The scientific exploration of saliva in the post-proteomic era: from database back to basic function

The proteome of human saliva can be considered as being essentially completed. Diagnostic markers for a number of diseases have been identified among salivary proteins and peptides, taking advantage of saliva as an easy-to-obtain biological fluid. Yet, the majority of disease markers identified so far are serum components and not intrinsic proteins produced by the salivary glands. Furthermore, despite the fact that saliva is essential for protecting the oral integuments and dentition, little progress has been made in finding risk predictors in the salivary proteome for dental caries or periodontal disease. Since salivary proteins, and in particular the attached glycans, play an important role in interactions with the microbial world, the salivary glycoproteome and other post-translational modifications of salivary proteins need to be studied. Risk markers for microbial diseases, including dental caries, are likely to be discovered among the highly glycosylated major protein species in saliva. This review will attempt to raise new ideas and also point to under-researched areas that may hold promise for future applicability in oral diagnostics and prediction of oral disease.     

Proteomic analysis of gingival crevicular fluid for discovery of novel periodontal disease markers

The protein composition of gingival crevicular fluid (GCF) may reflect the pathophysiology of periodontal diseases. A standard GCF proteomic pattern of healthy individuals would serve as a reference to identify biomarkers of periodontal diseases by proteome analyses. However, protein profiles of GCF obtained from apparently healthy individuals have not been well explored. As a step toward detection of proteomic biomarkers for periodontal diseases, we applied both gel-based and gel-free methods to analyze GCF obtained from healthy subjects as compared with supragingival saliva. To ensure optimized protein extraction from GCF, a novel protocol was developed. The proteins in GCF were extracted with high yield by urea buffer combined with ultrafiltration and the intensity of spots with supragingival saliva and GCF was compared using agarose two-dimensional electrophoresis. Eight protein spots were found to be significantly more intense in GCF. They included superoxide dismutase 1 (SOD1), apolipoprotein A-I (ApoA-I), and dermcidin (DCD). Moreover, GCF proteins from healthy subjects were broken down into small peptide fragments and then analyzed directly by LC-MS/MS analysis. A total of 327 proteins including ApoA-I, SOD1, and DCD were identified in GCF. These results may serve as reference for future proteomic studies searching for GCF biomarkers of periodontal diseases.     

Capillary electrophoresis mass spectrometry-based saliva metabolomics identified oral,breast and pancreatic cancer-specific profiles

Saliva is a readily accessible and informative biofluid, making it ideal for the early detection of a wide range of diseases including cardiovascular, renal, and autoimmune diseases, viral and bacterial infections and, importantly, cancers. Saliva-based diagnostics, particularly those based on metabolomics technology, are emerging and offer a promising clinical strategy, characterizing the association between salivary analytes and a particular disease. Here, we conducted a comprehensive metabolite analysis of saliva samples obtained from 215 individuals (69 oral, 18 pancreatic and 30 breast cancer patients, 11 periodontal disease patients and 87 healthy controls) using capillary electrophoresis time-of-flight mass spectrometry (CE-TOF-MS). We identified 57 principal metabolites that can be used to accurately predict the probability of being affected by each individual disease. Although small but significant correlations were found between the known patient characteristics and the quantified metabolites, the profiles manifested relatively higher concentrations of most of the metabolites detected in all three cancers in comparison with those in people with periodontal disease and control subjects. This suggests that cancer-specific signatures are embedded in saliva metabolites. Multiple logistic regression models yielded high area under the receiver-operating characteristic curves (AUCs) to discriminate healthy controls from each disease. The AUCs were 0.865 for oral cancer, 0.973 for breast cancer, 0.993 for pancreatic cancer, and 0.969 for periodontal diseases. The accuracy of the models was also high, with cross-validation AUCs of 0.810, 0.881, 0.994, and 0.954, respectively. Quantitative information for these 57 metabolites and their combinations enable us to predict disease susceptibility. These metabolites are promising biomarkers for medical screening.     

Bright Fluorescence Monitoring System Utilizing Zoanthus sp. Green Fluorescent Protein (ZsGreen) for Human G-Protein-Coupled Receptor Signaling in Microbial Yeast Cells

G-protein-coupled receptors (GPCRs) are currently the most important pharmaceutical targets for drug discovery because they regulate a wide variety of physiological processes. Consequently, simple and convenient detection systems for ligands that regulate the function of GPCR have attracted attention as powerful tools for new drug development. We previously developed a yeast-based fluorescence reporter ligand detection system using flow cytometry. However, using this conventional detection system, fluorescence from a cell expressing GFP and responding to a ligand is weak, making detection of these cells by fluorescence microscopy difficult. We here report improvements to the conventional yeast fluorescence reporter assay system resulting in the development of a new highly-sensitive fluorescence reporter assay system with extremely bright fluorescence and high signal-to-noise (S/N) ratio. This new system allowed the easy detection of GPCR signaling in yeast using fluorescence microscopy. Somatostatin receptor and neurotensin receptor (implicated in Alzheimer’s disease and Parkinson’s disease, respectively) were chosen as human GPCR(s). The facile detection of binding to these receptors by cognate peptide ligands was demonstrated. In addition, we established a highly sensitive ligand detection system using yeast cell surface display technology that is applicable to peptide screening, and demonstrate that the display of various peptide analogs of neurotensin can activate signaling through the neurotensin receptor in yeast cells. Our system could be useful for identifying lead peptides with agonistic activity towards targeted human GPCR(s).     

Determination of uric acid and p-aminohippuric acid in human saliva and urine using capillary electrophoresis with electrochemical detection: potential application in fast diagnosis of renal disease

The monitoring of uric acid (UA) and p-aminohippuric acid (PAH) levels in biological samples is routinely carried out in clinical laboratories as an indication of renal disease. With the aim of investigation of the correlation between the trace amounts of UA and PAH in human saliva or urine and renal diseases, we carried out the determination of UA and PAH in human saliva and urine by using capillary electrophoresis with electrochemical detection (CE-ED) in this work. Under the optimum conditions, UA, PAH and three coexisting analytes could be well separated within 21 min at the separation voltage of 14 kV in 80 mmol/L borax running buffer (pH 9.2). Good linear relationship was established between peak current and concentration of analytes over two orders of magnitude with detection limits (S/N = 3) ranged from 5.01 x 10(-7) to 2.00 x 10(-6) mol/L for all analytes. The result shows that this proposed method could be successfully applied for the study on the correlation between the levels of UA and PAH in human saliva and urine and renal diseases, and provide an alternative and convenient method for the fast diagnosis of renal disease.     

SELDI-TOF-MS of saliva: methodology and pre-treatment effects

Interest in saliva as a diagnostic fluid for monitoring general health and for early diagnosis of disease has increased in the last few years. In particular, efforts have focused on the generation of protein maps of saliva using advanced proteomics technology. Surface-enhanced laser-desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) is a novel high throughput and extremely sensitive proteomic approach that allows protein expression profiling of large sets of complex biological specimens. In this study, large scale profiling of salivary proteins and peptides, ranging from 2 to 100kDa was demonstrated using SELDI-TOF-MS. Various methodological aspects and pre-analytical variables were analysed with respect to their effects on saliva SELDI-TOF-MS profiling. Results show that chip surface type and sample type (unstimulated versus stimulated) critically affect the amount and composition of detected salivary proteins. Factors that influenced normal saliva protein profiling were matrix composition, sample dilution and binding buffer properties. Delayed processing time experiments show certain new peptides evolving 3h post-saliva donation, and quantitative analyses indicate relative intensity of other proteins and peptides changing with time. The addition of protease inhibitors partly counteracted the destabilization of certain protein/peptide mass spectra over time suggesting that some proteins in saliva are subject to digestion by intrinsic salivary proteases. SELDI-TOF-MS profiles also changed by varying storage time and storage temperature whereas centrifugation speed and freeze-thaw cycles had minimal impact. In conclusion, SELDI-TOF-MS offers a high throughput platform for saliva protein and peptide profiling, however, (pre-)analytical conditions must be taken into account for valid interpretation of the acquired data.     

Comparative proteomic analysis of whole saliva from chronic periodontitis patients

Chronic periodontal disease is a chronic inflammatory process affecting tooth supporting tissues in the presence of pathogenic bacterial biofilm. There is some evidence for changes in the protein composition of whole saliva from chronic periodontitis patients, but there have been no studies using a proteomic approach. Hence, the aim of this study was to compare the protein profiles of unstimulated whole saliva from patients with periodontitis and healthy subjects by two complementary approaches (2D-gel electrophoresis and liquid chromatography). Protein spots of interest were analyzed by MALDI-TOF-TOF, and the data was complemented by an ESI-Q-TOF experiment. The analyses revealed that subjects with periodontal disease have increased amounts of blood proteins (serum albumin and hemoglobin) and immunoglobulin, and they have a lower abundance of cystatin compared to the control group. A higher number of protein spots were observed in the periodontitis group, of which most were identified as alpha-amylase. This higher number of alpha-amylase variants seems to be caused by hydrolysis by cysteine proteases under such inflammatory conditions. This approach gives novel insights into alterations of salivary protein in presence of periodontal inflammation and may contribute to the improvement of periodontal diagnosis.     

Simple and high-sensitivity detection of dioxin using dioxin-binding pentapeptide

The purposes of this study are to construct a simple dioxin detection system using peptides that bind to dioxin, and to test the system on real environmental samples. In this method, dioxin and N-NBD-3-(3',4'-dichlorophenoxy)-1-propylamine (NBD-DCPPA) are competitively bound to the peptides synthesized on beads. The fluorescence intensity of the bead decreases with increasing dioxin concentration. The concentration of dioxin is determined by measuring the fluorescence intensity using a fluorescence microscope equipped with a CCD camera. The fluorescence microscope system was equipped with a motor-driven stage and could be used with 96-well microplates and analytical software that automatically measured the fluorescence intensity of the bead images in the wells. Dioxin detection conditions, reaction temperature, number of beads and concentration of the organic solvent were optimized. About 0.5 nM (150 pg mL(-1)) of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TeCDD) could be detected under the optimized conditions. Environmental soil samples were subjected to the detection system using the peptide beads. Although the results obtained correlated poorly with the toxicity equivalency quantity (TEQ) concentration obtained by a GC/MS method, our method is robust enough as a prescreening method to detect at least 250 pg-TEQ g(-1), the survey level for soil as stipulated in the law concerning special measures against dioxins in Japan.     

A short proregion of trialysin, a pore-forming protein of Triatoma infestans salivary glands, controls activity by folding the N-terminal lytic motif

Triatoma infestans (Hemiptera: Reduviidae) is a hematophagous insect that transmits the protozoan parasite Trypanosoma cruzi, the etiological agent of Chagas' disease. Its saliva contains trialysin, a protein that forms pores in membranes. Peptides based on the N-terminus of trialysin lyse cells and fold into alpha-helical amphipathic segments resembling antimicrobial peptides. Using a specific antiserum against trialysin, we show here that trialysin is synthesized as a precursor that is less active than the protein released after saliva secretion. A synthetic peptide flanked by a fluorophore and a quencher including the acidic proregion and the lytic N-terminus of the protein is also less active against cells and liposomes, increasing activity upon proteolysis. Activation changes the peptide conformation as observed by fluorescence increase and CD spectroscopy. This mechanism of activation could provide a way to impair the toxic effects of trialysin inside the salivary glands, thus restricting damaging lytic activity to the bite site.     

Competitive binding assay using fluorescence resonance energy transfer for the identification of calmodulin antagonists

The ubiquitous calcium regulating protein calmodulin (CaM) has been utilized as a model drug target in the design of a competitive binding fluorescence resonance energy transfer assay for pharmacological screening. The protein was labeled by covalently attaching the thiol-reactive fluorophore, N-[2-(1-maleimidyl)ethyl]-7-(diethylamino)coumarin-3-carboxamide (MDCC) to an engineered C-terminal cysteine residue. Binding of the environmentally sensitive hydrophobic probe 2,6-anilinonaphthalene sulfonate (2,6-ANS) to CaM could be monitored by an increase in the fluorescence emission intensity of the 2,6-ANS. Evidence of fluorescence resonance energy transfer (FRET) from 2,6-ANS (acting as a donor) to MDCC (the acceptor in this system) was also observed; fluorescence emission representative of MDCC could be seen after samples were excited at a wavelength specific for 2,6-ANS. The FRET signal was monitored as a function of the concentration of calmodulin antagonists in solution. Calibration curves for both a selection of small molecules and a series of peptides based upon known CaM-binding domains were obtained using this system. The assay demonstrated dose-dependent antagonism by analytes known to hinder the biological activity of CaM. These data indicate that the presence of molecules known to bind CaM interfere with the ability of FRET to occur, thus leading to a concentration-dependent decrease of the ratio of acceptor:donor fluorescence emission. This assay can serve as a general model for the development of other protein binding assays intended to screen for molecules with preferred binding activity.     

Construction of a more sensitive fluorescence sensing material for the detection of vascular endothelial growth factor, a biomarker for angiogenesis, prepared by combining a fluorescent peptide and a nanopillar substrate

We have established a highly sensitive and selective protein detection technology in combination with the nanofabrication technique. A silica nanopillar chip with a 200-nm pitch and 1000-nm height pillar substrate was fabricated by electron beam lithography and deep reactive ion etching method. Fluorescent peptides, with high affinity towards vascular endothelial growth factor (VEGF), were immobilized on nanopillar chip via a self-assembled monolayer made from 3-aminopropyltrimethoxysilane and glutaraldehyde under optimal conditions. The fluorescence intensity of the fluorescent peptide on the nanopillar substrate increased with increasing VEGF concentrations, as determined by a fluorescence spectrophotometer and fluorescent scanning image analysis. The dissociation constant (K(d) value) calculated by the non-linear least square curve fitting method was 6.0 × 10(-9)M, which contributed to the highly sensitive detection of VEGF. The fluorescence intensity of the fluorescent reagent on the nanopillar substrate upon binding to VEGF was higher than that obtained using the flat substrate because the dense and tall nanopillar array increased the virtual protein binding area. The reproducibility tests and lifetime measurement indicate the fluorescent reagent to be a useful biosensor for the detection of VEGF in this system. These experimental results clearly showed that the combination of a fluorescent reagent and a nanopillar substrate may be widely applicable as a convenient method for the detection of VEGF.     

Multiplexed Fluorescent Microarray for Human Salivary Protein Analysis Using Polymer Microspheres and Fiber-optic Bundles

Herein, we describe a protocol for simultaneously measuring six proteins in saliva using a fiber-optic microsphere-based antibody array. The immuno-array technology employed combines the advantages of microsphere-based suspension array fabrication with the use of fluorescence microscopy. As described in the video protocol, commercially available 4.5 μm polymer microspheres were encoded into seven different types, differentiated by the concentration of two fluorescent dyes physically trapped inside the microspheres. The encoded microspheres containing surface carboxyl groups were modified with monoclonal capture antibodies through EDC/NHS coupling chemistry. To assemble the protein microarray, the different types of encoded and functionalized microspheres were mixed and randomly deposited in 4.5 μm microwells, which were chemically etched at the proximal end of a fiber-optic bundle. The fiber-optic bundle was used as both a carrier and for imaging the microspheres. Once assembled, the microarray was used to capture proteins in the saliva supernatant collected from the clinic. The detection was based on a sandwich immunoassay using a mixture of biotinylated detection antibodies for different analytes with a streptavidin-conjugated fluorescent probe, R-phycoerythrin. The microarray was imaged by fluorescence microscopy in three different channels, two for microsphere registration and one for the assay signal. The fluorescence micrographs were then decoded and analyzed using a homemade algorithm in MATLAB.     

Multiplex pathogen detection based on spatially addressable microarrays of barcoded resins

Suspension microsphere immunoassays are rapidly gaining recognition in antigen identification and infectious disease biodetection due to their simplicity, versatility and high-throughput multiplex screening. We demonstrate a multiplex assay based on antibody-functionalized barcoded resins (BCRs) to identify pathogen antigens in complex biological fluids. The binding event of a particular antibody on given bead (fluorescence) and the identification of the specific pathogen agent (vibrational fingerprint of the bead) can be achieved in a dispersive Raman system by exciting the sample with two different laser lines. Anthrax protective antigen, Franciscella tularensis lipopolysaccharide and CD14 antigens were accurately identified and quantified in tetraplex assays with a detection limit of 1 ng/mL. The rapid, versatile and simple analysis enabled by the BCRs demonstrates their potential for multiplex antigen detection and identification in a reconfigurable microarray format.     

Immunochromatographic assay of T-2 toxin using labeled anti-species antibodies

A new scheme of immunochromatographic assay was developed for the highly sensitive detection of low-molecular-weight analytes. This scheme includes the following two steps: the formation of complexes of free specific antibodies with an antigen and their detection by anti-species antibodies conjugated to gold nanoparticles as the label. This scheme was tested with mycotoxin T-2 toxin in maize extracts. The use of specific antibodies and a label as two individual components made it possible to independently vary their concentrations with a simultaneous decrease in the detection limit and an increase in the color intensity. The assay did not require additional reagents and manipulations. The instrumental and visual detection limits of the designed test system were 0.1 and 5.0 ng/mL, respectively (2 and 90 ng per gram of analytes), which are two orders of magnitude lower compared to conventional immunochromatography using the same reagents.     

Tools for the Quantitative Analysis of Sedimentation Boundaries Detected by Fluorescence Optical Analytical Ultracentrifugation

Fluorescence optical detection in sedimentation velocity analytical ultracentrifugation allows the study of macromolecules at nanomolar concentrations and below. This has significant promise, for example, for the study of systems of high-affinity protein interactions. Here we describe adaptations of the direct boundary modeling analysis approach implemented in the software SEDFIT that were developed to accommodate unique characteristics of the confocal fluorescence detection system. These include spatial gradients of signal intensity due to scanner movements out of the plane of rotation, temporal intensity drifts due to instability of the laser and fluorophores, and masking of the finite excitation and detection cone by the sample holder. In an extensive series of experiments with enhanced green fluorescent protein ranging from low nanomolar to low micromolar concentrations, we show that the experimental data provide sufficient information to determine the parameters required for first-order approximation of the impact of these effects on the recorded data. Systematic deviations of fluorescence optical sedimentation velocity data analyzed using conventional sedimentation models developed for absorbance and interference optics are largely removed after these adaptations, resulting in excellent fits that highlight the high precision of fluorescence sedimentation velocity data, thus allowing a more detailed quantitative interpretation of the signal boundaries that is otherwise not possible for this system.     

Determination of the human salivary peptides histatins 1, 3, 5 and statherin by high-performance liquid chromatography and by diode-array detection

A reversed-phase high-performance liquid chromatography (HPLC) method with diode-array detection for the quantification of several human salivary peptides is described. Sample pretreatment consisted of the acidification of whole saliva by phosphate buffer. This treatment produced precipitation of mucins, α-amylases and other high-molecular-mass salivary proteins, simultaneous inhibition of intrinsic protease activities and reduction of sample viscosity. Direct HPLC analysis by diode-array detection of the resulting acidic sample allowed one to quantify histatin 1, histatin 3, histatin 5, statherin, as well as uric acid, in normal subjects. Moreover, the groups of peaks pertaining to proline-rich proteins and cystatins were tentatively identified. The method can be useful in assessing the concentration of salivary peptides from normal subjects and from patients suffering oral and/or periodontal diseases.     

Development of a convenient and supersensitive high-throughput screening system for genetically encoded fluorescent probes of small molecules using a confocal microscope

Monitoring the dynamic patterns of intracellular signaling molecules, such as inositol 1,4,5-trisphosphate (IP₃) and Ca²⁺, that control many diverse cellular processes, provides us significant information to understand the regulatory mechanism of cellular functions. For searching more sensitive and higher dynamic range probes for signaling molecules, convenient and supersensitive high throughput screening systems are required. Here we show the optimal “in Escherichia coli (E. coli) colony” screening method based on the twin-arginine translocase (Tat) pathway and introduce a novel application of a confocal microscope as a supersensitive detection system to measure changes in the fluorescence intensity of fluorescent probes in E. coli grown on an agar plate. To verify the performance of the novel detection system, we compared the changes detected in the fluorescent intensity of genetically encoded Ca²⁺ indicator after Ca²⁺ exposure to two kinds of conventional fluorescence detection systems (luminescent image analyzer and fluorescence stereomicroscope). The rate of fluorescence change between Ca²⁺ binding and unbinding detected by novel supersensitive detection system was almost double than those measured by conventional detection systems. We also confirmed that the Tat pathway-based screening method is applicable to the development of genetically encoded probes for IP₃. Our convenient and supersensitive screening system improves the speed of developing florescent probes for small molecules.     

Sensitive detection of the redox state of copper proteins using fluorescence

The blue copper protein azurin from Pseudomonas aeruginosa has been covalently labelled with the fluorescing dye Cy5. The optical spectrum of the azurin changes markedly with its redox state. These changes are reflected in the fluorescence intensity of the dye through fluorescence resonance energy transfer (FRET). This provides a sensitive way to monitor biological redox events. The method shown to work in the nanomolar range of protein concentrations, can be easily extended into the sub-nanomolar regime and holds promise for single-molecule detection.     

Protein detection by Western blot via coiled-coil interactions

We propose an approach for the detection of proteins by Western blot that takes advantage of the high-affinity interaction occurring between two de novo designed peptides, the E and K coils. As a model system, K coil-tagged epidermal growth factor (EGF) was revealed with secreted alkaline phosphatase (SeAP) tagged with E coil (SeAP-Ecoil) as well as with biotinylated E coil. In that respect, we first produced purified SeAP-Ecoil and verified its ability to interact with K coil peptides by surface plasmon resonance biosensing. We demonstrated that protein detection with Ecoil-biotin was more specific than with SeAP-Ecoil. We then showed that our approach is as sensitive as conventional detection strategies relying on nickel-nitrilotriacetic acid-horseradish peroxidase (Ni-NTA-HRP), anti-His-HRP, or anti-EGF. Altogether, our results indicate that the E/K coiled-coil system is a good alternative for protein detection by Western blot.     

Analysis of several fluorescent detector molecules for protein microarray use

The utility of several streptavidin‐linked fluorescent detector molecules was evaluated on two protein microarray platforms. Tested detector molecules included: Alexa Fluor 546; R‐phycoerythrin (RPE), orange fluospheres; Cy3‐containing liposomes (Large Unilamellar Vesicles, LUV) labelled with Cy3; and an RPE–antibody complex. The two array architectures tested consisted of an array of murine Fc–biotin and an array of murine IgG (the murine IgG array was probed with a biotinylated rabbit anti‐murine IgG). These platforms allowed for the direct comparison of detector utility by detector recognition of array‐bound biotin. All of the fluorescent detectors examined demonstrated utility on each of the array platforms. For the Fc–biotin array, detector signal intensity (background adjusted) was as follows: RPE–antibody complex > fluospheres > RPE > liposomes > Alexa 546: for the IgG array: RPE/antibody complex > RPE > fluospheres > Alexa546 > liposomes. The RPE–antibody complex fluoresced 67% and 150% more intensely than the next closest detector molecule for the Fc–biotin and the murine IgG arrays, respectively. A marked increase in background fluorescence (as compared to RPE alone) did not accompany the increase in signal intensity gained through RPE–antibody complex use (a true increase in signal:noise ratio). These results suggest that the RPE–antibody complex is superior to other molecules for fluorescent detection of analytes on protein microarrays. Copyright © 2002 John Wiley & Sons, Ltd.