well dishes (6,000 cells/well) with 100 µl/well of their standard growth medium (DMEM plus 10% serum). Once they reached 75% confluency, their growth medium was removed from each well and cells were washed once with PBS containing Ca2+ and Mg2+. Then the cells were growth-arrested in 0.5% serum media for 24 h. If pretreatment was desired, NPY was added to the media at that time. For the next 24 h the cells were treated with NPY, NPY agonists, or NPY plus antagonists diluted in the same media. The controls consisted of 0.5% serum- (negative control) and 10% serum-containing media (positive control). 0.5 µCi [3H]-thymidine/well was added 6 h before completion of the experiment, diluted in the same 0.5% serum containing media. At the end of the experiment, media were removed and cells were washed with PBS (100 µl/well). Then 50 µl of 0.05% trypsin was added to each well, incubated for 5 min, followed by the addition of 100 µl/well of distilled water to lyse the cells. The plates were then either frozen in –20°C until the time of quantification or quantified by harvesting the samples into filter mats (Wallac, Turku, Finland) using a Tomtec Harvester 96 Mach II (Orange, CT, USA). [3H]-thymidine uptake (cpm/well) was measured using a ß-plate liquid scintillation counter.
In vitro capillary tube formation on Matrigel
Cells were incubated for 18 h at 37°C on Matrigel-coated 24-well plates, at 40,000 cells/well in the 2.5% serum containing media (DMEM or medium 199) with NPY (in the absence or presence of either receptor antagonist or antagonists), NPY analogues, or antagonists alone. For antagonist treatment, cells were preincubated with the antagonist 15 min prior to loading onto Matrigel. Cells were fixed and stained with DiffQuick Fixative and Solution II and the area of the tube network was quantified at X4 magnification with a Nikon microscope using NIH image (28)
. Each data point was assayed in triplicate and the experiments were repeated two times.
RT-PCR of human NPY, and Y1, Y2, and Y5 receptor mRNAs in endothelial cells
RNA was isolated from subconfluent cells using TRI REAGENT (Molecular Research Center, Inc., Cincinnati, OH, USA) to determine NPY or NPY receptor expression by RT-PCR. The RNA was first digested with DNase I (Ambion, Austin, TX, USA) and relative, semiquantitative RT-PCR was performed using 18S rRNA as an internal control. 7.5 µg of total RNA was reverse transcribed by denaturing at 65°C for 5 min and then incubating for 50 min at 42°C in 30 µl of 1 x reaction buffer containing dNTPs (0.5 mM each), 5 µM random hexamers (Perkin Elmer, Foster City, CA, USA), 2 U/µl of placental RNase inhibitor (Promega, Madison, WI, USA), 10 mM DTT and 1.5 µl of Super Script II Reverse Transcriptase (GIBCO Life Technologies, Rockville, MD, USA). Reactions were heated at 70°C for 15 min to inactivate reverse transcriptase. 2.5 µl of cDNA was amplified in 25 µl of 1 x reaction buffer containing 1.5 mM MgCl2, 0.2 mM each dNTP, 0.8 µM each primer, and 1.5 U of Taq polymerase (Promega). Reactions were carried out as follows: 1) 94°C, 1 min; 60°C, 1 min 20 s; 72°C, 1 min; 35 cycles (murine Y1, Y2, Y5 receptors); 2) 94°C, 1 min; 52°C, 1 min 20 s; 72°C, 1 min 15 s; 35 cycles (human Y1, Y2, Y5 receptors, murine NPY, DPPIV). The following primers were used in RT-PCR reactions: Y1 receptor: 5'-CTCTTGCTTATGGRGATGTGA-3'; 5'-CTGGAAGTTTTTGTTCAGGAAYCCA-3'; Y2 receptor: 5'-CCTACTGCTCCATCATCTTGC-3'; 5'-GTAGTTGCTGTTCATCCA GCC-3'; Y5 receptor: 5'-ATGGAGTTTAAGCTTGAGGAGC-3'; 5'-TGTGTAGGCA GTGGATAAGGG-3'; 18s RNA: 5'-TCAAGAACGAAAGTCGGAGG-3'; 5'-GGACA TCTAAGGGCATCACA-3'; NPY: 5'-TACCCCTCCAAGCCGGACAA-3'; 5'-TCTCATTT CCCATCACCACATG-3'. polymerase chain reaction (PCR) products were analyzed by electrophoresis in 2.5% agarose gels and visualized by ethidium bromide staining.
Materials
Human NPY, NPY3–36, [Leu31-Pro34]-NPY, hPP, PYY, and des-amidated NPY were purchased from Peninsula Labs (Belmont, CA, USA). Matrigel was purchased from Becton Dickenson (Bedford, MA, USA). Y1 receptor antagonist (H409/22) and Y5 antagonist (CGP71683A) were gifts from AstraZeneca and Y2 antagonist (BIIE0246TF) was from Berhinger Ingelheim. [Ala31,Aib32]NPY and [Ahx 5–24]NPY were gifts from Dr. Beck-Sickinger (University of Leipzig, Germany).
Statistical analysis of data
Raw data were analyzed with 1- or 2-way ANOVA, followed by post hoc t tests: Dunnett’s t test when compared with the control and Bonferroni’s t test when compared pairwise using SigmaStat 2.03 software (Statistical Packages for the Social Sciences Science, Chicago, IL, USA), and considered significant at P < 0.05. Data were presented as mean ± SE of either raw data or ratios of raw data over control.
RESULTS
NPY bimodally stimulates migration of HUVECs and SVEC4–10s in vitro
SVEC4–10s’ chemotaxis to NPY was tested in parallel with that of HUVECs and the representative dose-response curves and the NPY receptor expression are shown in (Fig. 1
). NPY enhanced HUVECs migration bimodally by 2-fold (Fig. 1A
). The first peak occurred at pM concentrations with a 2.2-fold increase (84.3±3.1 cell/mm2) over the baseline (38.3±6.0 cell/mm2) (P<0.001) at 10–12M. The second peak occurred at 10–8 M with a 1.9-fold increase (72.7±1.5 cell/mm2) over the baseline (P<0.001). RT-PCR data of HUVECs prior to the experiment showed expression of Y1, Y2, and Y5 receptor subtypes and the NPY peptide (Fig. 1A
, insert). SVEC4–10s’ chemotaxis to NPY varied with response seen only to higher concentrations of the peptide (10–8 and 10–7 M) (Fig. 1B
). Only the low-affinity peak with a 2.2-fold increase (425±21.6 cell/mm2) over the baseline (195.3±10.7) (P<0.001) occurred at 10–8 M. RT-PCR of these cells prior to the experiment revealed an absence of the Y5 receptor expression (Fig. 1B
, insert). However, the bimodality of SVEC4–10s’ chemotaxis to NPY was restored after a 24 h preincubation in low serum (0.5%) with NPY (10–8 M) (Fig. 1C
). In this case, 10–12 M NPY increased SVEC4–10s’ chemotaxis by 1.8-fold (112±5.0 cell/mm2) over the baseline (62±6.4) (P<0.001). Likewise, at 10–8 M, a 2-fold increase (123±4.7 cell/mm2) over the baseline (P<0.001) was observed. The chemotaxis "valley" occurred at 10–10M for both cell types, with a significant difference (P<0.001) between migration at this concentration and at the two peaks. RT-PCR of SVEC4–10 cells (after a 24 h incubation but prior to the migration assay) revealed expression of all three receptor subtypes Y1, Y2, and Y5 and the peptide (Fig. 1C
, insert). NPY administration did not alter the expression of any of its receptors within the 2 h of migration (data not shown).
|
NPY-stimulated migration of HUVECs and SVEC4–10s involve Y1, Y2, and Y5 receptors
To elucidate which NPY receptor(s) mediate HUVEC and SVEC4–10 cells’ migration to NPY, the NPY effect at both peaks of the migratory response (10–12M and 10–8M) was compared with those of various NPY receptor agonists and antagonists. At the concentration of 10–12 M, only NPY3–36 (Y2/Y3/Y5 agonist) (1.9-fold) was as effective as NPY1–36 (2.0-fold) (P<0.001 compared with baseline). Responses to both were partially inhibited by the addition of the Y1 receptor antagonist (H409/22)(from 2.0-fold over the baseline to 1.5-fold). Complete inhibition was observed by the addition of Y2 (BIIE0246TF)(to 0.9-fold), Y5 (CGP71683A)(to 1.0-fold), and a combination of Y1+Y2 (to 0.7-fold) or Y2+Y5 (to 1.3-fold) receptor antagonists (P<0.001 compared with NPY) (Fig. 1D
). On the other hand, NPY’s action at 10–8 M (2.0-fold) was mimicked by human PP (Y5 agonist) (2.0-fold) (P<0.001 compared with baseline), and similarly blocked by the addition of each of the Y1, Y2, and Y5 antagonists, as well as by a combination of these antagonists (P<0.001 compared with NPY) (Fig. 1E
). Similar results were found with SVEC4–10 cells (data not shown). hPYY (Y1/Y2/Y4/Y5, but not Y3 agonist), deamidated NPY (NPY without the PP fold, not binding to known NPY receptors and not bioactive), [Ala31,Aib32]NPY (specific Y5 agonist), and [Ahx 5–24]NPY (specific Y2 agonist) did not influence the migratory response (data not shown).
NPY bimodally stimulates proliferation of SVEC4–10 cells in vitro
NPY’s growth promoting effect on SVEC4–10s was also tested by measuring DNA synthesis ([3H]thymidine incorporation). After being growth arrested in 0.5% serum for 24 h, cells were exposed to NPY at a concentration range of 10–14 to 10–7M for another 24 h. As shown in Fig. 2
A, NPY increased growth of SVEC4–10s in a bimodal fashion. The first peak was around 10–12-10–11M NPY with a [3H]thymidine incorporation of 203–206% (n=30, P<0.001) over the baseline (100%). The second peak was around 10–8M with a [3H]thymidine incorporation of 186% (n=30, P<0.001). The valley was at 10–10M (169%), which was significantly different from the first peak (P<0.05). The minimum effective NPY concentration was 10–14M, with a 60% increase in mitogenesis (n=30, P<0.001). The growth response to the positive control (SVEC4–10 media) was 329% (n=30, P<0.001). Moreover, semiquantitative RT-PCR was used to assess the NPY receptor expression. As shown in Fig. 2
(insert), these cells constitutively expressed Y1 and Y2 receptors and DPPIV. However, in response to NPY, expression of the Y5 receptor and the NPY peptide showed a bimodal response, detected mainly at NPY concentrations corresponding with the two peaks of the mitogenic response.
|
NPY up-regulates Y5 receptor synthesis and NPY-mediated proliferation
SVEC4–10s were growth arrested in 0.5% serum media for 24 h. During this period, one-third of endothelial cells received 10–12M NPY, one-third 10–8M NPY, and the other one-third none (which served as the control). In the next 24 h, all cells were exposed to NPY at a concentration range of 10–14 to 10–7M. The negative control (0) was DMEM and the positive control was SVEC4–10 media (10% FBS) (not shown). DNA synthesis measurement and RT-PCR were done at the end of the preincubation period (24 h), and at the end of the experiment (48 h).
SVEC4–10 cells exposed to NPY during the preincubation period had a comparable growth to nonpretreated cells as measured after the preincubation period (initial 24 h, Fig. 2B
). However, in the second 24 h, NPY pretreated cells had a higher growth response to NPY. For example, in response to 10–12M NPY, NPY 10–12 M-pretreated cells responded with a [3H]thymidine uptake of 539% (5.3-fold over the baseline) (n=12, P<0.001), which differed significantly from the nonpretreated cells, with a [3H]thymidine uptake of 253% (2.5-fold over the baseline) (n=12, P<0.001). NPY 10–8M pretreated cells had a similar increase in response with a 502% [3H]thymidine uptake (n=12, P<0.001).
On the other hand, RT-PCR data revealed that prior to the initiation of the experiment, SVEC4–10s growing in their full serum (fibrous sheath) media expressed Y1 and Y2 receptors, NPY, and DPPIV (Fig. 2B
, insert: 0 h, fibrous sheath (FS)). After the preincubation period (24 h), the Y5 receptor expression was detected in both the non-NPY-pretreated (Fig. 2B
, insert: 24 h, a) and the NPY-pretreated cells (Fig. 2B
, insert: 24 h, b); although, higher levels were seen in the NPY-pretreated cells. Furthermore, cells without any NPY treatment for 48 h lost the Y5 receptor and the NPY peptide expression (Fig. 2B
, insert: 48 h, c). NPY-treated cells, on the other hand, expressed all three receptors plus the NPY peptide and DPPIV (Fig. 2B
, insert: 48 h, d). Higher expression of the Y5 receptor was detected in the NPY-pretreated cells vs. the nonpretreated cells (Fig. 2B
, insert: 48 h, e).
NPY-induced proliferation of SVEC4–10 cells involves Y1, Y2, and Y5 receptors
To elucidate which NPY receptor(s) mediate the mitogenic response of SVEC4–10s to NPY, we tested multiple NPY receptor agonists and antagonists in the [3H]thymidine uptake assay alongside NPY1–36. At first, dose-response curves of NPY receptor agonists [Leu31Pro34]-NPY (Y1/Y3/Y5 agonist) and NPY3–36 (Y2/Y3/Y5 agonist) were compared with NPY1–36 (Y1/Y2/Y3/Y5 agonist). Both agonists closely mimicked both the high-affinity (pM range) and the low-affinity peaks (nM range) of NPY1–36 (Fig. 2C
). Next we tested NPY’s response in the presence of Y1, Y2, and Y5 receptor antagonists. Our data show that the NPY1–36 dose-response curve was not affected by single Y1, Y2, or Y5 receptor antagonists (Fig. 2D, a-c
). Conversely, the combination of any two receptor antagonists, Y1+Y2, Y1+Y5, or Y2+Y5 completely inhibited NPY-induced mitogenesis (Fig. 2E
: a–c). All three receptor antagonists together also fully blocked the response (not shown). Moreover, none of the NPY receptor antagonists had endogenous mitogenic effects, or affected the mitogenic response of serum (data not shown). Furthermore, the expression of NPY, DPPIV, or NPY receptor subtypes did not change with antagonists’ treatment (not shown).
NPY bimodally induces capillary tube formation in Matrigel of HUVECs and SVEC4–10 cells, and Y1, Y2, and Y5 receptors are involved
Capillary tube formation in Matrigel requires migration as well as differentiation of endothelial cells into networks. In this assay, NPY, at all four concentrations tested (10–14, 10–12, 10–10, 10–8) significantly increased the density of the capillary network compared with baseline (P<0.001). However, a bimodal pattern could still be observed as there was a significant difference in the capillary tube formation between concentrations of 10–14M and 10–12M (P<0.05), 10–12M and 10–10M (P<0.001), and 10–10M and 10–8M (P<0.05). The minimum increase was 1.7-fold (10–14M and 10–8M, P<0.001) and the highest increase was 2.1-fold (10–12 M, P<0.001), comparable to those of bFGF (2.1-fold, P<0.001) and VEGF (2.1-fold, P<0.001) at the same concentration (10–12M) (Fig. 3
A, B). NPY’s effect at the lowest effective concentration was compared with those of agonists and antagonists. Similar to their effects on migration, Y2/Y5 agonist (1.6-fold, P<0.001), and to a lesser degree, Y1/Y5 agonist (1.3-fold, P<0.001) mimicked the stimulatory response of NPY1–36 (1.7-fold, P<0.001) (Fig. 3C
). However, unlike in the Boyden chamber assay, inhibition of this response was observed when all three antagonists were used together. A combination of Y1, Y2 and Y5 antagonists significantly reduced NPY-stimulated tube formation by 66% (from 1.7-fold to 0.6-fold, P<0.001), to even below the baseline (1.0), but did not affect the positive controls (HUVEC media, data not shown).
|
DISCUSSION
The angiogenesis cascade is multistep and complex because of the large number of participating proteins that function to promote cell migration, proliferation, and differentiation. VEGF and bFGF are two ubiquitous and potent angiogenic peptides that participate (or are even essential) in many angiogenic cascades. These angiogenic molecules are secreted by the mesenchymal and many tumor cells and are stored in the extracellular matrix (ECM). NPY, by being present in the brain, the endothelium, and the sympathetic nerves, which accompany all blood vessels except the aorta and are tonically active at angiogenic sites, is a potential participant of many angiogenic cascades. We have discovered previously that NPY is an angiogenic factor and exerts strong stimulatory effects on endothelial cell migration, proliferation, and differentiation into capillary tubes in vitro. NPY also significantly enhances angiogenesis in various ex vivo and in vivo models of angiogenesis, such as aortic sprouting, Matrigel plug assay, and ischemic hind limb revascularization. In vitro and in vivo studies revealed that NPY’s angiogenic potential is comparable to that of bFGF and VEGF.
In this study we demonstrate that NPY’s angiogenic activity in vitro is bimodal. A high-affinity (pM) and a low-affinity (>10 nM) peak, with an average of a 2-fold increase occurred in migration, proliferation, and differentiation of both endothelial cell types (HUVEC and SVEC4–10), when, all three Y1, Y2, and Y5 receptors were expressed (RT-polymerase chain reaction). The presence of two maxima, one at very low NPY concentrations (pM) and one at very high concentrations (10 nM) not only indicates that NPY is one of the most potent angiogenic peptides known, but also that it can induce blood vessel formation over a wide range of concentrations. For instance, NPY-induced angiogenesis can occur in tissues that contain low NPY concentrations, such as the nonsympathetically innervated aorta or growing organs lacking a mature NPY system (29)
. Nevertheless, it can occur in tissues containing high NPY concentrations, such as the heart, mature vessels, or muscles during high sympathetic activity induced by stress, ischemia, or injury (11
, 18
, 30
, 31)
. NPY also proved to be a potent growth factor for endothelial cells: 10–14M of this peptide significantly increased proliferation of SVEC4–10s by up to two-fold. The growth-promoting action of NPY in such a low concentration implies that NPY may be a physiological growth factor for endothelial cells. Whether diffused from local nerve terminals or synthesized and released from endothelial cells, these cells are exposed to this peptide at low levels during normal hemeostatic conditions (11
, 18)
. In fact, it has been shown that neurons in the central nervous system (CNS) secrete pM (10–12M) concentrations of NPY in basal conditions with a higher secretion due to the occurrence of stress (32)
. Therefore, NPY may act alone, or synergize with other growth factors such as VEGF (33)
and bFGF (34
, 35)
to promote both survival of endothelial cells and angiogenesis.
Our agonist/antagonist data shows that both the high- and the low-affinity peaks of NPY activity are mediated via activation of multiple NPY receptor subtypes: Y1, Y2, and Y5. Migration requires activation of all of these receptors. The high-affinity (10–12 M) peak of migratory response of NPY1–36 was mimicked by NPY3–36 (Y2/Y3/Y5 agonist), while no single receptor agonist, such as [Ala31, Aib32]-NPY (specific Y5 agonist) and [Ahx 5–24]-NPY (specific Y2 agonist), was able to produce a migratory response. On the other hand, this response was significantly inhibited by any of Y1, Y2, or Y5 single antagonists. The low-affinity (10–8M) migratory peak of NPY1–36 was also fully inhibited by either of these receptor antagonists. These results suggest a combined Y1/Y2/Y5' receptor activity for both migratory peaks. This combined receptor had a similar pharmacology to the putative Y3 receptor (NPY-preferring receptor), since, among the tested peptides, PYY1–36 (Y1/Y2/Y5, but not Y3 receptor agonist) was the only one that did not produce the two migratory peaks. PYY1–36 and PYY3–36 (truncated by the DPPIV enzyme) are analogous in activity (except at the Y3 receptor) to NPY1–36 and NPY3–36, respectively. However, their tertiary structure varies to some degree with that of NPY (PubMed’s structure database). To date, no single NPY receptor has been cloned with the pharmacology of the Y3 receptor. However, despite its debatable presence, several tissues with Y3 receptor activity have been identified, including but not limited to human adrenal chromaffin cells (36)
, nucleus tractus salitarii (NTS) (37)
, cardiac ventricular membrane (38)
, and rat aortic endothelial cells (39)
. In adrenal chromaffin cells, the sum of the individual effects of Y1, Y2, Y4, and Y5 receptors has been suggested as the Y3 receptor effect (15)
. In these reports, inactivity of PYY was used as the criteria to infer Y3 receptor activity.
NPY-induced proliferation of endothelial cells, on the other hand, necessitates simultaneous activation of two NPY receptor subtypes (Y1+Y2, Y1+Y5, or Y2+Y5). This is supported by the observation that minimum of two receptors have to be antagonized to block the proliferation but no further inhibition occurs with antagonizing all three. NPY-induced proliferation of SVEC4–10s in PC-12 (pheochromocytoma cell line)-conditioned media (rich in NPY secondary to release from PC12 cells) is also inhibited only with a combination of two receptor antagonists (40)
. Nevertheless, NPY-induced endothelial cell differentiation appears to require only one active receptor subtype (Y1, Y2, or Y5), since inhibition occurs only by antagonizing all three. Therefore, the NPY receptors Y1, Y2, and Y5 are all capable of signaling migration, proliferation, and differentiation of endothelial cells to NPY.
Our RT-PCR data also suggest a multireceptor mode of action for NPY in endothelial cells as well as an important role for the Y5 receptor. HUVECs growing in HUVEC media constitutively expressed Y1, Y2, and Y5 receptors in addition to the NPY peptide and DPPIV. On the other hand, SVEC4–10s growing in their full media differed from HUVECs in that they did not express the Y5 receptor constitutively. However, their Y5 receptor expression became detectable after a 24 h incubation in low serum or NPY, and it correlated directly with their migration toward the pM concentrations of NPY. In the proliferation study, NPY given to SVEC4–10s after 24 h incubation in low serum (the second 24 h) dose-dependently increased its own expression and the expression of the Y5 receptor, with higher levels at the peaks of the proliferative response (pM and 10 nM concentrations). Moreover, additional NPY given during the initial 24 h incubation in low serum further increased SVEC4–10s Y5 receptor level, and their proliferation. Similar regulation of NPY receptors by growth factor starvation and NPY itself has been observed in vitro in rat and human vascular smooth muscle cells (VSMCs) (41)
. In vivo, during ischemia or vascular injury, particularly when superimposed on NPY local administration, the expression of all three NPY receptor subtypes (Y1, Y2, and Y5) is increased (11
, 42)
. NPY receptor plasticity has also been seen in other tissue types, such as the brain, when all three receptors are up-regulated or induced following induction of seizures (43)
.
Bimodal dose-response curves with high- and low-affinity peaks have been reported in the scientific literature since 1973 (44
45
46
47
48
49
50
51)
. Some in vitro examples include, norepinephrine-stimulated lipolysis in rat adipocytes (44)
, concanavalin A (Con A) induced proliferation of Human Peripheral Blood Lymphocytes (PBL) (45)
, leukotrienes C4 stimulated LH-releasing hormone (LH-RH) secretion from rat median eminence (46)
, and human growth hormone (hGH) induced production of Insulin-like growth factor (IGF) I (insulin-like growth factor-I) by human lymphoid cell line IM-9 (51)
. Example hypotheses provided for the observed high- and low-affinity peaks include presence of different cell populations with multiple affinity receptors (45)
, bimodal concentration of intracellular messengers (44)
, water-mediated ligand-receptor transmission in super low (10–17-10–15 M) ligand concentrations (47)
, and presence of multiple binding sites on one receptor (49)
.
We propose that NPY’s bimodal angiogenic effect on the endothelium in vitro is through oligomerization of NPY receptor subtypes Y1, Y2, and Y5; heterotrimers for migration, heterodimers for proliferation, and monomers for differentiation. Moreover, we speculate that NPY receptor oligomers are part of larger receptor complexes that include other receptors (i.e., integrins), enzymes (i.e., DPPIV and eNOS), and receptor tyrosine kinases (i.e., EGFR or VEGF). Accordingly, the peaks of NPY activity may correspond to the oligomeric state of the NPY receptors with enhanced intracellular signaling, whereas the valleys may correspond to either desensitization/dissociation of the oligomer or a transitional state. The high- and the low-affinity peaks may differ in a number of ways: in the presence or absence of regulatory/accessory proteins or enzymes, or as to how the oligomers are assembled (directly via contact, domain swapping, or adaptor or scaffold proteins, or indirectly through intracellular messengers). Furthermore, the Y5 receptor plays a key role in the NPY receptor complex in its formation and stabilization, its ligand binding affinity, and its signaling capacity. Alternatively or additionally, coupling of Y receptors may occur downstream secondary to converging signaling pathways.
NPY receptor subtype dimerization has been observed via fluorescent microscopy in our laboratory in Chinese hamster ovary (CHO) cells cotransfected with the Y1 and the Y5 receptors (J. Pons et al., unpublished data) as well as by others in cotransfected baby hamster kidney (BHK) cells (BHK) (52
, 53)
, and in human embryonic kidney cells (HEK 293) (53)
. In the latter, although unaltered by NPY or PYY, Y1/Y5 heterodimers seemed to have existed constitutively and their levels decreased in the presence of Y1 receptor antagonists and increased with specific Y5 agonists or Y2/Y5 agonist, NPY2–36. Y1/Y5 cotransfected cells also had enhanced adenylate cyclase inhibition in response to NPY2–36 compared with either Y1 or Y5 transfected cells (53)
. Higher order receptor complexes composed of NPY receptors and other receptors/proteins have also been postulated based on pharmacological studies (54)
. For example, NPY-induced proliferation of Y1-transfected CHO cells has been coupled to the epidermal growth factor (EGF) receptor (EGFR) activation in clathrin-coated pits (55)
. We currently have evidence that NPY-mediated aortic ring sprouting requires VEGF receptor (flt-1) activation and eNOS (11)
, and in vitro migration of HUVECs requires DPPIV activation (20)
. However, the exact nature of NPY receptor cluster formation and the presence of other proteins remain to be elucidated. Knowledge of the receptor subtypes mediating NPY-induced angiogenesis and their possible interaction with each other, and with other angiogenic receptors such as those of VEGF and bFGF, can be invaluable in designing more effective angiogenesis-based therapeutics.
ACKNOWLEDGMENTS
This study was supported by National Institutes of Health grants HL67357 and HL0553, awarded to Z. Z.
Received for publication January 12, 2006. Accepted for publication April 30, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  D. Stanic, G. Paratcha, F. Ledda, H. Herzog, A. S. Kopin, and T. Hokfelt Peptidergic influences on proliferation, migration, and placement of neural progenitors in the adult mouse forebrain PNAS, March 4, 2008; 105(9): 3610 - 3615. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.-Y. Chen, G.-G. Feng, K. Nishiwaki, Y. Shimada, Y. Fujiwara, T. Komatsu, and N. Ishikawa Possible roles of neuropeptide Y Y3-receptor subtype in rat aortic endothelial cell proliferation under hypoxia, and its specific signal transduction Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H959 - H967. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
