Search

Background: Neonatal total parenteral nutrition (TPN) is associated with animals with low glucose tolerance, body weight, and physical activity at adulthood. The early life origin of adult metabolic perturbations suggests a reprogramming of metabolism following epigenetic modifications induced by a change in the pattern of DNA expression. We hypothesized that peroxides contaminating TPN inhibit the activity of DNA methyltransferase (DNMT), leading to a modified DNA methylation state., Methods: Three groups of 3-d-old guinea pigs with catheters in their jugular veins were compared: (i) control: enterally fed with regular chow; (ii) TPN: fed exclusively with TPN (dextrose, amino acids, lipids, multivitamins, contaminated with 350 ± 29 μmol/l peroxides); (iii) H2O2: control + 350 μmol/l H2O2 intravenously. After 4 d, infusions were stopped and animals enterally fed. Half the animals were killed immediately after treatments and half were killed 8 wk later (n = 4-6 per group) for hepatic determination of DNMT activities and of 5'-methyl-2'-deoxycytidine (5MedCyd) levels, a marker of DNA methylation., Results: At 1 wk, DNMT and 5MedCyd were lower in the TPN and H2O2 groups as compared with controls. At 9 wk, DNMT remained lower in the TPN group, whereas 5MedCyd was lower in the TPN and H2O2 groups., Conclusion: Administration of TPN or H2O2 early in life in guinea pigs induces a sustained hypomethylation of DNA following inhibition of DNMT activity.

Introduction: In preterm neonates, peroxides contaminating total parenteral nutrition (TPN) contribute to oxidative stress, which is suspected to be a strong inducer of hepatic complications related to prematurity. Recently, others reported that hexapeptides derived from human milk (HM) exerted free radical-scavenging activities in vitro. Therefore, the aim of this study was to assess the capacity of these hexapeptides to limit the generation of peroxides in TPN and to prevent TPN-induced hepatic oxidative stress., Methods: At 3 d of life, guinea pigs were infused, through a catheter in jugular vein, with TPN containing or not peptide-A (YGYTGA) or peptide-B (ISELGW). Peroxide concentrations were measured in TPN solutions, whereas glutathione, glutathionyl-1,4-dihydroxynonenal (GS-HNE) and mRNA levels of interleukin-1 (IL-1) and tumor necrosis factor-α (TNFα) were determined in liver after 4 d of infusion., Results: The addition of peptide-A to TPN allowed a reduction in peroxide contamination by half. In vivo, peptide-A or peptide-B corrected the hepatic oxidative status induced by TPN. Indeed, both peptides lowered the hepatic redox potential of glutathione and the level of GS-HNE, a marker of lipid peroxidation. As compared with animals infused with TPN without peptide, the hepatic mRNA levels of IL-1 and TNFα were lower in animals infused with TPN containing peptide-A or peptide-B., Discussion: These results suggest that the addition of YGYTGA or ISELGW to TPN will reduce oxidative stress in newborns. The reduction in mRNA of two proinflammatory cytokines could be important for the incidence of hepatic complications related to TPN.

Failure to protect total parenteral nutrition (TPN) from ambient light exacerbates the generation of peroxides, which affects blood glucose and plasma triacylglyceride (TG) in neonates. Based on the concept that the origin of adult diseases can be traced back to perinatal life, it was hypothesized that neonatal exposure to peroxides may affect energy availability later in life. Three-day-old guinea pigs, fitted with a jugular catheter, were fed regular chow (sham) +/- i.v. 350 microM H2O2 (sham + H2O2) or nourished with light-protected TPN [TPN(-)L, 209 +/- 9 microM peroxides] or light-exposed TPN [TPN(+)L, 365 +/- 15 microM peroxides]. After 4 d, infusions were stopped and animals fed chow. Spontaneous ambulatory movements, fasting blood glucose, glucose tolerance, TG, hepatic activities of glucokinase, phosphofructokinase (key enzymes of glycolysis), and acetyl-CoA carboxylase (key enzymes of lipogenesis) were determined at 12-14 wk and compared by ANOVA (p < 0.05). Relative to sham, the animals from sham + H2O2, TPN(-)L and TPN(+)L groups had lower plasma TG explained for 36% by low phosphofructokinase activity; they had lower glucose tolerance, lower body weight, and lower physical activity. In conclusion, neonatal exposure to oxidant molecules such as peroxides has important consequences later in life on lipid and glucose metabolism leading to a phenotype of energy deficiency.

The i.v. lipid emulsion (LIP) is a source of oxidants, which may stimulate inflammation. Coadministration of parenteral multivitamins (MVP) with LIP prevents lipid peroxidation in light-exposed total parenteral nutrition (TPN). We hypothesized that this modality of TPN administration affects systemic inflammation, which may be modulated by exposure to oxygen. Premature infants were allocated to three TPN regimens: control regimen - MVP coadministered with amino acid/dextrose exposed to ambient light, LIP provided separately (n = 9) - LIP+MVP light exposed (LE): MVP coadministered with light-exposed LIP (n = 9) - LIP+MVP light protected (LP): MVP coadministered with light-protected LIP (n = 8). In LE and LP, amino acid/dextrose was provided separately. On reaching full TPN, infants were sampled for IL-6 and IL-8 in plasma and the redox potential of glutathione in whole blood (E, mV). Data were compared (ANOVA) in infants exposed to low (<0.25) versus high (> or =0.25) FiO2. Patients (mean +/- SD: birth weight 797 +/- 172 g; GA 26 +/- 1 wk) had similar clinical characteristics in TPN groups. Cytokine levels correlated positively (p < 0.01) with FiO2 and E. High FiO2 stimulated an increase (p < 0.01) in cytokines in control regimen, whereas these markers remained unaffected by oxygen in the LE and LP groups. The choice of a TPN admixture may have important consequences on the systemic inflammatory response triggered by an oxidant stress.

Photooxidation of multivitamin solutions results in the generation of peroxides. Because peroxides are associated with hepatic steatosis and fibrosis as well as cholestasis, we questioned whether multivitamins are implicated in hepatic complications of parenteral nutrition. Guinea pig pups were assigned to groups receiving intravenously either total parenteral nutrition, photo-protected or not, or a control solution (5% dextrose + 0.45% NaCl) supplemented with either a) multivitamins; b) photo-protected multivitamins; c) multivitamins without riboflavin; or d) peroxides (H(2)O(2), tert-butylhydroperoxide). After 4 d, liver was sampled for histology and isoprostane-F(2alpha) levels, a marker of radical attack. Multivitamins as well as total parenteral nutrition were associated with steatosis (scored 0-4), the severity of which was reduced (p < 0.05) by photo protection. Although H(2)O(2) is the major peroxide contaminating multivitamins, it did not induce steatosis scores different than the controls. Compared with controls, hepatic isoprostane-F(2alpha) content increased in animals infused with H(2)O(2) (p < 0.05), but not in those infused with Multi-12 pediatric multivitamins or total parenteral nutrition. Results suggest that peroxides and/or free radicals are not mediators of the induction of steatosis observed with infusion of photo-exposed multivitamins, as there was no correspondence between histologic findings and hepatic levels of isoprostanes. It is suspected that a component of the multivitamin solution becomes hepato-toxic after photo-exposure, as indicated by the protective effect observed when withdrawing riboflavin. Photo-oxidation of multivitamins might be the common link between reports involving amino acids, lipids, and light exposure in the ethiology of hepatic complications of parenteral nutrition.

The imbalance between high oxidant loads and immature antioxidant defenses is associated with long-term complications of prematurity. Glutathione is a central element among the antioxidants. Depletion of pulmonary glutathione accelerates the development of oxygen-induced lung injury in neonatal animal models. After the observation that newborn infants exposed to oxygen have low glutathione levels, a study was designed to test the hypothesis that in neonates from a species susceptible to oxygen toxicity, the lethal effect of hyperoxia is related to a low availability of substrates for glutathione production rather than an impairment in synthetic activity. One-day-old guinea pigs, randomly assigned to room air or oxygen (>95%), were fed by their mothers (n = 16) or i.v. by dextrose (n = 14) or by total parenteral nutrition (TPN, n = 20). After 3 d, glutathione and activities of enzymes involved in maintaining intracellular glutathione levels were determined in lungs and liver. The lethal effect of oxygen (p < 0.05) observed in animals without TPN was not related to glutathione depletion, as oxygen induced a 33% increase in lung glutathione, positively correlated (r2 = 0.35) with enhanced synthesis. With TPN, the animals were protected against the lethal effects of hyperoxia and lung glutathione increased by 67% in oxygen. The results suggest that the glutathione demand by the lungs in the presence of an oxidant stimulus was met by the increased (p < 0.001) hepatic production supported by TPN. Under hyperoxic conditions, early nutritional support is of vital importance.

In the presence of light, a multivitamin preparation is the main source of peroxides in solutions of total parenteral nutrition (TPN). This preparation contains two photosensitive products, 5'-phosphate flavin mononucleotide (FMN) and polysorbates (PS), as well as electron donors such as ascorbate (AH). We hypothesized that the admixture of FMN or PS with electron donors generates peroxides in TPN and alters the quality of nutrients. Using xylenol orange, peroxide concentrations were measured in solutions containing AH, FMN, and/or PS in water, a dextrose solution, an amino acid preparation, and a lipid emulsion. Thiol functions were evaluated by reduction of 5,5-dithiobis(2-nitrobenzoic acid) in the amino acid preparation. After 24-h light exposure, dextrose solutions with admixtures of AH + FMN or AH + FMN + PS generated peroxides at concentrations similar to those observed in a 1% multivitamin solution, and over three times higher than those observed with FMN, PS, or AH alone. However, in the presence of amino acids, FMN alone induced a generation of peroxides comparable to that observed with FMN + AH. In the lipid emulsion, peroxides increased over 3-fold in the presence of FMN or FMN + AH. The addition of catalase suggested that lipid peroxides and H2O2 were produced, and the loss of thiol function suggested that an oxidation of amino acids occurred. When exposed to light, FMN induces reactions with amino acids, polyunsaturated fatty acids, and even AH, altering the quality of nutrients. Paradoxically, AH without FMN has a protective effect on peroxide generation in TPN.

Antioxidant properties of bisulfite are used to stabilize parenteral amino acid solutions. After reports of adverse reactions to dietary sulfites, we evaluated whether the infusion of bisulfite had biologic effects on human vascular tissue. Because an endothelial oxidative injury can affect mediators of vasoreactivity, vascular pressure and prostaglandin production were studied in an intact human vein model infused with clinically relevant amino acid solutions differing only by their metabisulfite content (0 versus 300 mg/L). The amino acid solution containing bisulfite presented higher venous pressure (p < 0.01) and prostaglandin production (6-keto-prostaglandin F1 alpha, p < 0.01; and prostaglandin E2, p < 0.05). A hydroxyl radical-generating system added to the solutions did not modify the pressure readings, but it resulted in an overall decrease in prostacyclin production (p < 0.05). Despite this known inhibitory effect on oxidative challenge on prostaglandin I2 production, prostaglandins remained higher in the presence of bisulfite. The results suggest that the effect of bisulfite takes place before prostaglandin H2 synthesis in the eicosanoid cascade, whereas the oxidative challenge affects specifically the synthesis of prostaglandin I2, after prostaglandin H2, indicating that there is no interaction between bisulfite and the hydroxyl generating system. Bisulfite has local vascular effects on endothelial mediators, separate from its antioxidant properties.