HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Electronic Appendix Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Silwood, C.J.L. Articles by Grootveld, M.C. Search for Related Content PubMed PubMed Citation Articles by Silwood, C.J.L. Articles by Grootveld, M.C.

¹H and ¹³C NMR Spectroscopic Analysis of Human Saliva

¹ Medical Unit, St. Bartholomews and the Royal London School of Medicine and Dentistry, London E1 1BB, UK; and
² School of Dentistry, The Queen's University of Belfast, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BP, Northern Ireland;

*corresponding author, Room AW518, Medical Unit, 5th floor, Alexandra Wing, Royal London Hospital, London E1 1BB, UK, m.grootveld{at}qmul.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
We have explored the ability of high-resolution NMR techniques to (1) index salivary biomolecules and (2) provide valuable data regarding intra- and inter-subject variability in the concentrations of a series of components readily determinable by this technique (organic acids and malodorous amines). Experiments were conducted on ‘whole’ saliva samples collected from 20 patients, either randomly during their daily activities, or, for investigations involving the quantification of salivary biomolecules, immediately after they woke in the morning throughout a three-day period. These NMR techniques permitted us to detect greater than 60 metabolites, together with agents arising from dietary, oral health care product, and pharmaceutical sources. Highly significant "between-subject" differences in the a.m. waking salivary metabolite concentrations were found for 9 out of 11 components monitored. It is concluded that NMR spectroscopy serves as a powerful technique for the multicomponent analysis of human saliva.

KEY WORDS: human saliva • multicomponent analysis • NMR spectroscopy • periodontal diseases • metabolic profile

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
High-resolution NMR spectroscopy is now an established analytical tool which has been extensively used for the purpose of probing the metabolic status of biofluids (Lindon et al., 1999). Indeed, this technique possesses many advantages over alternative analytical methods, since: (1) it is virtually non-invasive; (2) it provides simultaneous multicomponent information regarding the metabolic profiles of biofluids and appropriate tissue sample extracts; (3) it requires minimal sample preparation, and the analysis is fast (requiring just 10-15 min); (4) components containing more than 1 class of ¹H nuclei display 2 or more "fingerprint" signals in spectra acquired (spin-spin coupled if the nuclei are adjacent to each other), facilitating their rapid identification; (5) it has a high degree of spectral dispersion and sensitivity (less than or equivalent to µmol/L levels); and (6) molecules which would not necessarily be anticipated to be present in biological samples can be identified.

The organic acid concentrations of human saliva specimens have been previously monitored by labor-intensive, relatively time-consuming laboratory methods such as those involving gas-liquid chromatography (GLC) (Tyler, 1971; Lambert and Moss, 1972; Botta et al., 1985), column chromatography (Vratsanos, 1981; Vratsanos and Mandel, 1982), and high-performance liquid chromatography (HPLC) (Linke and Moss, 1992; Linke et al., 1997; Linke and Birkenfeld, 1999). For these methods, however, much information concerning the particular biomolecules present in such samples is a pre-requisite of analysis. Consequently, such analytical methods generally offer only a partial characterization of the metabolic status of biofluids and are essentially sample-destructive. NMR spectroscopy is therefore ideally suited as an analytical technique for saliva.

With the exception of an extremely limited number of studies (Harada et al., 1987; Yamadanosaka et al., 1991), saliva has been largely overlooked as a biofluid for NMR spectroscopic analysis. Therefore, we report here the use of high-field, high-resolution NMR techniques for the purposes of (1) indexing salivary biomolecules, and (2) evaluating both intra- and inter-subject variabilities in the salivary concentrations of a range of such components in dental patients. To the best of our knowledge, this work includes the first report of the applications of two-dimensional (2D) NMR techniques to the analysis of human saliva.

	MATERIALS & METHODS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Sample Collection and Preparation
For investigations involving the ¹H NMR determination of salivary biomolecules, 20 patients attending the Dental School for regular dental care and who were not medically compromised were recruited. These subjects ranged in age from 21 to 49 yrs (mean age ± standard deviation, 34.6 ± 9.7 yrs), and none of them had any active periodontal disease or active dental caries. To avoid any interferences arising from the introduction of exogenous agents into the oral environment, we asked the patients to collect all saliva available (ca. 2-4 mL), i.e., ("whole") saliva expectorated from the mouth, into a plastic universal tube immediately after waking in the morning, each day for a three-day period. Each patient was also requested to refrain completely from oral activities (i.e., eating, drinking, toothbrushing, oral rinsing, smoking, etc.) during the short period between awakening and sample collection (< 5 min). Each collection tube contained sufficient sodium fluoride (15 µmol) to ensure that metabolites are not generated or consumed via the actions of bacteria or bacterial enzymes present in whole saliva during periods of sample preparation and/or storage. Specimens were transported to the laboratory on ice and centrifuged immediately on their arrival to remove cells and debris, and the supernatants were then stored at –70°C for a maximum of 18 hrs prior to NMR analysis.

The ¹H NMR profiles and biomolecule concentrations of salivary supernatant specimens subjected to analysis immediately after collection into the fluoride-containing tubes and rapid centrifugation were compared with those of the same samples stored as described above, and no differences were discernible, i.e., none of the criteria investigated changed significantly during these periods of storage.

Further experiments involved the random collection of saliva samples from each of the above volunteers during their normal daily activities. These specimens were collected, transported, treated, and stored in the same manner as those obtained above.

The use of human materials conformed to an informed consent protocol that was approved by the Research Ethics Committee of the East London and City Health Authority (reference no. P98 057).

NMR Measurements
Both one-dimensional (1D) and 2D NMR spectra were acquired on a Bruker AMX-600 spectrometer [University of London Intercollegiate Research Services (ULIRS), Queen Mary, University of London facility, UK] operating at a frequency of 600.13 (¹H) and 150.93 (¹³C) MHz and a probe temperature of 298°K. We prepared samples by taking 0.60 mL of salivary supernatant, to which deuterium oxide (0.07 mL, providing a field frequency lock) and a 5.00 mM solution of sodium 3-trimethylsilyl [2,2,3,3-²H₄] propionate (TSP) in deuterium oxide (0.03 mL, chemical shift reference for both ¹H and ¹³C spectra, = 0.00 ppm) were added. Experimental conditions for the acquisition of 1D spectra acquired on these specimens appear in Appendix A (www.dentalresearch.org), as do those regarding samples subjected to ¹H-¹H total correlation (TOCSY) and J-resolved (JRES) and heteronuclear multiple quantum coherence (HMQC) spectroscopies utilizing the pulse sequences of Braunschweiler and Ernst (1983), Aue et al. (1976), and Bax et al. (1983), respectively.

Statistical Analysis
For each metabolite and subject, the mean salivary concentration and its corresponding "between-days" standard error were computed on the raw (untransformed) data.

Biomolecule concentration data were also subjected to the transformation y = log_e (1 + x) to satisfy assumptions of normality and homogeneity of "between-days within-patient" variances, and a one-way analysis of variance (components of variance model) was conducted. Subsequently, "between-patients" and "between-days" components of variance (s_p² and s², respectively) were estimated and the significance of the former determined.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
1D ¹H NMR Spectroscopy of Human Saliva
Complete and expanded 0.80-2.50, 2.55-4.20, and 6.85-8.50 ppm regions of the 600.13 MHz 1D ¹H NMR spectrum of a typical unstimulated human saliva sample are shown in Fig. 1. This spectrum contains many prominent, sharp resonances attributable to a wide range of low-molecular-mass components, together with the molecularly mobile, NMR-responsive portions of certain macromolecules. Indeed, signals assignable to short-chain organic acid anions—such as n-butyrate, iso-butyrate, propionate, lactate, acetate, pyruvate and succinate, amino acids (e.g., alanine, valine, leucine, histidine, phenylalanine, and tyrosine), and carbohydrates—are readily observable. Signal assignments were confirmed by the methods given in Appendix A (www.dentalresearch.org).

View larger version (26K): [in this window] [in a new window]   Figure 1. Single-pulse 1D 1H NMR spectrum of a human salivary supernatant specimen. (a-c) Expanded 0.80-2.50, 2.55-4.20, and 6.85-8.50 ppm regions, respectively, of the 600.13-MHz single-pulse 1H NMR spectrum of a human salivary supernatant specimen (pH value 6.78). A typical spectrum is shown. Abbreviations: A, acetate-CH3; Ala I, alanine-CH3; Bu I, ß-hydroxybutyrate -CH3 group protons; Bu II, III, and IV, ß-hydroxybutyrate ß, ß', and protons, respectively (ABX coupling system); iso-But I and II, iso-butyrate-CH3 and -CH group protons, respectively; n-But I, II, and III, n-butyrate , ß, and protons, respectively; Chol, choline-N+(CH3)3; DMeN, dimethylamine-CH3; Eth I and II, ethanol-CH3 and -CH2 group protons, respectively; Form, formate-H; Gly, glycine-CH2; His I and II, histidine ABX ß protons; His III and IV, histidine imidazole ring protons; Lac I and II, lactate-CH3 and -CH protons, respectively; Leu I, II, III, and IV, leucine , , ß, and protons, respectively; MeGu, methylguanidine-CH3; MeN, methylamine-CH3; Meth, methanol-CH3; N-Ac, spectral region for acetamido methyl groups of N-acetyl sugars; Phe I and II, phenylalanine ABX ß protons; Phe III, phenylalanine ABX proton; Phe IV, V, and VI, phenylalanine aromatic ring protons; Prop I and II, propionate-CH3 and -CH2 group protons, respectively; Pyr, pyruvate-CH3; Sar I and II, sarcosine-CH3 and -CH2 group protons, respectively; Suc, succinate-CH2; TMAO, trimethylamine oxide ON(CH3)3; TMeN, trimethylamine-CH3; Tyr I and II, tyrosine ABX ß protons; Tyr III, tyrosine ABX proton; Tyr IV and V, tyrosine aromatic ring protons; and Val I and II, n-valerate and protons, respectively. Assignments of 1H NMR resonances to methylguanidine, sarcosine, n-valerate, and N-acetylsugars are tentative in this instance.

Two-dimensional ¹H-¹H Correlation Spectroscopy of Human Saliva
The 0.50-5.50 ppm region of a typical 600 MHz ¹H-¹H TOCSY NMR spectrum of human saliva is shown in Fig. 2, and strong connectivities for the ¹H nuclei of many low-molecular-mass metabolites and catabolites are readily observable. Such spectra readily aid the assignment of resonances from free amino acids, e.g., alanine, glutamine, isoleucine, lysine, ornithine, and proline. Many further low-molecular-mass components are detectable in such TOCSY spectra, including -amino-n-butyrate and propane-1,2-diol (OHCP agent). Connectivities for adjacent aliphatic protons in saturated fatty acids, i.e., CH₃(CH₂)_nCH₂CH₂CO-, are also readily observable.

View larger version (35K): [in this window] [in a new window]   Figure 2. Partial 1H-1H TOCSY NMR spectrum of a human salivary supernatant specimen. A typical spectrum is shown. KEY: 1, saturated fatty acid (terminal end spin system); 2, leucine; 3, saturated fatty acid (intermediate carbon spin system); 4, lysine; 5, unassigned; 6, -aminobutyric acid; 7, ornithine; 8, unassigned; 9, isoleucine; 10, propionate; 11, valine; 12, glutamate; 13, glutamine; 14, proline; 15, unassigned; 16, propane-1,2-diol; 17, unassigned; 18, threonine; 19, unassigned; 20, phosphorylethanolamine; 21, tyrosine; 22, phenylalanine; 23, aspartic acid; 24, alanine; 25, ethanol; 26, lactate; and 27, 28, 29, and 30, unassigned.

Two-dimensional ¹H-¹³C Heteronuclear Multiple Quantum Coherence Transfer Spectroscopy of Human Saliva
The 600-MHz ¹H-¹³C inverse-detected HMQC spectra of human saliva contained a range of prominent cross-peaks for many low-molecular-mass components, e.g., organic acid anions and amino acids (a typical spectrum is shown in Appendix B, www.dentalresearch.org). The 2D spectrum shown in Appendix B (www.dentalresearch.org) clearly demonstrates the ability of the HMQC technique to resolve overlapping signals, the use of the 8-mm ¹³C probe in this case undoubtedly aiding the detection of many components present at low concentrations. The corroboration of prior ¹H NMR assignments with the corresponding ¹³C NMR data readily facilitates the identification of particular biomolecules, the ¹³C assignments and related HMQC correlated resonances being confirmed via reference to known chemical shift values for authentic model compounds (Lindon et al., 1999).

Intra- and Inter-individual Variability of ¹H NMR-determined Salivary Biomolecule Concentrations in Dental Patients
Table A1 (Appendix C, www.dentalresearch.org) gives the mean ± "between-days" standard error values for the concentrations of the organic acid anions acetate, iso-butyrate, n-butyrate, formate, lactate, propionate, pyruvate, and succinate, and the malodorous amines methyl-, dimethyl-, and trimethylamine, in saliva specimens collected immediately after a.m. awakening, for each of the 20 patients investigated. The "between-subjects" ranges of these "between-days" mean values were: acetate, 31.4-307.0 mM; iso-butyrate, 0.01-1.76 mM; n-butyrate, 0-2.94 mM; formate, 0.24-61.3 mM; lactate, 0.08-100.9 mM; propionate, 5.92-69.2 mM; pyruvate, 0.10-10.6 mM; succinate, 0.06-4.46 mM; methylamine, 0-316 µM; dimethylamine, 9-262 µM; and trimethylamine, 9-309 µM.

One-way analysis of variance performed on log_e(x + 1)-transformed data revealed that, with the exception of n-butyrate and dimethylamine, the estimated "between-subjects" component of variance (s_p²) was significantly greater than that "between days" (s²) for all biomolecules examined in this manner (p less than 0.001-0.025), indicating that differentiation between individual dental patients is readily achievable. These estimated components of variance and the statistical significance of the s_p² values are listed in Table A2 (Appendix C, www.dentalresearch.org) for each metabolite determined.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
The use of a combination of 1D and three separate 2D NMR spectroscopic methods for the analysis of salivary components provides much valuable molecular information which may offer valuable diagnostic information. Indeed, the NMR analysis of human biofluids has proved to be a valuable aid in the diagnosis of a range of clinical conditions, e.g., diabetes mellitus (Bell et al., 1989), inflammatory joint diseases (Naughton et al., 1993), and renal failure (Holmes et al., 1990).

In addition to the identification of greater than 60 endogenous salivary biomolecules, agents arising from dietary sources, oral health care products, and pharmaceutical preparations are also readily detectable. The short-chain organic acid anions detectable reflect, at least in part, the growth, preponderance, and metabolism of micro-organisms (Guerrant et al., 1982), and it is conceivable that selected individual or patterns of these metabolites represent chemotaxonomic markers of microbial infiltration. Similarly, the N-acetylsugars detectable are presumably derived from the actions of the bacterial enzymes hyaluronidase and neuraminidase, respectively.

Although the consumption of alcoholic beverages undoubtedly represents a source of salivary ethanol, this component may also arise from carbohydrate metabolism by selected bacteria [e.g., Streptococcus mutans (Williams and Elliott, 1979)], and the detectable methanol is derived from the passive or direct inhalation of cigarette smoke, which contains this xenobiotic. A complete list of assignments for resonances present in 600 MHz ¹H NMR spectra of human salivary supernatants, together with their corresponding spin-spin coupling patterns, are given in Table A3 (Appendix C, www.dentalresearch.org). Of course, this list should be considered inexhaustive in view of the advent of spectrometers of higher operating frequencies. The TOCSY technique afforded the unambiguous identification of complete molecular "backbones", while the JRES and HMQC methods overcame problems associated with spectral signal overlap.

The TOCSY spectroscopic technique permits the transfer of magnetization from a ¹H nucleus [or magnetically equivalent group (2 or 3) of such nuclei] bonded to one specific carbon atom (C₁), to one or more magnetically distinct nuclei located two or more carbon positions further along a molecular chain (C₃, C₄, C₅ position, etc.), i.e., the latter nucleus/nuclei is/are not directly coupled to the C₁-bearing ¹H nucleus.

Moreover, the ¹H-¹H JRES technique successfully resolves many complex overlapping multiplet signals by a dispersion of chemical shift and coupling constant data into two orthogonal frequency domains, an advantage which readily facilitates spectral assignment.

The direct observation of ¹³C nuclei is of a poor sensitivity in view of its low natural abundance (1.11%), particularly for salivary components present at relatively low concentrations. However, the recent development of inverse-geometry probes and relevant accompanying pulse sequences has generally overcome this limitation, since HMQC spectroscopy and related techniques offer a marked increase in sensitivity over conventional 1D spectroscopy.

Metabolite concentration data determined by 1D NMR spectroscopy clearly offer the ability to detect highly significant differences between individuals for all but 2 of the biomolecules examined, and experiments aimed at establishing its diagnostic capacity, i.e., its ability to discriminate between different periodontal diseases, are currently in progress. Indeed, the nature and levels of salivary organic acids may serve as markers of the susceptibility of patients to dental caries. Moreover, the amines determined here may represent one or more potentially toxic agents generated by bacteria implicated in the etiology of periodontal diseases.

Both 1D and 2D NMR analyses of human saliva specimens collected prior and subsequent to the administration of oral health care products to patients with periodontal diseases may demonstrate a reduction in the salivary concentrations of microbial-derived catabolites (for example, short-chain organic acids), a process conceivably contingent on the removal of cariogenic micro-organisms such as lactobacilli, streptococci, and Gram-positive pleomorphic rods (Guerrant et al., 1982). Indeed, recent pilot studies conducted by the authors have shown that the techniques outlined here are readily applicable to such investigations.

In conclusion, high-resolution, high-field 1 and 2D ¹H NMR techniques offer many advantages over alternative time-consuming, labor-intensive analytical methods, since they allow for the rapid, virtually non-invasive and simultaneous examination of a very wide range of components present in saliva.

	ACKNOWLEDGMENTS

 
We are very grateful to the University of London Intercollegiate Research Services (ULIRS) for the provision of NMR facilities, and to Peter Haycock for excellent technical assistance. This research was supported by institutional funding.

	FOOTNOTES

 
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org.

Received June 20, 2001; Last revision April 1, 2002; Accepted April 18, 2002

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Aue WP, Karhan J, Ernst RR (1976). Homonuclear broad band decoupling and two-dimensional J-resolved NMR spectroscopy. J Chem Phys 64:4226–4227.

Bax A, Griffey RH, Hawkins BL (1983). Correlation of ¹H and ¹⁵N chemical shifts by multiple quantum NMR. J Magnet Reson 64:547–552.

Bell JD, Brown JCC, Nicholson JK, Sadler PJ (1987). Assignment of resonances for ‘acute-phase’ glycoproteins in high resolution proton NMR spectra of human blood plasma. FEBS Lett 215:311–315.[Medline]

Bell JD, Brown JCC, Sadler PJ (1989). NMR studies of body fluids. NMR Biomed 2:246–256.[Medline]

Botta GA, Radin L, Costa A, Schito G, Blasi G (1985). Gas-liquid chromatography of the gingival fluid as an aid in periodontal diagnosis. J Periodontal Res 20:450–457.[Medline]

Braunschweiler L, Ernst RR (1983). Coherence transfer by isotropic mixing—application to proton correlation spectroscopy. J Magnet Reson 53:521–528.

Guerrant GO, Lambert MA, Moss CW (1982). Analysis of short-chain acids from anaerobic bacteria by high performance liquid chromatography. J Clin Microbiol 16:355–360.[Abstract/Free Full Text]

Harada H, Shimizu H, Maeiwa M (1987). ¹H NMR of human saliva. An application of NMR spectroscopy in forensic science. Foren Sci Int 34:189–195.

Holmes E, Foxall PJD, Nicholson JK (1990). Proton NMR analysis of plasma from renal-failure patients—evaluation of sample preparation and spectral-editing methods. J Pharm Biomed Anal 8:955–958.[Medline]

Lambert JE, Moss CW (1972). Gas-liquid chromatography of short-chain fatty acids on Dexsil 300GC. J Chromatogr74:335–338.[Medline]

Lindon JC, Nicholson JK, Everett JR (1999). NMR spectroscopy of biofluids. Ann Rep NMR Spec 38:1–88.

Linke HA, Birkenfeld LH (1999). Clearance and metabolism of starch foods in the oral cavity. Ann Nutr Metab 43:131–139.[Medline]

Linke HA, Moss SJ (1992). Quantitation of carbohydrate sweeteners and organic acids in human oral fluid using HPLC analysis. Zeit für Ernahrung 31:147–154.

Linke HA, Moss SJ, Arav L, Chiu PM (1997). Intra-oral lactic acid production during clearance of different foods containing various carbohydrates. Zeit für Ernahrung 36:191–197.

Naughton DP, Haywood R, Blake DR, Edmonds S, Hawkes GE, Grootveld M (1993). A comparative evaluation of the metabolic profiles of normal and inflammatory knee-joint synovial fluids by high resolution proton NMR spectroscopy. FEBS Lett 332:221–225.[Medline]

Tyler JE (1971). Quantitative estimation of volatile fatty acids in carious enamel by gas chromatography of their methyl esters. J Dent Res 50:1189–1194.

Vratsanos SM (1981). Chromatographic microanalysis of organic acids in plaque related to food cariogenicity. In: Foods, nutrition and dental health. Vol. 3. Hefferren JJ, Ayer WA, Koehler HM, editors. Chicago, IL: Pathotox, pp. 89-91.

Vratsanos SM, Mandel ID (1982). Comparative plaque acidogenesis of caries-resistant vs. caries-susceptible adults. J Dent Res 61:465–468.[Abstract/Free Full Text]

Williams RAD, Elliott JC (1979). Basic and applied dental biochemistry. Edinburgh: Churchill Livingstone, p. 180.

Yamadanosaka A, Fukutomi S, Uemura S, Hashida T, Fujishita M, Kobayashi Y, et al. (1991). Preliminary nuclear magnetic resonance studies on human saliva. Arch Oral Biol 36:697–701.[Medline]

This article has been cited by other articles:

				M. C Walsh, L. Brennan, J P. G Malthouse, H. M Roche, and M. J Gibney Effect of acute dietary standardization on the urinary, plasma, and salivary metabolomic profiles of healthy humans. Am. J. Clinical Nutrition, September 1, 2006; 84(3): 531 - 539. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Electronic Appendix Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Silwood, C.J.L. Articles by Grootveld, M.C. Search for Related Content PubMed PubMed Citation Articles by Silwood, C.J.L. Articles by Grootveld, M.C.