The gut hormone apelin is a major therapeutic focus for several diseases involving inflammation and aberrant cell growth. We investigated whether apelin-36 contained alternative bioactive peptides associated with normal physiology or disease. Amino acid sequence analysis of apelin-36 identified an amidation motif consistent with the formation of a secondary bioactive peptide (SCNH2). SCNH2 is proven to be mitogenic and chemotactic in normal/malignant cells and augments angiogenesis via a PTX-resistant/CT-X-sensitive G protein-coupled receptor (GPCR). Notably, SCNH2 is substantially more potent and sensitive than apelin-13 and vascular endothelial growth factor-A. Endogenous SCNH2 is highly expressed in human tumors and placenta and in mouse embryonic tissues. Our findings demonstrate that SCNH2 is a new apelinergic member with critical pluripotent roles in angiogenesis related diseases and embryogenesis via a non-APJ GPCR.
Apelin-13 (APL-13) is a gastrointestinal tract peptide that regulates a variety of physiological functions via its interaction with the APJ receptor [1-3]. APL-13 is derived from a 77 amino acid preprohormone, which is enzymatically processed into apelin-36 (APL-36), apelin-17, and apelin-16, prior to being converted to its final modified state [
Peptide amidation selectively tracks with biological activity and involves consecutive enzymatic processing events that target a recognizable amino acid motif within the precursor molecule [9-11]. We have identified a classic peptide amidation motif (GRRK or GRKK) within the APL-36 prohormone that encodes for a carboxy-terminal (C-terminal) glycine amide and is highly conserved in mammalian/marsupial species but is not found in lower vertebrates. The human peptide, denoted as “selective apl-36 cutting and amidation” peptide (SCNH2), has a predicted amino acid sequence of L-V-Q-P-R-GS-R-N-G-P-G-P-W-Q-G-NH2.
Given that other peptide amides, such as adrenomedullin (AM), are proven mitogenic factors and have alternative C-terminal derivatives with either low or absent bioactivity [12-16], SCNH2 (amide), SCOH (free-acid) and SCGly (glycine-extended intermediate) peptides were synthesized to evaluate their effects on cell proliferation. Only the peptide amide derivative proved to be bioactive. Accordingly, we evaluated whether SCNH2 is involved in tumor angiogenesis and metastasis using tumor and endothelial cell proliferation, migration, and in vitro/in vivo angiogenesis assays. In addition, selected SM inhibitors blocked SCNH2 activated phosphorylation of PI3/Akt and p44/42 MAPK signaling pathways as well as that of APL-13 [17,18]. Using a specifically designed fluorescent activated cell-sorting (FACS) assay, we enriched the SCNH2 receptor positive cells from a human breast carcinoma cell line (MB435). Importantly, using a CTX/PTX sensitivity assay, the SCNH2 putative receptor was characterized as a CTX-sensitive and PTX-resistant G protein-coupled receptor, which is distinctly different from the APL-13 receptor, APJ receptor, because the latter is a PTX-sensitive G protein-coupled protein [
SCOH, SCGly, and SCNH2 were synthesized by GenScript; biotinylated SCNH2 (bSCNH2) were synthesized by Princeton BioMolecules; gastrin-releasing peptide (GRP), adrenomedullin (AM), APL-36, APL-13, luteinizing hormone-releasing hormone (LHRH), arginine vasopressin (AVP) and oxytocin (OXT) were purchased from Bachem, and vascular endothelial growth factor A (VEGF-A) was obtained from the R&D Systems; LY2228820, LY294002, U01216, AMD3100, and PTX were purchased from Sigma-Aldrich; normal rabbit IgG (IgGNRS) and IgGHIRS were produced by Epitomics; the anti-APJ receptor antibody was purchased from Santa Cruz; the human umbilical vein endothelial cell line (HUVEC) was purchased from Lonza; HMEC-1s was authorized by the Centers of Disease Control and Prevention; the immortalized porcine aortic endothelial cell line (PAE) was agift from Dr. C.H. Heldin, Sweden; the immortalized human mast cell line (HMC-1) was kindly provided by Dr. J. H. (Butterfield, MN, USA); the human ocular melanoma (92.1) was a gift of Dr. Libutti, (NCI). The rest of the cell lines were acquired from the ATCC. An official NCI/MTA was obtained for the acquisition of all cell lines.
The ATPlite one-step firefly luciferase assay (PerkinElmer) was used for all cell proliferation studies. In short, 1250 cells were seeded at a volume of 50 μl per well in appropriate media containing 0.5% FBS in a white wall 96-well plate with a clear bottom (Costar 3610). Following overnight incubation, the peptide was added to the each well to obtain a final volume of 100 μl and the cells were incubated for 3 - 5 days at 37˚C/5% CO2 and terminated according to the company protocol.
ELISA assays were a modification of our previously used RIA [
Freshly prepared HUVEC cells were seeded at a concentration of 25,000 cells per well of a 96-well culture plate. The wells were first coated with 50 μl of GeltrexTM basement membrane matrix (Invitrogen). The cells were resuspended in medium at a concentration of 5 × 105 cells/ml. 25,000 cells were added per well. Positive control cells were cultured with medium supplemented with 1.0% fetal bovine serum (FBS). Negative control cells were cultured with serum-free medium. Cells were incubated at 37˚C/5% CO2 with. SCNH2 at the following concentrations: 1.0 pM, 10 pM, 100 pM, 1 nM, 10 nM, 100 nM, 1 mM, and 10 mM (n = 3). After 6 hours of incubation at 37˚C, images of the experiments were taken using an inverted epi-fluorescent microscope (Olympus, IX70) and the tube length was measured using the WimTube software (WIMASIS, Germany).
This procedure followed the previously described protocols [
Angiogenesis assays were performed as described previously [
NMuMG, MCF-7, or HMEC-1 cells were seeded in 60 mm plates (6 × 105 to 1.2 × 106 cells/plate) and they were serum starved for 24 hours. The following day, the cells were stimulated with various concentrations of APL-13, SCNH2, SCOH, and SCGly peptides over various different times at 37˚C/5% CO2. Protein extraction and Western blot analysis were performed as previously described [
For both cell migration and invasion assays, 96-well Chemo Tx 101-8 plates (Neuroprobe) were used pursuant to the previously reported protocols [
MB435 cells were dissociated when they reached 70% - 80% confluency in the culture. The cells were washed 3x in RPMI 1640 medium with 5%FBS and resuspended in the same medium. Following 1 hour incubation on a shaker for antigen recovery at room temperature, the cells were washed 3x in PBS and suspended in PBS with 1% BSA at 1 × 107 cells/ml. 5 × 105 cells were incubated with 1:100 streptavidin-Alexa 488 (SA 488, Invitrogen) and 5 μM of bSCNH2 + 10 μg/ml of SA488 (1:100) and placed on ice for 15 minutes and centrifuged for 5 minutes at 4˚C. The cells were suspended in 0.5 ml of PBS with 1% BSA for analyzing and sorting using a FACS (ARIA II, BD). Cells with positive signals in the 488-channel (now referred to as enriched MB435 cells) were sorted for further culturing. We used the same procedures to perform a binding inhibition assay to determine the specificity of the SCNH2 receptor recognition in MB435 cells. The enriched MB435 cells were incubated with 50 μM of SCNH2, APL-13, other glycine amide peptides LHRH, AVP and OXT, and other C-terminal amides such as AM and GRP following the incubation of 5 μM of bSCNH2 + 10 μg/ml of SA488. The positive control was incubated with the incubation of 5 μM of bSCNH2 + 10 μg/ml of SA488 and the negative control was incubated with 10 μg/ml SA488. All of the experiments were then analyzed using the FACS.
Human tumor cell lines/tumor tissue arrays and human placenta tissue were acquired through IRB approved protocols (TARP, LP, CCR NCI). Paraffin sections were deparaffinized in xylene and rehydrated in a graded series of ethanol. To prepare the cytospinning slides of MB435 cells, we suspended 50,000 dissociated cells in 1 ml of PBS and added a 100 μl solution to a Cytoslide loaded onto a Cytofunnel. The cells were centrifuged for 5 minutes at speed 1,000 rpm using a Cytocentrifuge (Thermo Scientific). The slide was washed in PBS before any further IHC analysis. Endogenous peroxidase was blocked using a 3% hydrogen peroxidase-methanol solution. A staining procedure was performed using a Histostain-Plus Kit (Invitrogen). 1:6000 diluted IgGHIRS and IgGNRS and 1:2000 diluted anti-APJ antibody were used for the IHC staining. SCNH2 or APL-36 was used for absorption at 50 μM. Slides were mounted using GVA mounting media.
Statistical analysis was carried out using either Microsoft Excel or Prism using a two-tailed Student’s t test. P values less than 0.05 were considered statistically significant.
Peptide amidation is one of the most important biosynthetic events in the production of bioactive peptide hormones [9,11]. The conserved amidation motif indicates multiple bioactivities [9,10,22]. Our C-terminal glycine amide SCNH2 was shown to be highly conserved in mammalian/marsupial species but is not expressed in lower vertebrates (
Cell proliferation is a key process driving angiogenesis, carcinogenesis and embryogenesis [
To identify the binding specificity of purified hyperimmune rabbit anti-SCNH2 IgG (IgGNRS) should be “IgGHIRS” for SCNH2, we performed modified ELISA (
Angiogenesis is the primary pathological process for the development of nearly all solid tumor growth and a critical step for tumor metastasis [24,25]. Moreover, because APL-13 promotes angiogenesis [6-8], we assessed whether SCNH2 had a similar function. First, we evaluated the effects of SCNH2 on HUVEC cell tubule formation. During a 6-hour exposure, compared to the control, we observed a classic rise/fall response with SCNH2 inducing maximal tube formation at 10 nM (p < 0.05,
the angiogenic compounds tested in vitro necessarily have angiogenic effects in vivo. Therefore, we investigated the in vivo angiogenic activity of SCNH2 and APL-13 using the CAM assay. SCNH2 and APL-13 dramatically induced neovascularization at all of the tested concentrations (p < 0.005, left in
APL-13 modulates the PI3K/Akt and p44/42 MAPK signaling pathways in NMuMG [26,27]. For purposes of comparison, we assessed whether and how SCNH2 induces p44/42 MAPK and PI3K/Akt phosphorylation (P-p44/42 MAPK and P-PI3K/Akt) using the NMuMG cells and found differences in the potency of peptide-induced signaling (
Previous studies showed that the SM inhibitors to p44/42 MAPK (U0126) and PI3K/Akt (LY294002) block the phosphorylation of these signaling pathways [17,18]. We therefore evaluated whether these SM inhibitors are able to block the phosphorylation activities induced by SCNH2 in the MCF-7 (
Cell proliferation, migration, and invasion are critical events of embryogenesis, wound healing, and malignant diseases [24,25]. However, these events may not share the same signaling pathways. Therefore, in addition to evaluating SCNH2 on cell proliferation, we examined the effects of SCNH2 and APL-13 on human endothelial/tumor cell migration/invasion, as well as the characteristics and suppressive activities of IgGHIRS and SM inhibitors on these activities (
To characterize the putative SCNH2 receptor, we employed a FACS assay with (bSCNH2) and (SA488) to target responding MB435. Initially, we tested whether bSCNH2 is actively similar to SCNH2. The results showed that both peptides are mitogenic (
6(b)). 4.8% of the SCNH2 receptor positive cells were identified in the positive control (Parent,
Furthermore, we compared the IHC staining of SCNH2 and the APJ receptor using the MB435 cells (
Finally, to assess the anatomical expression of SCNH2, we used IgGHIRS IHC analysis of paraffin embedded tissue from a variety of sources (
An analysis of ninety-four patients with non-small cell lung cancer showed that the APL mRNA expression was significantly elevated in tumor tissue compared to normal lung tissue and that messenger levels track with increased microvessel density and poorer prognosis [
The apelinergic system (APL/APJ) is the new therapeutic focus for several human diseases involving inflammation and aberrant cell growth [
factors and have alternative C-terminal derivatives with either low or absent bioactivity [12-16], we assessed SCNH2, SCOH, and SCGly derivatives for the trophic function. SCNH2 proved to be the only derivative with mitogenic activity, which confirmed the critical importance of an intact C-terminal amide residue for bioactivity and receptor recognition capability [10,11].
APL-13 plays an important role in stimulating epithelial cell growth, activating PI3K/Akt and p44/42 MAPK signaling pathways, modulating angiogenesis, regulating cell migration/invasion [
APL-13 activates the phosphorylation of PI3K/Akt and p44/42 MAPK signaling pathways via a Gi/oPCR, the APJ receptor [1-3]. We investigated whether SCNH2 was able to regulate the same signaling pathways and whether its regulation was mediated by the same receptor. Interestingly, we found that SCNH2 activated the PI3K/Akt and p44/42 MAPK signaling pathways through a receptor distinctly different from the one mediating APL-13 function. The PTX sensitivity assay has been used for Gi/o protein characterization and the APJ receptor is a PTX-sensitive Gi/o protein [19,28,30]. SCNH2 activation of PI3K/Akt and p44/42 MAPK signaling pathway is PTX resistant while being CTX sensitive (Gs component), which indicates that SCNH2 and APL-13 use different GPCRs.
Although SCNH2 was synthesized using a predicted amino acid sequence, its antibody (IgGHIRS) was able to specifically detect the endogenous expression of SCNH2 in normal human tissue, placenta, and tumors, and in mouse embryonic tissues. It is noteworthy that there is significant augmentation of SCNH2 expression in the mouse embryonic organs and in human tumors. Because there is a divergence in SCNH2 expression between viviparous and oviparous animals, we concluded that there was probably a significant evolutionary development in the implantation process given SCNH2 expression in fetal trophoblasts and its functional role in cell migration/invasion. Also, SCNH2 is able to execute its bioactivities at low nanomolar to picomolar concentrations. This suggests that SCNH2 could potentially play critical physiological roles in embryogenesis and carcinogenesis and may serve as a clinically relevant biomarker for certain malignant events.
We thank J. Prescott (Epitomics, Inc.) for assisting in the development of rabbit anti-SCNH2 reagents and for multiple discussions on assay design. We thank M. Shiue (Princeton Biomolecules) for synthesizing SCNH2/bSCNH2 peptides and a helpful hint on peptide chemistry. We acknowledge D. Goldstein and S. Segal (National Cancer Institute) for establishing the NCI contract with Epitomics, Inc. A.C.A./R.F.N. grant support: NIH/NHLBI HL52585 and Merit Review Grant from Veterans Affairs Medical Research Services.
C.F., I.A., C.B., D.S.S., and F.C. designed the research. I.A. performed cell proliferation assays. C.B., N.H., and D.S.S. assessed phosphorylation of signaling pathways, time course analysis and PTX studies. C.F. performed migration/invasion assays and developed bSCNH2/SA 488/FACS ligand binding assay. A.C.A. and R.F.N. ran the rat aortic ring assays and generated the quantitative data. L.A.F. and J.D.L. completed the CAM analysis for in vivo angiogenesis testing and generated the associated quantitative data. F.C. characterized the binding specificity of rabbit anti-SCHN2 reagents. W.G.S.-S. did the protein A purification and desalting of IgG-NRS/HIRS reagents. J.M., K.Y., S.M.H., C.F. and I.A. performed the tissue preparation and IHC staining. M.A. involved with assay design and experimental discussions. W.G.S.-S. board certified pathologist who critiqued IHC of human tumor tissue. C.B. and D.S.S. helped prepare the manuscript. C.F., I.A., and F.C. wrote the manuscript.