Highly ordered titania nanotube arrays for phosphoproteomics

Abstract

Titania nanotube arrays are useful for phosphopeptide enrichment and separation. These highly ordered titania nanotube arrays are a low cost and highly effective alternative to the use of liquid chromatography mass spectrometry (LC-MS) methods using meoporous titania beads or particles. The highly ordered TiO.sub.2 nanotubes are grown on surfaces coated with Ti metal, or preferably on Ti wires, by methods that preferably include anodic oxidation.

Claims

1. A device for isolation of phosphopeptides in a sample, comprising: an ordered TiO.sub.2 nanotube array comprising ordered TiO.sub.2 nanotubes grown on a Ti surface, wherein the nanotubes point outward from the Ti surface, wherein the Ti surface is a Ti wire, and wherein the nanotubes point radially outward from the Ti wire; and a container for phosphopeptide isolation, wherein the ordered TiO.sub.2 nanotube array is located within the container, and wherein the sample is placed in the container to contact the ordered TiO.sub.2 nanotube array.

2. The device of claim 1, wherein the ordered TiO.sub.2 nanotubes have a length of about 100 nm to about 500 m and a pore diameter of about 10 nm to about 400 nm.

3. The device of claim 1, wherein the ordered TiO.sub.2 nanotubes have a length of about 10 to about 20 m, a pore diameter of about 110 nm, and a wall thickness of about 20 nm.

4. The device of claim 1, wherein the Ti wire has a diameter of about 0.01 mm to about 1 mm.

5. The device of claim 1, wherein the Ti wire has a diameter of about 0.25 mm.

6. The device of claim 1, wherein the ordered TiO.sub.2 nanotube array is immobilized on the Ti surface.

7. A method for isolation of phosphopeptides in a sample, comprising: exposing a sample containing phosphopeptides to the device of claim 1, wherein the sample is placed in the container and wherein the sample contacts the ordered TiO.sub.2 nanotube array to produce bound phosphopeptides attached to the ordered TiO.sub.2 nanotube array; and releasing the bound phosphopeptides from the ordered TiO.sub.2 nanotube array to produce isolated phosphopeptides.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a SEM image of a commercially available mesoporous TiO.sub.2 bulk material bead.

(2) FIG. 2 shows a SEM image of a tilted surface view of an example of a TiO.sub.2 nanotube array.

(3) FIG. 3 shows a schematic of bidentate interaction of phosphopeptide with TiO.sub.2.

(4) FIG. 4 shows (A) a cross-sectional SEM image of an as-anodized Ti wire sample showing a titania nanotube layer (having a thickness of about 15 m) covering the titanium wire (having a diameter of about 0.25 mm) and (B) a schematic representation of the orientation of the nanotubes on the surface of the wire substrate.

(5) FIG. 5 shows SEM images of (A) a side view of a Ti wire covered with TiO.sub.2 nanotubes, with a higher magnification shown in the inset, (B) a side view of nanotubes (having a length of about 10 m) attached to the Ti wire substrate, (C) a magnified lateral view of the nanotubes, and (D) a view of the top surface of the nanotubes, with a view of the bottom surface of the nanotubes detached from the wire substrate shown in the inset.

(6) FIG. 6 shows SEM images of nanotubes and a rift caused by stress at (A) a magnification of 78.73 K X and (B) a magnification of 50.00 K X.

(7) FIG. 7 shows (A) a high resolution transmission electron microscope (HRTEM) image of a nanotube grown on Ti wire, (B) a HRTEM image of the polycrystalline lattice, and (C) selected area diffraction pattern (SAD) from a few nanotubes showing reflections from anatase phase of titania.

(8) FIG. 8 shows a glancing angle X-ray diffractometer (GAXRD) pattern obtained from a single wire coated with nanotubes.

(9) FIG. 9 shows XPS survey spectrum from a titania nanotube film covered Ti wire, with insets showing high resolution scans of the O 1s and Ti 2p peaks.

(10) FIG. 10 shows a schematic of an experimental workflow for Matrix assisted laser desorption ionization mass spectrometry (MALDI-MS) analysis of tryptic peptides of a phosphoprotein standard, -casein, 500 g was digested using trypsin, to compare phosphopeptide separation of commercially available bulk material beads and examples of a Ti-wire surface grown with TiO.sub.2 nanotubes.

(11) FIG. 11 shows (A) MALDI-mass spectrum of -casein tryptic peptides, between m/z range 1500-2100, indicating the prominent peptide, 8-HQGLPQEVLNENLLR-22 (SEQ ID NO: 1) at m/z 1760.05, less prominent phosphorylated peptide, 104-YKVPQLEIVPN(pS)AEER-119 (SEQ ID NO: 2) at m/z 1952.07, and infinitesimally small response for the phosphorylated peptide 106-VPQLEIVPN(pS)AEER-119 (SEQ ID NO: 3) at m/z 1660.94; and MALDI-mass spectrum of -casein tryptic peptides, between m/z range 1500-2100, after phosphopeptide separation using (B) commercially available bulk material beads and (C) an example Ti-wire surface grown with TiO.sub.2 nanotubes.

(12) FIG. 12 shows a schematic of an experimental workflow for liquid chromatography-mass spectrometry/tandem mass spectrometry (LC-MS/MS) analysis and identification of mouse liver proteome tryptic peptides to compare phosphopeptide separation capacity of commercially available bulk material beads (Beads) and examples of Ti-wire surface grown with TiO.sub.2 nanotubes (Wires).

(13) FIG. 13 shows bar-graphs representing, with unique and high confidence (>95%): (I) average number of phosphopeptides, (II) average number of phosphoproteins, (III) average number of Ser-phosphopeptides, and (IV) average number of Thr-phosphopeptides.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

(14) The present disclosure relates to titania nanotube arrays useful for phosphopeptide enrichment and separation. These highly ordered titania nanotube arrays are a low cost and highly effective alternative to the use of liquid chromatography mass spectrometry (LC-MS) methods using meoporous titania beads or particles. These beads or particles are expensive and their irregular pore structure offers very limited opportunities for surface manipulation for any further improvement in performance.

(15) In a preferred embodiment, radially aligned nanotubes were grown by anodic oxidation of titanium wires and their performance was compared to widely used commercially available bulk mesoporous titania beads. Peptides generated from a standard phosphoprotein, -casein, as well as mouse liver complex tissue extracts were used for the comparison. Example titania nanotubes of length about 10 to about 20 m, with a pore diameter of about 110 nm and a wall thickness of about 20 nm, demonstrated their capacity to perform on par with the commercially available beads, with further indications that the nanotubes having optimum dimensions could outperform the commercially available phosphopeptide enrichment materials. However, other titania nanotubes having other lengths, pore diameters, and wall thicknesses may also be used, since the optimum dimensions for maximum separation capacity may be outside the ranges recited above. For example, the lengths of the titania nanotubes may range from about 100 nm to about 500 m, and the pore diameter may range from about 10 nm to about 400 nm.

(16) Thus, the highly porous nature of the commercially available bulk material, which has been effective for phosphopeptide separation, is also achieved in preferred embodiments of the highly ordered TiO.sub.2 nanotubes grown on Ti metal surfaces. In preferred examples, TiO.sub.2 nanotubes coated on Ti-wires were tested for their capacity for phosphopeptide enrichment, and compared to the phosphopeptide separation capacity of the commercially available bulk material beads as a reference. Ti-wires were chosen for nanotube growth because their quantity and effective surface area for phosphopeptide binding could be simply and appropriately tuned based on the wire length. Ideally, controlled/optimized amounts of TiO.sub.2-beads for improved phosphopeptide separation and detection should be used. Thus, instead of controlling the weight of small sample amounts of bulk material beads, the length of Ti-wires bearing titania nanotubes are readily optimized for phosphopeptide separation. Furthermore, the nanotubes coated on Ti wires are immobile, so they do not mix into the solution, hence providing the added benefit of avoiding practical difficulties in separating beads from solvent media. The nanotube array can be immobilized on any suitable solid surface, or it may be utilized without any surface immobilization as a self-stranding membrane. In preferred embodiments, the Ti wire has a diameter from about 0.01 mm to about 1 mm. In additional preferred embodiments, the Ti wire has a diameter of about 0.25 mm.

(17) In additional preferred embodiments, the titania nanotubes are not grown on TI wires. Rather, they are grown inside a container such as a vial. The inside of the container is first coated with titanium, then the titanium is anodized to form the TiO.sub.2 nanotubes. Phosphorylated proteins placed inside the container would then attach to the inside of the container.

(18) Experiments demonstrate that the highly ordered titania nanotube arrays grown on Ti surfaces are highly suitable for isolating phosphopeptides. The arrays perform at an appreciable level as compared to bulk material beads while also possessing other desirable attributes. Importantly, the nanotube dimensions can be further varied in length and diameter, thus one can precisely tune the parameters to further optimize their functionality. In addition, the nanotube-on-wire geometry of preferred embodiments facilitates the use of length of the wire as a way to easily assess the surface area for phosphopeptide separation experiments and thus eliminate the need for weighing precisely very small amounts of enrichment material or the variability associated with using slurry suspensions of material as in the case of beads. Furthermore, the nanotube arrays may be made available at a much lower cost than the present commercially available materials. The TNTs on Ti wire embodiments are shown to offer similar efficacy for phosphopeptide enrichment compared to the current best approach, while at the same time offering an enhanced ease-of-use. It should be noted that titania nanotubes are not the only nanotubes expected to show these advantages. For example, nanotubes made of oxides of alloys of titanium and zinc oxide may also be useful.

EXAMPLE 1

Materials, Preparation, and Characterization of TiO2 Nanotubes on Ti Wire

(19) Guanidine:hydrogen chloride (GHCl), ammonium bicarbonate (NH4HCO3), Phosphatase inhibitor cocktail 2 (cat. no. P5726), dithiothreitol (DTT), iodoacetamide (IAA), formic acid (HCOOH, FA), triflouroacetic acid (TFA), -casein from bovine milk (cat. no. C6780, as-casein minimum 70%), acetonitrile (CHROMASOLV, for HPLC, gradient-grade, >99.9%), water (CHROMASOLV-Plus, for HPLC), -cyano-4-hydroxycinnamic acid (CHCA), and glycolic acid were purchased from Sigma-Aldrich (St. Louis, Mo.). For proteomics sample preparation work-flow, deionized water was obtained from an in-house Milli-Q system (Millipore, Bedford, Mass.). All centrifugation steps were completed in an IEC Micromax RF microfuge at 14,600 RCF (Relative Centrifugal Force). Modified trypsin was obtained from Promega (Madison, Wis.). Oligo R3 reversed-phase material was obtained from Applied Biosystems (Foster City, Calif.). For the packing of Oligo R3 reversed phase material, Bio-select extraction columns (reversed phase C4) were obtained from GRACE-VYDAC (W.R. Grace & Co., Deerfield, Ill.). For TiO.sub.2-chromatography using beads, Titansphere TiO.sub.2-beads were obtained from GL Sciences Inc. For TiO.sub.2-chromatography using wires, titanium wire of diameter 0.25 mm (99.7% pure) purchased from Sigma Aldrich was used. The electrolyte for anodization of the wire consisted of ammonium fluoride (ACS reagent, 98%, Sigma Aldrich) and ethylene glycol (anhydrous, 99.8%, Sigma Aldrich) and deionized water.

(20) Ammonium hydroxide (trace-metal grade, assay: 20-22% as NH.sub.3) was obtained from Fisher Scientific (Hampton, N.J.) for phosphopeptide elution during TiO.sub.2-chromatography. Whole mouse liver samples were dounce homogenized in the presence of both protease and phosphatase inhibitor cocktails (Jarrold et al. 2005). Protein concentrations were also determined using NI (Non-Interfering) Protein Assay-Kit purchased from G-BIOSCIENCES.

(21) The Ti wire (diameter 0.25 mm) was cut into 25 mm length and degreased by sonication in acetone and then in isopropanol. The degreased wires were again cleaned sequentially in water and Micro-90, isopropanol and acetone and dried with nitrogen gas. Anodization was conducted at room temperature in an electrolyte consisting of 0.3 wt % NH.sub.4F, 2 vol % H.sub.2O in Ethylene Glycol. The titanium wire was used as anode and platinum foil as cathode. The anodization was performed for 4 h with 60 V applied between the electrodes. The anodized wires were washed and sonicated in isopropanol to remove debris formed on the surface of the nanotubes during the anodization process. The cleaned samples were annealed at 530 C. in oxygen for 3 h (Varghese et al. 2003).

(22) The morphology of the nanotubes on Ti wire was studied using a field emission scanning electron microscope (FESEM; LEO 15125). The crystal structure was identified using a high resolution transmission electron microscope (HRTEM; JEOL 2010) and glancing angle x-ray diffractometer (GAXRD; Rigaku, Smartlab, Cu K-alpha). An array of three nanotube coated wire was used for GAXRD measurements. The incident angle was 0.5. The x-ray photoelectron spectroscopy (XPS; Physical Electronics, model 5700) was used to determine the composition of the samples.

(23) In order to understand the phosphopeptide separation efficiency of the titania nanotube arrays relative to the most utilized material in the field, TiO.sub.2-chromatography was performed in parallel using, (1) the most popularly used and commercially available Titansphere TiO.sub.2 Bulk Material-beads as a reference material (Beads) and, 2) Ti-wire pieces grown with TiO.sub.2 nanotubes (Wires), on a standard phosphoprotein, -casein and also on mouse liver lysates, to illustrate their phosphopeptide separation capacity, and their applicability in studying complex biological proteomes, respectively. Surfaces of Ti wire pieces were modified with TiO.sub.2-nanotubes (see FIG. 4), and the wires were particularly chosen so that they could be readily cut into pieces so that effective surface area can be simply be controlled by specific lengths.

(24) Anodization of titanium wires per the conditions given above resulted in the growth of highly ordered nanotubes pointing radially outward from the surface. The SEM image of the cross section of a nanotube-covered Ti wire sample (diameter 0.25 mm) is given in FIG. 4(A) and an expanded view of the nanotubes on the wire surface is schematically represented in FIG. 4(B). The growth and self-organization of the nanotubes take place during the anodization process and, without wanting to be bound by theory, are believed to be due to the interplay between electric field assisted oxidation of titanium metal and chemical dissolution as well as electric field assisted dissolution of the oxide. An alternate model suggests a stress induced displacement of material rather than field assisted dissolution as responsible for the anodic growth of porous structures. Nevertheless, the nanotubes are grown more or less perpendicular to the substrate surface by consuming the metal and converting into oxide regardless of the growth conditions.

(25) The anodization of titanium in organic electrolytes such as ethylene glycol generally produces anodization debris in the form of particles or nanowires or bunched/broken tubes on the surface of the sample. To remove the debris from the surface of the nanotubes, the wire samples were subjected to ultrasonic agitation at 35 kHz as used normally for other substrates. However, the nanotube films peeled during sonication due to the stress at the oxide/metal interface. The problem was eliminated by performing the ultrasonication at 130 kHz at a reduced power level for 1 to 2 hours. The resulting films were heat treated in oxygen ambient at 530 C. for stoichiometric TiO.sub.2 formation and crystallization (Varghese et al. 2003).

(26) The low magnification SEM images of a heat treated nanotube coated wire sample are shown in FIG. 5(A) and the high magnification images of the nanotubes are given in FIGS. 5(B)-5(D). The nanotube coating appeared like the bark of a tree (see FIG. 5(A) inset) with crevices formed between groups of nanotubes. Studies showed that the crevices were primarily formed during the growth of the nanotubes and not during heat treatment. The images in FIGS. 6(A) and 6(B) give a closer view of the rift between nanotubes. The nanotubes grown using ethylene glycol electrolyte in planar substrates have a more or less hexagonal close packed geometry. Such a configuration cannot be maintained in a cylindrical substrate geometry especially when the radius is small. The lateral stress across the nanotube film increases as the nanotube length increases and as a result partition is formed between nanotubes to relieve the stress. These observations and conclusions are consistent with those previously reported in which cracks were observed in TNT films on wire substrates. Nonetheless, the nanotubes were not detached from the substrate, rather their alignment changed in the partition region. The nanotubes between the crevices were closely packed as evident from the side and top views of the nanotubes given in FIG. 5(B)-5(D). The close packing is maintained from the bottom of the nanotube (see FIG. 5(D) inset) to the top. Nanotubes of length about 10 to 20 m, pore diameter 110 nm and wall thickness 20 nm (measured from SEM images) were used for the study.

(27) In order to understand the structure and composition of the heat treated nanotubes on wire substrates, HRTEM, GAXRD and XPS studies were performed. FIG. 7(A) shows the HRTEM images of a nanotube mechanically separated from the wire substrate. The high resolution image of a nanotube region given in FIG. 7(B) shows the polycrystalline lattice. The selected area diffraction pattern (SAD) from a few nanotubes (FIG. 7(C)) revealed reflections from anatase phase of titania. GAXRD patterns obtained from a sample of three nanotube coated wires (FIG. 8) confirmed that the nanotubes consisted of anatase phase of titania. No other phase was found in the samples. This result was further substantiated by XPS studies. The XPS survey spectrum given in FIG. 9 shows the peaks of titanium, oxygen and carbon only. The high resolution scans of O1s and Ti2P peaks are given in the inset. The carbon peak is centered at 288 eV, which can be attributed to CO bonds. It is likely that this peak arises from the carbon left on the surface after the burning of the organic electrolyte during heat treatment in oxygen ambient. The stoichiometry of the nanotubes was not estimated as XPS does not provide the composition accurately for nanoporous samples. Nonetheless, significant deviation is not expected from stoichiometry as the pristine oxide samples were heat treated in oxygen ambient. In short, the nanotubes composed of pure titanium dioxide in anatase phase.

EXAMPLE 2

Trypsin Digestion, Desalting of Tryptic Peptides, and Separation of Phosphopeptides using TiO2-Chromatography

(28) As reported previously (Wijeratne et al. 2013), 500 g aliquots of protein were precipitated with 8 volumes of cold acetone (20 C.) in 1.5 mL Eppendorf tubes. After centrifugation (14,600 RCF, 5 min), supernatants were discarded and pellets were washed three times using 20 C. acetone (100 L for each wash). Sample tubes were then kept open in a fume-hood for 2 min to ensure any residual acetone vaporization. The pellets were reconstituted in 3 M Guanidine:HCl in 100 mM NH.sub.4HCO.sub.3 (90 L) containing phosphatase inhibitor cocktail (2 L). The solutions were subsequently reduced with DTT (1 mM final concentration, incubated at 37 C., for 45 min) and then alkylated with iodoacetamide (5.5 mM final concentration, incubated at 37 C., for 30 min). The solutions were finally diluted with ddH.sub.2O to 1 mL before trypsin-based digestion. 100 g of modified trypsin was dissolved in 300 L of 0.1 M NH.sub.4HCO.sub.3, and 10 g aliquots were added into each 500 g protein sample (i.e. 1:50 weight ratio). Samples were then incubated overnight at 37 C., and the digestion was quenched by adding 20 L of formic acid (to bring the pH of solutions to less than 5). After centrifugation, the supernatants were recovered for further processing. Similarly, 500 g of bovine -casein was subjected to trypsin digestion for qualitative comparison of phosphopeptide separation using Titansphere TiO.sub.2 Bulk Material-beads (beads) and Ti-wire surface grown with TiO.sub.2 nanotubes (wires).

(29) Oligo R3 reversed-phase material was dispersed in ACN/H.sub.2O/TFA 70/29.9/0.1 (v/v/v) to make a 60 mg/mL slurry as previously described (Wijeratne, et al. 2013; Thingholm, et al. 2008), and divided into 500 L aliquots each containing 30 mg of Oligo R3 beads in 1.5 mL Eppendorf tubes. The beads prepared for peptide desalting by sequential vortex, spin and removal of the supernatant followed by two wash steps using 200 L of 0.1% TFA in MilliQH.sub.2O. Peptide solutions were added onto washed Oligo R3 beads and incubated for 30 min at room-temperature using end-over-end rotation. GRACE-VYDAC BIOSELECT-C4 columns (CAT. NO. 214SPE1000) were adapted onto an extraction manifold (Waters Manifold, Mass., USA), washed sequentially with 1) ACN (500 L), 2) ACN/TFA/H.sub.2O 70/0.1/29.9 (v/v/v, 200 L), 3) 0.1% TFA (500 L) and 4) dd H.sub.2O (500 L), and then packed with peptide-bound Oligo R3 beads by a gentle application of vacuum into the extraction manifold vacuum chamber. Subsequently, the peptide-bound beads were washed with 500 L ddH.sub.2O and eluted by sequentially passing 200 L of ACN/TFA/H.sub.2O 90/0.1/9.9 (v/v/v) for one time and then 200 L of ACN/TFA/H.sub.2O 70/0.1/29.9 (v/v/v) for two times. All elution fractions were collected into 1.5 mL Eppendorf tubes. Prior to TiO.sub.2-chromatography, these elution fractions were subjected to vacuum centrifugation for complete dryness.

(30) Phosphopeptide separation of the peptide mixtures was carried out using an optimized strategy adapted from previous reports ((Wijeratne, et al. 2013; Thingholm, et al. 2008; Li, et al. 2009), and was performed in triplicate using identical protein samples for each TiO.sub.2-chromatographic method. In using the beads for phosphopeptide separation, briefly for each replicate, Titansphere TiO.sub.2 beads were dispersed in ACN/H.sub.2O/TFA 80/15/5 (v/v/v) to make a 100 g/L slurry and then divided into 5 L aliquots each containing 500 g of TiO.sub.2 beads in 0.5 mL Eppendorf tubes. Each vial was subjected to a vortex and spin with supernatants discarded, followed by 2 additional wash steps using 200 L of 0.1% TFA in MilliQH.sub.2O. Dried peptide mixtures (500 g) were reconstituted in 200 L of 1 M glycolic acid in ACN/H.sub.2O/TFA 80/15/5 (v/v/v), and loaded onto the pre-washed Titansphere TiO.sub.2-beads. The peptides were allowed to interact with the TiO.sub.2 for 30 min at room temperature using end-over-end rotation. The TiO.sub.2 beads were then sequentially washed with 400 L ACN/H.sub.2O/TFA 80/15/5 (v/v/v) with a spin and removal of the supernatant followed by an additional 400 L wash with the same solvent. Finally the phosphopeptides captured on the TiO.sub.2-beads were eluted 1 time with 200 L of 5% NH.sub.4OH. In using the wires for phosphopeptide separation, briefly for each replicate, Ti-wire pieces (40.5 cm, i.e. 2.0 cm in length) grown with TiO.sub.2 nanotubes were placed inside 0.5 mL Eppendorf tubes and subjected to similar washing steps. Following reconstitution of dried peptide mixtures (500 g) in 200 L of 1 M glycolic acid in ACN/H.sub.2O/TFA 80/15/5 (v/v/v), peptide mixtures were loaded onto the pre-washed Ti-wire pieces grown with TiO.sub.2 nanotubes. TiO.sub.2 wires with loaded peptides were then sequentially washed with 400 L ACN/H20/TFA 80/15/5 (v/v/v) with a spin and removal of the supernatant followed by an additional 400 L wash with the same solvent. Finally the phosphopeptides captured on the TiO.sub.2-wires were also eluted 1 time with 200 L of 5% NH.sub.4OH. The NH.sub.4OH elution fractions were dried by vacuum centrifugation prior to nanoLC-MS/MS analysis. For the standard phosphopeptide mixture obtained from -casein trypsin digestion, 2.5 L aliquots of the elution fractions were removed from each sample, desalted by ZipTip(C-18) as described by the manufacturer (Millipore) and evaluated by Matrix-assisted Laser Desorption IonizationTime of FlightMass Spectrometry (MALDI-TOF-MS) to qualitatively investigate phosphopeptide separation.

(31) MALDI-MS analysis was performed on a 4800 MALDI-TOF/TOF instrument (AB Sciex, Foster city, Calif.). Mass spectra were obtained in positive ion reflector mode. MALDI-matrix solution was prepared by dissolving -cyano-4-hydroxy-cinnamic acid (CHCA, 5 mg) in 10 mM ammonium phosphate (monobasic) in ACN/FA/H.sub.2O 60/0.1/39.9 (v/v/v, 1 mL). In order to perform MALDI-MS analyses, desalted (using Oligo R3 reversed phase material or ZipTip (C-18 )) and isolated peptides in solution (0.5 L) were mixed with MALDI-matrix solution (1 L), and spots were placed on a calibrated MALDI plate.

(32) For the qualitative evaluation of phosphopeptide separation capacity of highly ordered TiO.sub.2 nanotubes on Ti-Metal wires prepared, phosphopeptide separation experiments were first performed using a standard phosphoprotein, -casein. As illustrated in FIG. 10, tryptic digest from 500 g of -casein was first separated into two identical aliquots, i.e. 250 g each, and subjected to OligoR3 based desalting as described in the methods. Prior to performing TiO.sub.2-chromatography, for comparison purposes, a MALDI-mass spectrum (or an MS finger print) was obtained for desalted -casein tryptic digest (FIG. 11(A)). During the TiO.sub.2-chromatography step, the desalted tryptic digests were pretreated with both beads and wires. After washing the captured phosphopeptides and eluting them with ammonia solutions as described in methods, MALDI-mass spectra (or MS fingerprints) were obtained (FIGS. 11(B) from beads and 11(C) from wires). In analyzing the mass spectral regions (m/z 1500 to 2100), after performing TiO.sub.2-chromatography either using beads or wires on -casein tryptic digests, the phosphopeptides, 106-VPQLEIVPN(pS)AEER-119 (SEQ ID NO: 3) and 104-YKVPQLEIVPN(pS)AEER-119 (SEQ ID NO: 2) represented by m/z values at 1661.01 and 1952.20, are markedly abundant from both TiO.sub.2-treatments (FIGS. 11(B) and 11(C)) compare to before TiO.sub.2 treatment (FIG. 11(A)). In FIG. 11(A), the most abundant peptide response for the -casein tryptic peptide before TiO.sub.2-treatment is, 8-HQGLPQEVLNENLLR-22 (SEQ ID NO: 1) (m/z 1760.05). The responses at m/z 1660.94 and 1952.07 are either infinitesimally small or relatively low in abundance, respectively. This qualitative illustration demonstrates that TiO.sub.2 nanotubes, grown on pieces of Ti-metal wire can be used to enrich phosphopeptides in a comparable manner to TiO.sub.2 beads.

EXAMPLE 3

Phosphopeptide Separation Capacity for Phosphoproteomes of Complex Tissue Samples

(33) In real biological protein samples or proteomes, it is known that phosphorylation is sub-stoichiometric or very low in abundance. Thus, separation of phosphopeptides from a purified standard phosphoprotein sample like the test case with -casein may have different dynamics to that of a complex digestion derived from a tissue extract. Hence, the capacity of the highly ordered TiO.sub.2 nanotubes in phosphopeptide separation for studying complex phosphoproteomes derived from tissue extracts was also compared with respect to the widely used Titansphere TiO.sub.2 Bulk Material-beads.

(34) Nano-LC-MS/MS analyses were performed on a TripleTOF 5600 (ABSciex, Toronto, ON, Canada) coupled to an Eksigent (Dublin, Calif.) nanoLC.ultra nanoflow system. Dried phosphopeptide samples were reconstituted in FA/H.sub.2O 0.1/99.9 (v/v,) and loaded onto IntegraFrit Trap Column (outer diameter of 360 m, inner diameter of 100, and 25 m packed bed) from New Objective, Inc. (Woburn, Mass.) at 2 l/min in FA/H20 0.4/99.2 (v/v) for 10 min to desalt and concentrate the samples. For the chromatographic separation of peptides, the trap-column was switched to align with the analytical column, Acclaim PepMap100 (inner diameter of 75 m, length of 15 cm, C18 particle sizes of 3 m and pore sizes of 100 ) from Dionex-Thermo Fisher Scientific (Sunnyvale, Calif.). The peptides were eluted using a varying mobile phase (MP) gradient from 95% phase A (FA/H.sub.2O 0.4/99.6, v/v) to 40% phase B (FA/ACN 0.4/99.6, v/v) for 70 min, from 40% phase B to 85% phase B for 5 min and then keeping the same MP-composition for 5 more minutes at 300 nL/min.

(35) Nano-LC mobile phase was introduced into the mass spectrometer using a NANOSpray III Source (AB Sciex, Toronto, On, Canada). Ion source gas 1 (GS1) was zero grade air while ion source gas 2 (GS2) and curtain gas (CUR) were both nitrogen. The gas settings were kept at 7, 0 and 25 respectively in vendor specified arbitrary units. Interface heater temperature and ion spray voltage was kept at 150 C., and at 2.3 kV. The mass spectrometer method was operated in positive ion mode set to go through 4156 cycles for 90 minutes, where each cycle consisted of one TOF-MS scan (0.25 s accumulation time, in a 400 to 1600 m/z window) followed by twenty information dependent acquisition (IDA) mode MS/MS-scans on the most intense candidate ions selected from initially performed TOF-MS scan during each cycle, having a minimum of 150 counts. Each product ion scan was operated under vender specified high-sensitivity mode with an accumulation time of 0.05 secs and a mass tolerance of 50 mDa. Former MS/MS-subjected candidate ions were excluded for 10 s after its first occurrence, and data were recorded using Analyst-TF (1.5.1) software.

(36) The nano-LC-MS/MS data (*.wiff file) from the enriched phosphopeptides were analyzed for peptide/protein identification using ProteinPilot software (version 4.2, revision 1297) that integrates the Paragon algorithm, searched against a SwissProt database of Mus Musculus protein sequences on a local 12-processor server. A custom sample-type was selected that specifies variable biological modifications as specified defaults in the ProteinPilot software. The vendor defined phosphorylation emphasis on serine/threonine/tyrosine was also used as a special factor. The resulting *.group files were then used to generate a spreadsheet as a peptide summary report. Only those phosphopeptides identified with a minimum of 95% confidence in identity (calculated by probability algorithms of ProteinPilot software), and phosphorylation as a modification were selected as viable phosphopeptide identifications. Unique phosphopeptides were then selected based on sequence, modifications, and mass to charge ratio (m/z-value) using available software tools on Microsoft Excel. Tools made available by Microsoft Excel were used to determine the number of unique phosphopeptides for each replicate and each TiO.sub.2-chromatography method employed.

(37) In order to examine the capability of TiO.sub.2 nanotubes for phosphopeptide separation in proteomes of complex samples, it was hypothesized that the number of high confidence (>95%) unique phosphopeptides identified using LC-MS/MS and database search algorithms is representative of the phosphopeptide separation capacity in a single phosphopeptide isolation. Hence, an experimental workflow, as depicted in FIG. 12, was implemented and analyzed for unique phosphopeptide identifications. More specifically, from a single pool of mouse liver lysate, as described in the methods, six identical protein solutions (500 g each) were sequentially subjected for protein precipitation, trypsin digestion, OligoR3 based desalting and phosphopeptide separation using TiO.sub.2-chromatography. In performing TiO.sub.2-chromatography with beads, 500 g equivalents of Titansphere TiO Bulk Material-beads were used. For the wires, 2 cm (0.5 cm4) of Ti-wire pieces grown with highly ordered TiO.sub.2 nanotubes were used. The surface of area of Titansphere TiO Bulk Material-beads given by the manufacturer is 100 m.sup.2/g and therefore, the surface area of 500 g beads is about 0.05 m.sup.2. The surface area of the nanotubes (pore diameter 110 nm, wall thickness 20 nm and length about 15 m) on a 2 cm long wire of diameter 0.25 mm was calculated using a method reported elsewhere (Varghese, et al. 2009; Shankar, et al. 2007) and is about 0.01 m.sup.2. Thus, the Titansphere TiO beads used were having surface area about 5 times the surface area of nanotubes. For comparison purposes, triplicate of 1 g equivalent tryptic digests from mouse liver proteome, were also analyzed using LC-MS/MS, so that non-phosphopeptide separation was compared to phosphopeptide separation by either beads or wire based TiO.sub.2-chromatography methods. It was assumed that 1 g equivalent of tryptic digest analysis in LC-MS from a mouse liver could be used to illustrate the sub-stoichiometric nature of phosphopeptides, and comparison of phosphopeptides identified in such an analysis is illustrative of the separation capacity in using either TiO.sub.2-chromatography methodologies. Nevertheless, following all LC-MS/MS analyses/runs of each sample, as described in the methods, each LC-MS/MS run was subjected to database searches to identify high confidence (>95%) unique phosphopeptides. The confidence values were assigned for phosphopeptide identifications from the Paragon probabilistic algorithms used in ProteinPilot search software.

(38) The average number of high confidence phosphopeptides (Paragon plus algorithm assigned confidence >95%) identified and their average number of representative phosphoproteins were presented as bar-graphs, FIG. 13 (I & II) for each beads and wire TiO.sub.2-chromatography. In each graph represented in FIG. 13, 1No Separation represents the number of phosphopeptides identified in analyzing 1 g equivalent of tryptic digest using LC-MS/MS with no TiO.sub.2-chromatography performed, 2Separation with Beads, and 3Separation with Wires represent number of phosphopeptides identified in analyzing 500 g equivalent of tryptic digest using LC-MS/MS with TiO.sub.2-chromatography performed using, Titansphere TiO.sub.2 Bulk Material-beads and Ti-wire surface grown with TiO.sub.2 nanotubes, respectively. In inspecting FIG. 13 (I & II), it is apparent that, although the surface area is lower, wires carry appreciable capacity (average number of phosphopeptides detected, 486.678.41) to isolate phosphopeptides in reference to that observed when spherical beads are utilized (average number of phosphopeptides detected, 541.0012.50). In performing TiO.sub.2-chromatography for phosphoproteomics studies, it has been well understood that phosphopeptides identified are phosphorylated mostly at serine (Ser or S) and then at threonine (Thr or T) amino acid residues. Thus, the average unique number of high confidence S-phosphopeptides and T-phosphopeptides identified were also plotted as bar-graphsFIG. 13 (III & IV) for evaluation. It is evident that the wires carry a more or less similar capacity to isolate Ser-phosphorylated peptides (390.333.67) and Thr-phosphorylated peptides (18.651.66), compared to peptides identified with Ser-phosphorylation (460.6711.86) and Thr-phosphorylation (13.882.18) when beads are employed. Overall, these results demonstrate that the highly ordered TiO.sub.2 nano tubes immobilization on Ti-metal wires carry significant potential as an alternate medium that can be utilized for mass spectrometry based phosphoproteomics workflows.