A cholesteryl-functionalized derivative of activity dependent neurotrophic factor-9 peptide (a nine amino acid core peptide of activity-dependent neurotrophic factor, acting against Alzheimer’s disease) was synthesized aiming at the improvement of its bioavailability. Therefore, its uptake was comparatively investigated with that of its parent peptide by employing mouse neuroblastoma Neuro-2a cells. Owing to the hydrophobic character of this cholesteryl-functionalized peptide, it exhibited enhanced permeability and intracellular uptake while it also retained its low cytotoxicity at concentrations up to 1 μm. FACS analysis also revealed that when Neuro-2a cells were treated with this activity dependent neurotrophic factor-9 derivative, at a concentration of 50 nm, an almost 100% uptake was obtained. In addition, in vitro biological activity experiments showed that the functionalized peptide retained its neurotrophic activity at femtomolar concentration range.
Constantly increasing human lifespan leads to an enhancement of central nervous system (CNS) disorders among the population. Interestingly, more than 98% of potential drugs for CNS disorders are ineffective owing to their inability to be effectively transported into the brain (1). The major obstacle to target brain is the presence of the blood–brain barrier (BBB), which screens blood components from penetrating into the brain. Thus, many hydrophilic compounds fail to enter the brain by systemic administration, and only small (<400 Da), lipid-soluble drugs are able to diffuse passively across the BBB (1,2). Various transport strategies, including carrier-mediated transport, receptor-mediated endocytosis, and adsorptive-mediated endocytosis systems, have been described (1–5).
For molecules to effectively cross cell membranes, it is crucial to have an appropriate balance between hydrophobic and hydrophilic moieties. Thus, permeability of the hydrophilic oligoarginine was facilitated, both in vitro (6) and in vivo (7), by the introduction of myristoyl moiety owing to hydrophobicity tuning. A combination of oligoarginine (7-mer) and fatty acid (C14) chain length afforded a seven- to eightfold higher uptake into cells compared to oligoarginine 7-mer. Additionally, modification of short bioactive peptides with lipoaminoacids, bearing chains from 12 to 20 carbons, resulted in increased stability and permeability (8,9). Last but not least, mitochondria-penetrating peptides bearing the right blend of hydrophobic and cationic moieties have been prepared. Thus, when cationic amino acid side chains were replaced with cyclohexane moieties, the obtained peptides entered mitochondria (10).
Many studies have emphasized the importance of neurotrophic factors on acting as neuropharmaceuticals for a variety of brain diseases (1,2,11). Among them, activity dependent neurotrophic factor (ADNF) exhibits neuroprotection and acts against Alzheimer’s disease (12). Structure–activity relationship studies have revealed that smaller peptides retain this activity. For instance, a fourteen amino acid peptide of ADNF (ADNF-14) (13), a nine amino acid peptide (ADNF-9) (12), or an eight amino acid peptide (NAP) (14) share this activity. Small peptides mimicking the parent protein activity are more potent owing to effective penetration and reduced enzymatic degradation sites (15).
In this study, for enhancing membrane transport of ADNF-9, a novel hydrophobic prodrug of this peptide was synthesized by the introduction of a hydrolysable cholesteryl moiety at its amino end-group. The transport of this derivative along with its neuroprotective activity against N-methyl-d-aspartate was comparatively investigated with those of the parent peptide, employing Neuro-2a cells. Although ADNF is a glia-derived neuroprotective protein (12), ADNF-9 maintains its neuroprotective action (13) in glia-depleted neuronal cultures. Therefore, Neuro-2a mouse neuroblastoma cell line was used as an in vitro neuronal model.
Materials and Methods
Fmoc-protected amino acids and Fmoc-chloride were purchased from Chemical and Biopharmaceutical Laboratories (Patras, Greece) and Novabiochem. The solvents and other chemicals used in the peptide synthesis and in RP-HPLC were products of Merck or Sigma. N-methyl-d-aspartate (NMDA), heat-inactivated fetal bovine serum (FBS), high and low glucose Dulbecco’s modified Eagle Medium (DMEM), MEM, RPMI, Trypsin-EDTA, l-glutamine, penicillin-streptomycin solution, and MTT were obtained from Sigma. MTS cell proliferation assay kit was purchased from Promega.
ADNF-9 peptide derivatives were synthesized on an in-house-prepared polystyrene resin (0.5 mmol/g) (Figure 1) (16). Solid-phase peptide synthesis was performed manually. The following side-chain-protected amino acids were used: Fmoc-Ser(OtBu), Fmoc-Arg(Pbf), and Fmoc-Lys(Dde)OH. Coupling was performed according to standard procedure (17). Coupling efficiency was checked by using the Kaiser ninhydrin test (18). After completion of ADNF-9 peptide synthesis (SALLRSIPA), N-terminal Fmoc group was removed using a (20% v/v) piperidine/DMF solution for 20 min, which allowed coupling of Fmoc-protected 6-aminohexanoic acid (Fmoc–Aca) to the free amino group. Subsequently, Fmoc–Aca was coupled to the free amino group. After Fmoc removal, the free amino group of Aca was reacted with fluorescein isothiocyanate (FITC) (2.5 equiv), dissolved in pyridine, for 1 h. The peptide FITC-Aca-ADNF-9 was cleaved from the resin using a TFA/H2O/TIPS (95/2.5/2.5% v/v) mixture for 2 h, and the crude peptide derivative was precipitated in ether and isolated following centrifugation. The solid compound was purified by semi-preparative RP-HPLC employing a Waters HPLC System (pump 600E, detector UV-484) and a 10 Nucleosil 7 C18 column (250 × 12.7 mm ID; Macherey Nagel). Peptide peaks were detected spectrophotometrically at 220 nm. The mobile phase consisted of 0.05% v/v TFA/H2O (solvent A) and 90% (0.05% v/v TFA/CH3CN)/10% A (solvent B). Elution was achieved by applying a linear gradient from 80% A to 0% A in 70 min. The flow rate was 3 mL/min. The retention time of FITC-Aca-ADNF-9 was 32.1 min. Analytical RP-HPLC was performed on a Waters HPLC System (pump 616E, detector 996 PDA) using a Symmetry300TM (Milford, MA, USA) C4 column (150 × 3.9 mm ID; 5 μm particle size, Waters). The solvent system consisted of 0.05% TFA/H2O (solvent A) and 90% (0.05% TFA/CH3CN)/10% A (solvent B). Elution was achieved by applying a linear gradient from 80% A to 10% A in 20 min. The flow rate was 1 mL/min. ESI-MS analysis was performed using an electrospray interface mass spectrometer (Finnigan AQA Thermoquest).
Cholesteryl-functionalized ADNF-9 peptide (CHOL-FITC-Aca-ADNF-9) was prepared by an analogous manner. In this case, Fmoc-Lys(Dde)OH was coupled to Aca-ADNF-9 peptide, and following Fmoc-group removal, cholesterol chloroformate dissolved in a DIEA/DCM solution was introduced to the free Nα- amino-group of the Lys residue. The side-chain Dde-protecting group was removed using a hydrazine/DMF solution (2% v/v) for 2 × 3 min, and subsequently, free Nε- amino-group of Lys was reacted with FITC. The peptide was cleaved, precipitated, and isolated as described above. The solid compound (CHOL-FITC-Aca-ADNF-9) was purified with semi-preparative RP-HPLC and characterized with mass spectrometry. The retention time of FITC-Aca-ADNF-9 was 61.5 min. The HPLC system as well as the mobile phase and the gradient conditions used were the same as described above. ESI-MS analysis was performed using an electrospray interface mass spectrometer (Finnigan AQA Thermoquest).
Mouse neuroblastoma (Neuro-2a) cells were maintained in high glucose DMEM, supplemented with 10% heat-inactivated FBS, 2 mm l-glutamine and antibiotics (100 units/mL penicillin and 100 μg/mL streptomycin). Cells were grown at 37 °C in a humidified atmosphere with 5% CO2.
Cell penetration experiments of functionalized peptides were performed in Neuro-2a cells, employing FITC-labeled peptides. The peptide was dissolved in 15-mm DMSO-containing PBS at a concentration of 1.5 mm. The cells were cultured in DMEM, supplemented with 10% FBS. Neuro-2a were seeded at a density of 2 × 104 cells per well in a 48-well plate and grown in 700 μL of growth medium 24 h prior to the incubation with peptides. After treating cells with the peptide (1 nm to 10 μm), in the presence of 1.5 mm DMSO, for 4 h in serum-free media (SFM), peptide and DMSO were removed by washing once with 700 μL SFM and with PBS. The medium was replaced with 700 μL per well of DMEM, supplemented with 10% FBS. After 4 h of further incubation, cells were observed under a fluorescence microscope (LEICA DM IL microscope with I3 filter cube, ×250 magnification).
Peptide uptake was assessed by flow cytometry. Cells were cultured in DMEM, supplemented with 10% FBS. Neuro-2a cells were seeded at a density of 2.5 × 105 cells per well in a 6-well plate and grown in 2 mL of growth medium 24 h prior to the incubation with the peptides. After treating cells with the peptide (25–150 nm) for 4 h in SFM, the growth medium was removed, and the cells were washed once with SFM. Subsequently, cells were trypsinized and resuspended in 2-mL completed growth medium. Cells sedimented after centrifugation, and the pellet was washed twice with PBS. Cells were resuspended in PBS, supplemented with 3% fetal calf serum (FACS buffer) and allowed at 4 °C until analysis. Results are expressed as the mean value of three independent experiments ± standard deviation (n = 3).
In vitro cytotoxicity studies
Cytotoxicity assay was performed by the modified MTS assay (19). Briefly, Neuro-2a cells were seeded at a density of 1 × 104 cells per well in a 96-well plate and grown in 200 μL of growth medium 24 h prior to the incubation with the peptides. Peptides were dissolved in 15-mm DMSO-containing PBS at a concentration of 1.5 mm. Medium was removed, and fresh medium was added to each well, containing the appropriate concentration of each peptide (50 μm to 1 mm) at a total volume of 100 μL. After treating cells with the peptides for 4 h in the presence of 1.5 mm DMSO, the medium was replaced with completed growth medium, and cells were incubated for 24 h. Following this period, the medium was replaced with 100 μL per well of MTS solution (0.5 mg/mL diluted in growth medium). Cells were further incubated for 3 h at 37 °C and 5% CO2. The absorbance was measured at 490 nm (reference filter 630 nm) using a microplate reader (SUNRISE, XFLUOR4). Measurements were converted to percent viability by comparison to control experiments in which peptides had not been added. Results from the MTS assays are expressed as the mean value of the absorption at 490 nm ± standard deviation (n = 3).
Neurotrophic activity assay
Cell survival after treatment with NMDA was determined with the MTT assay (20). Briefly, Neuro-2a cells were seeded at a density of 3 × 103 cells per well in a 96-well plate and grown in 100 μL of growth medium 24 h prior to the incubation with the peptides. Peptides were dissolved in DMSO-containing PBS at a concentration of 1.5 mm. Medium was removed, and fresh serum-free medium was added to each well containing the appropriate concentration of each peptide at a total volume of 200 μL. After pretreating the cells with the peptides, NMDA was added to a final concentration of 100 μm. Three days later, the medium was removed and replaced with 100 μL per well of MTT solution (1 mg/mL diluted in RPMI). The cells were further incubated for 4 h at 37 °C and 5% CO2. The absorbance was measured at 540 nm (reference filter 620 nm) using a microplate reader (Sirio S, SEAC Radim group). Measurements were converted to percent viability by comparison to control experiments in which peptides had not been added. Results from the MTT assays are expressed as the mean value of the absorption at 540 nm ± standard deviation (n = 4). Control experiments were also conducted in which the same procedure was followed in the presence of NMDA alone.
Results and Discussion
Synthesis of peptides
Synthesis of ADNF-9 peptide and its functionalized derivatives was performed according to solid-phase strategy (16). Both peptides (Figure 1) were labeled with FITC for tracking their intracellular uptake. After completion of the peptide’s sequence, a short alkyl chain was introduced for the peptide to retain its activity intact. Subsequently, Fmoc was removed, and free amino group of Aca was reacted with FITC. The isolated compound (FITC-Aca-ADNF-9) was purified by semi-preparative RP-HPLC, analyzed with RP-HPLC, and characterized with mass spectrometry. Under the conditions used, the retention time of FITC-Aca-ADNF-9 was 11.3 min. In the ESI-mass spectrum the doubly charged mass ion was observed (m/z = 715.7 for [M+2H]2+). Experimentally determined molecular mass (1429,6) was in agreement with the theoretical value (1429,4 for C68H96N14O18S). Pure peptide was obtained in a 45% yield.
For cholesteryl-functionalized derivative, a protected lysine residue (Fmoc-Lys(Dde)OH) was introduced after the spacer for attaching both the cholesteryl and FITC moieties. Thus, Fmoc-group of the Lys residue was selectively deprotected, and cholesterol chloroformate was introduced to the free Nα- amino-group of the Lys residue. The side-chain Dde-protecting group was removed, and the free Nε- amino-group of Lys was reacted with FITC. CHOL-FITC-Aca-ADNF-9 was purified by semi-preparative RP-HPLC, analyzed with RP-HPLC, and characterized with mass spectrometry. Under the conditions used, the retention time of CHOL-FITC-Aca-ADNF-9 was 18.4 min. In the ESI-MS spectrum, the doubly charged mass ion was observed (m/z = 986.1 for [M+2H]2+). The molecular mass (1970,2) was in agreement with the theoretical value (1970.4 for C102H152N16O21S). The yield of pure peptide was 31%. Both peptide derivatives were obtained in purity higher than 95%.
Cell penetration ability of the FITC-labeled ADNF-9 derivatives was investigated in Neuro-2a cells employing fluorescence microscopy. The peptides were dissolved in DMSO-containing PBS at a concentration of 1.5 mm. After 4 h incubation, CHOL-FITC-Aca-ADNF-9 demonstrated satisfactory intracellular uptake, while the parent FITC-Aca-ADNF-9 peptide (Figure 2A) was not detected. Fluorescent cells were only observed when Neuro-2a cells were treated with 5 and 10 μm CHOL-FITC-Aca-ADNF-9 (Figure 2B,C), while at lower concentrations fluorescent cells were not observed. In the absence of any peptide (control), fluorescent cells were not detected (Figure 2D). Cholesteryl-functionalized ADNF-9 peptide appears to be intracellularly uptaken, with marginal membrane staining (Figure 2).
For quantifying cellular association of cholesteryl-functionalized ADNF-9 peptide, fluorescence-activated cell-sorting analysis (FACS) was performed (Figure 3). The samples were thrypsinized and washed twice with PBS, thus excluding non-specifically bound fluorescence.
At a concentration of 25 nm functionalized peptide, almost 55% of the cells were fluorescently labeled (Figure 4A), while at a concentration of 75 nm, a plateau was reached (93%). A linear correlation (R2 = 0.99124) between the peptide’s concentration and the detected fluorescence was found, suggesting a quantitative interaction between the hydrophobic ADNF-9 and cellular membrane (Figure 4B).
At the same concentration range of the parent FITC-Aca-ADNF-9 peptide, only a marginal percentage of cell population was fluorescently labeled (9% and 6% at a concentration of 125 and 150 nm, respectively) (Figure 4A). It may, therefore, be assumed that at a first stage, the cholesteryl moiety of the peptide by its anchoring (21) inside cell membranes leads to peptide’s attachment on their surface, followed by its intracellular uptake at the second stage. In this connection, dealing with the mechanistic aspects of membrane transport, it should be noted that the small size of the hydrophobic ADNF-9 peptide does not favor endocytosis. In fact, it has been established (22) that the optimal nanoparticle radius for endocytosis to occur is of the order of 25–30 nm, that is, much bigger than the peptides of this study.
Concomitant to the enhanced permeability and intracellular accumulation of the cholesteryl-functionalized ADNF-9 peptide is the possibility of exhibiting cytotoxicity that was tested by modified MTS assay (19). On performing a cell proliferation assay, only marginal cytotoxicity was observed (Figure 5). After 4 h incubation with either parent peptide (FITC-Aca-ADNF-9) or hydrophobic peptide (CHOL-FITC-Aca-ADNF-9), more than 90% of the Neuro-2a cell viability was retained at concentrations up to 1 μm.
A biological activity assay was conducted to confirm whether the neuroprotective activity of the functionalized peptide is retained. Therefore, we have used as a positive control pure ADNF-9 peptide known to exert neuroprotective activity against NMDA (12,23). NMDA is an amino acid derivative which acts as a specific agonist at NMDA receptor. Abnormal activation of the NMDA receptor has been suggested to lead to neuronal cell death observed in many acute and chronic disorders such as ischemia, stroke, Alzheimer’s disease, and Huntington’s disease (24). Exposure to NMDA (100 μm) significantly reduced cell viability by 70%. The treatment with NMDA in the presence of pure ADNF, or its derivatives, FITC-Aca-ADNF-9 and CHOL-FITC-Aca-ADNF-9, recovered cell viability (Figure 6). In fact, at femtomolar concentrations (100 fm), at which the peptide’s neuroprotective action is primarily exerted, all the peptides examined showed comparable neuroprotective action, whereas at higher concentrations the parent peptide, ADNF-9, retained its activity, while the functionalized peptide was only partially neuroprotective.
Enhancing ADNF-9 hydrophobicity, through cholesteryl functionalization, proved a useful strategy for facilitating permeability of the peptide across mouse neuroblastoma Neuro-2a cell membranes. In addition, this hydrophobic peptide partially retained its neuroprotective activity being concentration-dependent. This behavior can be justified by the fact that cholesteryl moiety, which is attached to ADNF through a carbamate group, is hydrolyzed in the biological environment affording the active parent peptide (25,26). Thus, this approach of enhancing lipophilicity can be considered as an effective methodology for increasing peptides bioavailability (27). Our present results have shown that cholesteryl-functionalized ADNF-9 is a useful prodrug, justifying therefore, further biological studies to be performed in the near future.
The authors would like to thank Dr. Maria Evangelidou for her help in the FACS analysis. The work was partially supported by EU under the ‘NMP’ INTEGRATED PROJECT, NMP4-CT-2006-026723.