Single-crystalline hyperbranched nanostructure of iron hydroxyl phosphate Fe5(PO4)4(OH)3·2H2O for highly selective capture of phosphopeptides.

Single-crystalline hyperbranched nanostructures of iron hydroxyl phosphate Fe5(PO4)4(OH)3·2H2O (giniite) with orthorhombic phase were synthesized through a simple route. They have a well-defined dendrite fractal structure with a pronounced trunk and highly ordered branches. The toxicity test shows that the hyperbranched nanostructures have good biocompatibility and low toxicity level, which makes them have application potentials in life science. The study herein demonstrated that the obtained hyperbranched giniite nanostructures show highly selective capture of phosphopeptides and could be used as a kind of promising nanomaterial for the specific capture of phosphopeptides from complex tryptic digests with the detection of MALDI-TOF mass spectrometry.

Guiding nanoscaled building blocks such as nanoparticles, nanorods, nanowires and nanotubes into hierarchical complex architectures is a key step for building functional electronic and photonic nanodevices123456. Hyperbranched structures, as one type of complex architectures generally forming under nonequilibrium conditions, have a wide range of size scales in nature78. Recently, some of spontaneous pattern in nature have been investigated in the growth of crystals. For example, well-defined α-Fe2O3 dendritic fractals were fabricated from the hydrolysis of K3[Fe(CN)6] through spontaneous self-assembly of nanoscale building blocks9, and 3D PbS dendritic structures were synthesized by a surfactant/ligand-assisted hydrothermal method10. A survey of the literatures on hyperbranched structures reveals that most of hyperbranched crystals have hexagonal or cubic symmetry, but hyperbranched materials with low crystal symmetry (e.g., monoclinic or orthorhombic) have seldom been reported11. On the other hand, the hyperbranched structures are usually fabricated with the assistance of templates, microemulsions, surfactant micelles, and so on121314. Thus, it is a big challenge to develop simple methods to realize the synthesis of hierarchically hyperbranched architectures of those materials with low crystal symmetry.

Iron hydroxyl phosphates (Fex(PO4)y(OH)z·nH2O), such as barbosalite Fe3(PO4)2(OH)2, whitmoreite Fe3(PO4)2(OH)2·4H2O, giniite Fe5(PO4)4(OH)3·2H2O and beraunite Fe6(PO4)4(OH)5·6H2O, are well-known minerals with rich crystal chemistry and important oxidative dehydrogenation catalysts15161718. They are also significant in passivation of metal surfaces, corrosion inhibition and in the reactions of iron compounds contained in various soils with phosphate fertilizers1920. The control on the morphology of micro- and nano-sized Fex(PO4)y(OH)z·nH2O with well-defined shapes are important for their applications in catalysis, glass industries and environmental science and technology212223, but little literature has been focused on it, especially for the controlled synthesis of hyperbranched iron hydroxyl phosphate nanostructures. Herein, we reported a simple solution route to synthesize hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures with single crystallinity in nature and demonstrated that these hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures are biocompatible and can be used for highly selective capture of phosphopeptides.

Results

The hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures were synthesized on a large scale and in high purity by hydrothermally treating the aqueous solution of Fe(NO3)3·9H2O (0.202 g) and Na3PO4·12H2O (0.19 g) in 15 mL of distilled water at 220°C for 8 h. Figure 1a displays XRD pattern of the as-synthesized product, which can be easily indexed to be giniite Fe5(PO4)4(OH)3·2H2O (JCPDS card no. 45-1436). Figure 1b ~ d show SEM images of the obtained giniite Fe5(PO4)4(OH)3·2H2O. The low-magnification SEM image in Figure 1b shows that the product consists of hyperbranched nanostructures. The yield of the hyperbranched structures is high and almost all of the crystals in the sample have highly ordered structures. The high magnification SEM images shown in Figure 1c and d reveal the details of the hyperbranched nanostructures. They have a clear and well-defined dendrite fractal structure with a pronounced trunk and highly ordered branches distributed on both sides of the trunk.

Figure 1

(a) XRD pattern and (b~d) SEM images of the as-synthesized Fe5(PO4)4(OH)3·2H2O nanostructure.

TEM investigations confirm the results from SEM observations. The obtained giniite Fe5(PO4)4(OH)3·2H2O crystals have a hyperbranched nanostructures as the TEM image of a single dendrite in Figure 2a displays. Further structural characterizations of the dendrite were carried out by selected area electron diffraction (SAED). Figure 2b shows the SAED pattern of the whole single dendrite in Figure 2a, displaying a set of diffraction spots and confirming that the whole dendrite is single-crystalline in nature. The diffraction pattern indicates that the trunk of dendrite is oriented along [201] with the two branches along [20-1]. Figure 2c, e, and g show TEM images taken from the areas labeled c, e and g in Figure 2a; they respectively represent the structures of the tip and both side branches of the dendrite. Interestingly, the SAED patterns of these three parts shown in Figure 2d, f and h, are all as same as that of the whole dendrite, further confirming that the whole dendrite of giniite Fe5(PO4)4 (OH)3·2H2O is single-crystalline. The synthesis of single-crystalline Fe5(PO4)4(OH)3·2H2O hyperbranched structure enriches the nanostructure of iron hydroxyl phosphates and would bring more application potentials for giniite.

Figure 2

TEM images and corresponding SAED patterns taken from the different parts of the as-synthesized Fe5(PO4)4(OH)3·2H2O nanostructures.

Discussion

A series of experiments were carried out to research how the experimental conditions affect the morphologies and the structure of the products. When the used amounts of Fe(NO3)3·9H2O and Na3PO4· 12H2O are fixed at 0.202 g and 0.19 g, respectively, the obtained products are all hyperbranched nanostructures (SEM images showing in Figure 3a ~ b) when the reaction temperature changes from 220°C to 160°C under hydrothermal conditions for 8 hours. But there are still some differences between these novel structures. With the decrease of the reaction temperature, the trunk of the hyperbranched nanostructures appears to be much thicker and those branches encompassing with it seem fewer. Furthermore, with the decrease of the temperature, more and more particles other than hyperbranched structures appear and when the reaction temperature reduced to even lower at 120°C, only Fe5(PO4)4(OH)3·2H2O particles can be obtained (Figure 3c). But at a lower temperature (such as 180°C), prolonging the reaction time can make the small particles disappear and pure Fe5(PO4)4(OH)3·2H2O hyperbranched structures can be obtained (Figure 3d). These results suggest that the regulation of the experiment conditions, such as reagent ratio between Na3PO4 and Fe(NO3)3, reaction time and reaction temperature in a certain range would results in the formation of the hyperbranched structure. For example, the product prepared with 0.101 g of Fe(NO3)3·9H2O and 0.095 g of Na3PO4·12H2O in 15 mL of distilled water at 180°C for 24 hours (the SEM images are displayed in Figure 4) demonstrated that in spite of some detailed characteristics, the obtained giniite Fe5(PO4)4(OH)3·2H2O crystals still show hyperbranched structure with a trunk in the middle and many branches symmetrically at the sides. For further studying the possible formation mechanism of the hyperbranched nanostructures, we also carried out more experiments at 220°C but with different reaction times such as 1 h, 2 h, 4 h and 12 h, respectively. As shown in Figure 5, with the increase of the reaction time, the morphology of the products changed from particle-like to finally hyperbranched structures. From the results obtained with different reaction temperatures and reaction times, it can be deduces that small particles would be produced at early stage and the novel hyperbranched structure may grow from small particles and then isotropically develop to hyperbranched nanostructures in a proper time.

Figure 3

SEM images of the products prepared under different hydrothermal conditions: (a) at 180°C for 8 hours; (b) at 160°C for 8 hours; (c) at 120°C for 8 hours and (d) at 180°C for 12 hours.

Figure 4

SEM images of Fe5(PO4)4(OH)3·2H2O nanostructures prepared with 0.101 g of Fe(NO3)3·9H2O and 0.095 g of Na3PO4·12H2O in 15 mL of distilled water at 180°C for 24 hours.

Figure 5

SEM images of the products prepared at 220°C for different time: (a) 1 hours; (b) 2 hours; (c) 4 hours and (d) 12 hours.

Mass spectrometry (MS) is an important tool to investigate protein phosphorylation, which is known to play vital roles in regulations of normal cells or abnormal tissues. However, due to the low abundances and the ionization efficiency of phosphopeptides, generally phosphopeptides need to be isolated and concentrated prior to MS analysis24. Due to the affinity of metal ions to phosphate group, immobilized metal affinity chromatography (IMAC) currently is the most widely used technology for the isolation of phosphopeptides from peptide mixtures, but it often displays low specificity and sensitivity to targeted phosphopeptides due to the nonspecific binding of nonphosphorylated acidic peptides and the complexity of factors affecting phosphopeptide binding and release252627. As the development of nanoscience, it is found that some nanomaterials (e.g. TiO2, Fe2O3 or ZrO2 nanocrystals) have great functions for phosphopeptide enrichment282930. Due to specific chemisorption between phosphate groups and Fe3+, more abundant in nature of iron element than titanium, and relatively good biocompatibility and lower tocxity, in this study, we investigated the ability of the obtained giniite Fe5(PO4)4(OH)3·2H2O with the novel hyperbranched structure for selective enrichment of phosphopeptides from the peptide mixtures.

Before using giniite Fe5(PO4)4(OH)3·2H2O to enrich phosphopeptides, we investigated their biocompatibility since the biocompatibility of nanomaterials plays an important role in life science. We have assessed the biocompatibility of hyperbranched giniite Fe5(PO4)4(OH)3·2H2O nanostructures by the 5-dimethylthiazol-2-yl-2, 5-diphenyltetrazolium bromide (MTT) assay, which is a simple and effective colorimetric assay to measure cell cytotoxicity and viability. Figure 6 shows cell viabilities of HepG2 cells incubated without and with hyperbranched giniite Fe5(PO4)4(OH)3·2H2O nanostructures for different time. It can be seen that Fe5(PO4)4 (OH)3·2H2O shows low level of toxicity and the cell viability values are > 90% after 48 incubation. This phenomenon might be caused by the phosphate group (PO4) which is compatible with phosphoprotein on the cell membrane. With these expected results, the good biocompatibility of hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures makes the propose to capture phosphoproteins go forward.

Figure 6

Viabilities of HepG2 cells incubated with hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures for different time (Sample 1: as-synthesized hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures; and Sample 2: the product prepared with 0.101 g of Fe(NO3)3·9H2O and 0.095 g of Na3PO4·12H2O in 15 mL of distilled water at 180°C for 24 hours).

The hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures were applied in the capture of phosphopeptides from tryptic digests of β-casein, which possesses known phosphorylation sites as the standard phosphoprotein test with detection of MALDI-TOF mass spectrometry. Firstly, the tryptic digest of bovine β-casein was shaken in vortex at room temperature for 30 min with certain amount of hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures. Followed by series of rinsing steps for removing nonphosphorylated peptides, the MALDI matrix of 2, 5-DHB was added and mixed with the peptide solution eluted from hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures and deposited on MALDI target for mass spectrometry analysis. For comparison, the same amount of the tryptic digest of β-casein was directly analyzed by MALDI-TOF MS without the trapping steps by hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures. Figure 7a shows the MALDI-TOF mass spectrum of the tryptic digest of β-casein without any treatment of nanostructures. The peaks of nonphosphorylated peptides and phosphorylated peptides arising from tryptic digest of β-casein were all detected, where the peaks at m/z 2061.94 and 3122.56 presents β and β phosphorylated peptides and the peaks marked 1, 2, and 3 are β-(HPO3), β-(HPO3) β-2(HPO3), respectively. Meanwhile, The MALDI-TOF mass spectra of the tryptic digest of the same amount of β-casein (1 pmol) after being treated with two hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures (SEM images shown in Figure 1 and 3) were shown in Figure 7b and c. In comparison with Figure 7a, there were only strong and clear mass spectrometric patterns with only phosphopeptide peaks from the tryptic digest of β-casein after the specific capture by hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures. The peak intensity value in Figure 7b is about 9000 at m/z 2061.94 which is about 90% of that in Figure 7a. In Figure 7c, the peak intensity values of these isolated phosphorylated peptides are about 9800 at m/z 2061.94) and 5700 at m/z 3122.56, which are almost the same as those in Figure 7a, suggesting that almost all the phosphorylated peptides in the digest have been captured by the hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures. These results indicate that the hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures have the specific selectivity for capturing phosphopeptides. The affinity interactions between hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures and the phosphopeptides might result from the strong chelating interactions between iron ions and the phosphoryl groups of phosphopeptides. Furthermore, by changing the experimental conditions in a certain range, the obtained giniite Fe5(PO4)4(OH)3·2H2O products show the hyperbranched structure with different details which would lead to different affinity to phosphopeptides. Thus these novel structures can be used as the promising substrate for isolation of phosphorylated peptides from a proteolytic digest mixture and effectively increase the detection sensitivity of phosphopeptides with the detection of MALDI-TOF MS.

Figure 7

MALDI-TOF-MS spectra of (a) tryptic digests of β-casein; (b) tryptic digests of β-casein after being treated with as-synthesized hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures and (c) tryptic digests of β-casein after being treated with Fe5(PO4)4(OH)3·2H2O nanostructures prepared with 0.101 g of Fe(NO3)3·9H2O and 0.095 g of Na3PO4·12H2O in 15 mL of distilled water at 180°C for 24 hours.

In summary, the present work describes the simple synthesis of hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures by a mild hydrothermal reaction. The hyperbranched nanostructures are a well-defined dendrite structure and single-crystalline in nature. The special hyperbranched nanostructures show good biocompatibility and exhibit excellent capability for isolation of phosphorylated peptides from a proteolytic digest mixture and would have great application potentials in life science.

Methods

Preparation of hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures

0.202 g of Fe(NO3)3·9H2O and 0.19 g of Na3PO4·12H2O were mixed with 15 mL of distilled water. After being stirred for 5 minutes, the resulting solution was transferred into 50 mL stainless-steel autoclaves lined with poly(tetrafluoroethylene) (PTFE, Telfon) and maintained at 220°C for 8 hours and then cooled to room temperature. The products obtained at the bottom of the autoclave were collected, washed three times with deionized water and dried in the air.

MTT proliferation assay

HepG2 cells were grown in RPMI 1640 medium supplemented with 10% freshly inactivated fetal calf serum (FCS) and were seeded equivalently into a 96-well plate; the concentration of hyperbranched Fe5(PO4)4(OH)3·2H2O nanostructures were 10 μg/mL. The plates were kept at 37°C in a humidified atmosphere of 5% CO2 and incubated for 12, 24 and 48 h; then MTT solution of an appropriate concentration (1 mg/mL) was added to each well and the plates incubated at 37°C for 4 h. The measurements of absorbance of the solutions related to the number of live cells were performed on an ELISA spectrophotometer at 570 nm.

Tryptic digestion of proteins and capture of phosphopeptides using nanostructured materials

β-casein (1 mg) was dissolved in 1 mL of 50 mM ammonium bicarbonate buffer (pH = 8.2), and then was digested at 37°C for 16 h with trypsin at the ratio of enzyme-to-substrate of 1:40 (w/w). Protein digests were diluted with loading buffer containing 6% TFA and 80% (vol/vol) ACN. 2 μL of diluted protein digest solution was added into a 10 μL of 15 mg/mL Fe5(PO4)4(OH)3·2H2O nanostructures in loading buffer and shaken at room temperature for 30 min. After that, the nanostructured materials were centrifuged and rinsed twice with 20 μL of loading buffer solutions containing 6% TFA and 80% (vol/vol) ACN with and without 200 mM NaCl, respectively. Subsequently, the trapped phosphopeptides were eluted from the nanostructured materials with NH3·H2O solution (pH 11.5, 30 μL) under sonication for 15 min. By centrifugation, the eluted supernatant was collected and lyophilized. Then,1 μL of 25 mg/mL 2, 5-DHB solution containing 70% ACN and 1% H3PO4 were added and mixed homogeneously with 1 μL the resulting solution where the addition of H3PO4 in matrix solution was for the enhancement of phosphopeptide signals on MALDI-TOF MS analysis.

Mass spectrometry

All MALDI-TOF mass spectra were acquired on a Bruker Autoflex II time-of-flight mass spectrometer (Bruker, Bremen, Germany) equipped with a delayed ion-extraction device and a 337-nm pulsed nitrogen laser. All measurements were carried out in reflected positive-ion mode with delayed ion extraction. The delay time for ion extraction and the extraction voltage were set at 90 ns and 20 kV, respectively. Each MS spectrum was acquired by the accumulation of 100 laser shots.

Author Contributions

F.G. and Q.L. guided the entire project, carried out data analyses and co-wrote the manuscript. Q.C., C.W., Y.Z. and H.P. performed the experiments and characterizations. All the coauthors discussed the results and commented on the manuscript.