Article pubs.acs.org/jpr
1
H NMR-Based Urinary Metabolic Profiling Reveals Changes in Nicotinamide Pathway Intermediates Due to Postnatal Stress Model in Rat
Alberta Tomassini,† Annabella Vitalone,‡ Federico Marini,† Giulia Praticò,† Fabio Sciubba,† Marta Bevilacqua,† Maurizio Delfini,∥ Antonella Di Sotto,‡ Silvia Di Giacomo,‡ Paola Mariani,§ Caterina L. Mammola,∥ Eugenio Gaudio,∥ Alfredo Miccheli,*,† and Gabriela Mazzanti‡ †
Department of Chemistry, ‡Department of Physiology and Pharmacology “V. Erspamer, §Department of General and Specialized Surgery “P. Stefanini”, and ∥Department of Anatomical, Histological, Forensic and Orthopedic Sciences, Sapienza University of Rome, P.le Aldo Moro 5, 00185 Rome, Italy S Supporting Information *
ABSTRACT: The maternal separation protocol in rodents is a widely recognized model of early life stress allowing acute and chronic physiological consequences to be studied. An 1H NMR-based metabolomic approach was applied to urines to evaluate the systemic metabolic consequences of maternal separation stress in female rats after the beginning of weaning and 4 weeks later when the rats were reaching adulthood. Furthermore, because maternal separation is considered as a model mimicking the inflammatory bowel syndrome, the lactulose/ mannitol test was used to evaluate the influence of postnatal maternal separation on gut permeability and mucosal barrier function by 1H NMR spectroscopy analysis of urine. The results showed no statistical differences in gut permeability due to maternal separation. The application of ANOVA simultaneous component analysis allowed the contributions of physiological adaptations to the animal’s development to be separated from the metabolic consequences due to postnatal stress. Systemic metabolic differences in the maternally separated pups were mainly due to the tryptophan/NAD pathway intermediate levels and to the methyladenosine level. Urinary NMR-based metabolic profiling allowed us to disentangle the metabolic adaptive response of the rats to postnatal stress during the animal’s growth, highlighting the metabolic changes induced by weaning, gut closure, and maturity. KEYWORDS: metabolic profiling, metabolomics, ASCA, maternal separation, NMR spectroscopy, urine
■
INTRODUCTION
The hypothalamic−pituitary adrenal (HPA) axis, the main stress axis in mammals, involves the release of the corticotrophin-releasing factor (CRF) from the hypothalamus, eventually leading to an increased level of cortisol in the systemic circulation.13 CRF has been implicated as a key mediator of the stress-induced effects on gut-intestinal function, thus demonstrating the existence of a communication network between the central nervous system and the GIT.14 This gut− brain axis is a bidirectional one in which GIT then influences the central nervous system through hormonal, neural, and immunologic signaling to maintain correct homeostasis.15−17 Although several functional and hormonal effects of maternal separation have been studied, less attention has been paid to the systemic metabolic consequences of such emotional stress in an individual. In this regard, there is a need to comprehensively understand the impact of a stressful event, either acute or chronic, on the
It is now widely accepted that early life stress represents a predisposing factor in the development of many psychiatric disorders, ranging from depression to anxiety and behavioral alterations.1−4 Several animal studies and some human epidemiological data suggest that fetal programming and neonatal events can have long-term effects, causing permanent dysfunctions or a higher susceptibility to different diseases, such as cancer, diabetes mellitus, and obesity, in adulthood.5,6 Moreover, stress has been linked to alterations of the brain− gut axis, leading to the development and aggravation of irritable bowel disease (IBD), a common dysfunction of the gastrointestinal tract (GIT).7 In this context, much research has utilized the maternal separation model in rodents as an extensively studied model of early stress that efficiently mimics some events, such as the alteration of the intestinal barrier function and microflora composition, visceral pain, and hypersensitivity, which characterize IBD.8−12 © 2014 American Chemical Society
Received: July 23, 2014 Published: October 9, 2014 5848
dx.doi.org/10.1021/pr500748r | J. Proteome Res. 2014, 13, 5848−5859
Journal of Proteome Research
Article
measurement of hematic parameters. Gut samples (i.e., duodenum, ileum, and colon) were also taken for histomorphological observation. All animal handling and experimental procedures were performed in accordance with the Italian Animal Welfare Legislation (Ministry of Health, 1992) and with the Directive 2010/63/EU of the European Parliament and of the Council on the protection of animals used for scientific purposes.
physiology of the whole organism so that a metabolomic approach may well be invaluable to characterize the stress response at a system level. High-resolution 1H NMR-based spectroscopy on biofluids, coupled to a multivariate statistical analysis, actually allows the metabolic changes induced by diseases, toxicity, or nutritional intervention to be highlighted at a systemic level.18−21 An elegant NMR-based metabolomic approach to evaluate the systemic metabolic consequences in plasma of rats in response to exposure to a maternal separation stress has been previously reported. The biochemical response to stress was characterized by decreased levels of total lipoproteins and increased levels of amino acids, glucose, lactate, creatine, and citrate, suggesting a systemic metabolic response to maternal separation involving changes in energy and muscle metabolism.22 The work carried out in the present study focuses on characterizing the biochemical variations in the metabolic profiles induced by the maternal separation model of neonatal stress in rats through urinary 1H NMR-based metabolomics. The urinary metabolic profiles as well as reflecting multiorgan interactions can provide direct and indirect information on gut microbiota metabolism and its interaction with the host metabolism, allowing the impact of maternal separation stress on the complex metabolic system to be evaluated. Furthermore, because stress in early life has been hypothesized to induce a susceptibility to the development of diseases in adult life, the same animals were investigated just after the beginning of weaning (PND26) and at PND 51, when the rats were reaching adulthood.
■
Urine Collection
For urine collection, female rats (n = 8/group) were moved to the metabolic cages for 6 h (from 9:00 a.m. to 3:00 p.m.), during which water was available ad libitum, while they did not receive any food. Urine was collected at the end of the 6 h in a tube containing sodium azide (0.05% final dilution) to avoid contamination and stored at −80 °C until 1H NMR spectroscopy analysis. Gut Paracellular Permeability Test
Intestinal permeability was assessed in urine by measuring the excretion of non metabolizable probes D-mannitol and lactulose (Sigma, St. Louis, MO) according to Pernet and colleagues,24 with slight modifications. 12 h fasted female rats (n = 8/group) treated by gavage with a mixture of 66 mg of lactulose and 50 mg of D-mannitol (in 2 mL of distilled water) were moved to the metabolic cages for 6 h (from 9:00 a.m. to 3:00 p.m.) during which water was available ad libitum, while they did not receive any food. Urine was collected at the end of the 6 h and was stored at −80 °C until assayed by 1H NMR spectroscopy. The ratio of urinary excretion of lactulose and mannitol is a useful test for assessing small intestinal permeability (SIP), and 1 H NMR spectroscopy has been shown to be a reliable analytical tool with reasonable sensitivity and specificity for the assessment of SIP.25,26 The NMR signal at 3.86 ppm was used for the quantification of mannitol, while the signal at 4.57 ppm was used for the quantification of lactulose in the urine, as previously described.25,26 To appreciate any age-related differences, these procedures were repeated on two different postnatal days (PND28 and PND53).
EXPERIMENTAL METHODS
Experimental Protocol and Animal Handling
Four pregnant female Sprague−Dawley rats (Charles River, Lecco, Italy) were individually housed. The animals were kept in an artificial 12 h light/dark cycle, at 23 ± 1 °C throughout the experiment. Dams were randomly assigned to two experimental groups, a control group in which the offspring were not separated from their dams and another one in which pups were separated from their dams (maternal separated group − MS), as shown in Suppl. Figure 1 in the Supporting Information. The day of birth was taken as postnatal day (PND) 0. On PND 1 litters were culled to 8 pups, and, as females appeared more sensitive to stress events than males,10,23 only females were used in the study (four female/ litter/group). Maternal separation was carried out according to previously validated methods.7,9 Briefly, in the MS group, dams and their offspring were daily separated from PND 4 to PND 20 for 180 min (9:00−12:00 a.m.). Each whole litter was removed from its home cage and placed in a new cage with new bedding. During the separation period, the pups were kept in a separate room in an incubator, ensuring the maintenance of controlled temperature (31 ± 1 °C). The litters of the control group were not handled and were left undisturbed. Food and water were available ad libitum throughout the experimental period, except for pups during maternal separation. Pups were weaned on PND 21 and then moved to individual cages. Body weight and food intake were monitored three times a week, throughout the experimental period. On PND56, all rats were sacrificed and blood samples, drawn by intracardiac puncture, were used for biochemical
Blood Collections and Clinical Chemistry
The blood samples were immediately centrifuged after collection and analyzed in a blind procedure, as previously described.27 Each serum sample was tested for concentration of glucose, cholesterol, triglycerides, HDL, urea, albumin, AST, ALT, gamma-GT, and alkaline phosphatase using a routine automated COBAS 6000 analyzer (Roche Diagnostics, Italy). Insulin levels were determined using a specific enzyme-linked immunoassay for rat serum (Demeditec Diagnostics, Kiel, Germany), and the plates were read on a MPT-DV990 (GDV, Italy) at an OD of 450 nm, according to the manufacturer’s instructions. The homeostasis model of assessment (HOMA) insulin resistance index was calculated using the following international formula: [fasting serum glucose (mmol/L) × fasting serum insulin (mU/L)/22.5]. Tissue Specimen Collections and Histomorphological Study
The histomorphological examination was performed on all of the removed gut samples. Gut fragments were obtained from duodenum (immediately after the pylorus sphincter), jejunum (30 cm beyond the pylorus sphincter), ileum (immediately before the ileocecal valve), and colon (5 cm above the anus) and immediately fixed in 10% buffered formalin at room 5849
dx.doi.org/10.1021/pr500748r | J. Proteome Res. 2014, 13, 5848−5859
Journal of Proteome Research
Article
operates by partitioning the overall variation present in the experimental data into the individual contributions induced by the controlled factors and their interactions and by analyzing the resulting matrices by simultaneous component analysis (SCA), as previously described.33 Accordingly, ASCA is performed by comparing the sum of squares of each of the effect matrices (which is taken as a measure of the effect of the design terms) with its distribution under the null hypothesis as nonparametrically estimated by means of permutation tests.34,35 The effect of the treatment on each age group was then also confirmed by partial least-squares−discriminant analysis (PLSDA).36 Cross-model validation using nine cancellation groups in the external and eight in the internal loops (20 independent runs with random splitting of the samples) together with permutation tests with 10 000 randomizations was used to test the significance of the model and the relevance of supposedly important metabolites, which were identified on the basis of the values of PLS-DA regression coefficients and variable importance in projection (VIP) scores. VIP expresses the contribution of the individual variable in the definition of the model. Because the mean of squared VIP scores is equal to 1, values >1 are considered to be significant variables. The comparison between maternal separated and control groups of each age (PND26 and PND 51, respectively) was performed by univariate unpaired Student’s t test or by Mann− Whitney U test. P values
