Elaine Holmes

Development of a model for classification of toxin‐induced lesions using 1H NMR spectroscopy of urine combined with pattern recognition

Pattern recognition approaches were developed and applied to the classification of 600 MHz 1H NMR spectra of urine from rats dosed with compounds that induced organ‐specific damage in either the liver or kidney. Male rats were separated into groups (n = 5) and each treated with one of the following compounds; adriamycin, allyl alcohol, 2‐bromoethanamine hydrobromide, hexachlorobutadiene, hydrazine, lead acetate, mercury II chloride, puromycin aminonucleoside, sodium chromate, thioacetamide, 1,1,2‐trichloro‐3,3,3‐trifluoro‐1‐propene or dose vehicle. Urine samples were collected over a 7 day time‐course and analysed using 600 MHz 1H NMR spectroscopy. Each NMR spectrum was data‐reduced to provide 256 intensity‐related descriptors of the spectra. Data corresponding to the periods 8–24 h, 24–32 h and 32–56 h post‐dose were first analysed using principal components analysis (PCA). In addition, samples obtained 120–144 h following the administration of adriamycin and puromycin were included in the analysis in order to compensate for the late onset of glomerular toxicity. Having established that toxin‐related clustering behaviour could be detected in the first three principal components (PCs), three‐quarters of the data were used to construct a soft independent modelling of class analogy (SIMCA) model. The remainder of the data were used as a test set of the model. Only three out of 61 samples in the test set were misclassified. Finally as a further test of the model, data from the 1H NMR spectra of urine from rats that had been treated with uranyl nitrate were used. Successful prediction of the toxicity type of the compound was achieved based on NMR urinalysis data confirming the robust nature of the derived model.

Metabolic profiling strategy for discovery of nutritional biomarkers: proline betaine as a marker of citrus consumption

BACKGROUND New food biomarkers are needed to objectively evaluate the effect of diet on health and to check adherence to dietary recommendations and healthy eating patterns. OBJECTIVE We developed a strategy for food biomarker discovery, which combined nutritional intervention with metabolic phenotyping and biomarker validation in a large-scale epidemiologic study. DESIGN We administered a standardized diet to 8 individuals and established a putative urinary biomarker of fruit consumption by using (1)H nuclear magnetic resonance (NMR) spectroscopic profiling. The origin of the biomarker was confirmed by using targeted NMR spectroscopy of various fruit. Excretion kinetics of the biomarker were measured. The biomarker was validated by using urinary NMR spectra from UK participants of the INTERMAP (International Collaborative Study of Macronutrients, Micronutrients, and Blood Pressure) (n = 499) in which citrus consumption was ascertained from four 24-h dietary recalls per person. Finally, dietary patterns of citrus consumers (n = 787) and nonconsumers (n = 1211) were compared. RESULTS We identified proline betaine as a putative biomarker of citrus consumption. High concentrations were observed only in citrus fruit. Most proline betaine was excreted < or =14 h after a first-order excretion profile. Biomarker validation in the epidemiologic data showed a sensitivity of 86.3% for elevated proline betaine excretion in participants who reported citrus consumption and a specificity of 90.6% (P < 0.0001). In comparison with noncitrus consumers, citrus consumers had lower intakes of fats, lower urinary sodium-potassium ratios, and higher intakes of vegetable protein, fiber, and most micronutrients. CONCLUSION The biomarker identification and validation strategy has the potential to identify biomarkers for healthier eating patterns associated with a reduced risk of major chronic diseases. The trials were registered at clinicaltrials.gov as NCT01102049 and NCT01102062.

750 MHz 1H NMR spectroscopy characterisation of the complex metabolic pattern of urine from patients with inborn errors of metabolism : 2-hydroxyglutaric aciduria and maple syrup urine disease

750 MHz 1H NMR spectroscopy has been used to characterise in detail the abnormal low molecular weight metabolites of urine from two patients with inborn errors of metabolism. One case of the rare condition 2-hydroxyglutaric aciduria has been examined. There is at present no rapid routine method to detect this genetic defect, although NMR spectroscopy of urine is shown to provide a distinctive pattern of resonances. Assignment of a number of prominent urinary metabolites not normally seen in control urine could be made on the basis of their known NMR spectral parameters including the diagnostic marker 2-hydroxyglutaric acid, which served to confirm the condition. In addition, 750 MHz 1H NMR spectroscopy has been used to characterise further the abnormal metabolic profile of urine from a patient with maple syrup urine disease. This abnormality arises from a defect in branched chain keto-acid decarboxylase activity and results in a build up in the urine of high levels of branched chain oxo- and hydroxy-acids resulting from altered metabolism of the branched chain amino acids, valine, leucine and isoleucine. A number of previously undetected abnormal metabolites have been identified through the use of one-dimensional and two-dimensional J-resolved and COSY 750 MHz 1H NMR spectroscopy, including ethanol, 2-hydroxy-isovalerate, 2,3-dihydroxy-valerate, 2-oxo-3-methyl-n-valerate and 2-oxo-isocaproate. NMR spectroscopy of urine, particularly when combined with automatic data reduction and computer pattern recognition using a combination of biochemical markers, promises to provide an efficient alternative to other techniques for the diagnosis of inborn errors of metabolism.

Application of chemometrics to the 1H NMR spectra of apple juices : discrimination between apple varieties

Discrimination between apple juices produced from different varieties (Spartan, Bramley, Russet) has been achieved by applying principal components analysis (PCA) and linear discriminant analysis to 1H NMR spectra of the juices. The use of covariance and correlation matrix PCA methods was investigated and different regions of the spectrum were analysed in view of the large range of signal intensities. All the methods gave a high success rate of classification, with at least 24 out of 26 samples being correctly assigned when five principal components were used. Under optimum conditions a 100% success rate was achieved. Examination of the principal component loadings showed that the levels of malic acid and sucrose were two important chemical variables, but variations in the composition of the minor constituents were also found to make a significant contribution to the discrimination.

Automatic reduction of NMR spectroscopic data for statistical and pattern recognition classification of samples

A general method of automatically reducing NMR spectra to provide numerical descriptors of samples has been developed and investigated. These descriptors can be used as input to pattern recognition or multivariate algorithms for sample classification. The methods have been tested using 600 MHz one-dimensional 1H NMR spectra of biofluids which are complex mixtures. The approach is, in principle, applicable to multidimensional and heteronuclear NMR spectra and to other types of liquid samples such as oils and foodstuffs as well as to situations such as 1H or 31P NMR in vivo and solid state NMR in drug formulation analysis. The method relies upon apportioning the information in the spectra to individual contiguous segments and allowing specified regions of the spectra to be omitted. Three approaches, based on the number of peaks, the summed peak heights and the summed peak areas respectively in each segment, have been tested. The effect of segment width and overlap and the effects of manipulation of the NMR spectra have been evaluated in terms of the classification of the samples using principal components analysis. A simple method of generating NMR based spectral descriptors for object classification is thus proposed.

Metabolic profiling and the metabolome-wide association study: Significance level for biomarker identification

High throughput metabolic profiling via the metabolome-wide association study (MWAS) is a powerful new approach to identify biomarkers of disease risk, but there are methodological challenges: high dimensionality, high level of collinearity, the existence of peak overlap within metabolic spectral data, multiple testing, and selection of a suitable significance threshold. We define the metabolome-wide significance level (MWSL) as the threshold required to control the family wise error rate through a permutation approach. We used 1H NMR spectroscopic profiles of 24 h urinary collections from the INTERMAP study. Our results show that the MWSL primarily depends on sample size and spectral resolution. The MWSL estimates can be used to guide selection of discriminatory biomarkers in MWA studies. In a simulation study, we compare statistical performance of the MWSL approach to two variants of orthogonal partial least-squares (OPLS) method with respect to statistical power, false positive rate and correspondence of ranking of the most significant spectral variables. Our results show that the MWSL approach as estimated by the univariate t test is not outperformed by OPLS and offers a fast and simple method to detect disease-related discriminatory features in human NMR urinary metabolic profiles.

1H and 2H NMR spectroscopic studies on the metabolism and biochemical effects of 2-bromoethanamine in the rat

Male Fischer 344 rats were dosed with 2-bromoethanamine hydrobromide (BEA, N = 6) or [1,2,2,-2H4]-bromoethanamine hydrobromide (BEA-d4, N = 6) at 150 mg/kg i.p. and urine was collected -24 to 0 hr pre-dose and at 0-2 hr, 2-4 hr, 4-8 hr and 8-12 hr post-dose (p.d.). Urine samples were analysed directly using 500 and 600 MHz 1H NMR and 92.1 MHz 2H NMR spectroscopy. The major observed effect of BEA treatment was the induction of transient elevations in urinary glutaric acid (GTA) and adipic acid (ADA) excretion lasting up to 24 hr p.d. Most of the GTA was excreted in the 0-8 hr p.d. with maximal rates of 100-120 microM/hr for each rat occurring between 4 and 8 hr p.d. in animals treated with BEA or BEA-d4. GTA and ADA were shown to be of endogenous origin as there was no detectable incorporation of the 2H label into either compound following treatment of rats with BEA-d4. Following BEA-treatment there was an initial decrease in the levels of urinary citrate, succinate, 2-oxoglutarate and trimethylamine-N-oxide. A subsequent recovery of citrate and succinate was noted following the onset of medullary nephropathy. The abnormal urinary metabolite profiles were similar to that observed in the urine of humans with glutaric aciduria type II (an inborn error of metabolism) caused by a lack of mitochondrial fatty acyl coenzyme A dehydrogenases indicating that BEA or its metabolites have similar metabolic consequences. The BEA metabolite aziridine was detected by 1H and 2H NMR spectroscopy of the urine 8 hr p.d. together with BEA itself and two novel metabolites 2-oxazolidone (OX) and 5-hydroxy-2-oxazolidone (HOX). The formation of OX requires the reaction of BEA with endogenous bicarbonate followed by a cyclisation reaction eliminating HBr. Dosing rats with authentic OX resulted in the excretion of HOX but did not cause glutaric or adipic aciduria indicating that either aziridine or BEA itself was responsible for the presumed defect in mitochondrial metabolism.

Comparative studies on the nephrotoxicity of 2-bromoethanamine hydrobromide in the Fischer 344 rat and the multimammate desert mouse (Mastomys natalensis)

Renal papillary necrosis (RPN) was induced in Fischer 344 (F344) rats (n=4) using 2-bromoethanamine hydrobromide (BEA) dosed at 150 mg/kg, and in multimammate desert mice (Mastomys natalensis) at 150 and 250 mg/kg (n=4) per group). Control rats andMastomys were dosed with 0.9% saline (n=4 per group). Urine was collected at regular intervals for up to 4 days post-dosing and analysed for low MW metabolites using high resolution1H NMR spectroscopy. The urinary activity of lactate dehydrogenase, γ-glutamyl transpeptidase and alkaline phosphatase was determined using conventional biochemical assays. On termination, histopathological examination of papillae was performed showing the development of extensive lesions in F344 rats at 150 mg/kg BEA.Mastomys appeared much more resistant to BEA and showed normal renal histology at 150 mg/kg and patchy lesions at 250 mg/kg BEA. Enzyme analysis of control urine showed F344 rats to have > 1000% higher γ-glutamyl transpeptidase activity thanMastomys.1H NMR spectroscopic analysis showed that BEA caused a substantial decrease in urinary concentrations of succinate and citrate (0–24 h p.d.) and an increase in creatine (0–96 h p.d.) in both animal models. A decrease in the urinary concentration of 2-oxoglutarate with a subsequent increase by 72–96 h p.d. was also noted in both animal models. Glutaric and adipic aciduria were also induced in both F344 rats andMastomys 0–24 h post-BEA treatment, indicative of an enzyme deficiency in the acyl CoA dehydrogenases. Urinary taurine levels were elevated inMastomys following the administration of BEA, indicating some degree of liver toxicity. Urinary taurine was not elevated in F344 rats following BEA administration, demonstrating further species difference in BEA toxicity.

Quantitative structure-metabolism relationships (QSMR) using computational chemistry: pattern recognition analysis and statistical prediction of phase II conjugation reactions of substituted benzoic acids in the rat

1. Quantitative relationships between molecular physico-chemical properties of 22 substituted benzoic acids and the extent of excretion of their metabolites in rat urine have been investigated using computational chemistry and multivariate statistics. 2. A data set of 34 theoretically derived physico-chemical descriptors calculated was used to classify the benzoic acids according to their predominant urinary metabolic fate. 3. Quantitative structure-metabolism relationships were obtained by linear regression using combinations of physico-chemical descriptors allowing the prediction of % urinary excretion of glycine (r = 0.73) and glucuronide conjugates (r = 0.82) and % urinary excretion of the parent compound (r = 0.91).

1H NMR Spectroscopic and Histopathological Studies on Propyleneimine-Induced Renal Papillary Necrosis in the Rat and the Multimammate Desert Mouse (Mastomys natalensis)

The renal papillary toxin, propyleneimine (PI), was administered at 20 or 30 microliters/kg i.p. to male Sprague Dawley (SD) rats (n = 5), Fischer 344 (F344) rats (n = 4), and to multimammate desert mice (Mastomys natalensis, n = 4). Urine was collected at time points up to 4 days p.d. and the toxicological response of the different animal models to PI compared using 1H NMR spectroscopy of urine, renal histopathology, and urinary assays for alkaline phosphatase (ALP), lactate dehydrogenase (LDH), and gamma-glutamyl transpeptidase (gamma GT). The renal papillae of both F344 and SD rats showed extensive necrotic lesions 4 days post-dosing and in some cases sloughing of the papilla. However, only slight renal papillary necrosis (RPN) was observed in Mastomys treated with 20 microliters/kg PI and, although slight to moderate damage was observed at 30 microliters/kg, PI-treated Mastomys showed substantially less RPN than either group of PI-treated rats. 1H NMR urinalysis showed that PI treatment caused a decrease in the urinary concentrations of succinate (0-24 hr p.d.) and citrate (24-48 hr p.d.) and an increase in creatine (0-48 hr p.d.) in all animal models. Trimethylamine-N-oxide (24-48 hr) and 2-oxoglutarate concentrations decreased initially following the administration of PI and then rose above control levels. The 1H NMR-detected urinary biochemical effects of PI in all three models were similar. However, taurine concentrations were elevated in the urine of Mastomys following PI treatment, perhaps indicating a degree of liver damage, whereas taurinuria was not seen in either SD or F344 rats. These observations are discussed in relation to the potential mechanism of PI-toxicity.