Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Total parenteral nutrition (TPN²) is a therapeutic intervention designed to provide sufficient calories and nitrogen to sustain patients who are unable to consume adequate nourishment by mouth and are therefore at risk of developing malnutrition. Unfortunately, alterations in both hepatic and intestinal function accompany therapy with TPN. Bypassing the intestine and the processes involved in nutrient absorption lead to intestinal hypoplasia and both absorptive and immunological hypofunction characterized by decreases in mucosal mass (DNA, RNA, and protein content), intestinal enzymes, villus size, and mitotic index (Hughes et al., 1980; O'Dwyer et al., 1989). These intestinal effects are similar to those observed in starvation. Gut barrier function, critical in preventing bacterial translocation, is also compromised by TPN (Kudsk et al., 1983). Hepatobiliary complications of TPN include fat accumulation, steatosis, cholestasis, cholelithiasis, elevated serum transaminase activities, and serum bilirubin (Quigley et al., 1993) and may be related to TPN-induced changes in intestinal function (Grant et al., 1977; Leaseburge et al., 1992). The etiology of TPN-associated cholestasis is poorly understood and probably due to multiple factors, including lack of enteral feeding and possible toxicity from methionine. TPN-associated cholestasis remains a critical problem for infants with intestinal failure (Moss and Amii, 1999). TPN depresses both hepatic oxidative (Knodell et al., 1989; Raftogianis et al., 1996a,b) and conjugative (Raftogianis et al., 1996a,b) biotransformation.

Glutamine, the most abundant free amino acid in the body, is the preferred fuel for enterocytes and seems to be an essential amino acid for intestinal function (Platell et al., 1993; Horvath et al., 1996). However, glutamine is chemically unstable and therefore not included in standard TPN formulations. Administration of parenteral (Platell et al., 1993) or enteral (Horvath et al., 1996) nutrition without glutamine produces intestinal atrophy and ulcerations in laboratory animals. Glutamine supplementation prevents this intestinal atrophy. Moreover, glutamine protects the gut mucosa from injury due to experimental endotoxemia (Chen et al., 1994) or from 5-fluorouracil-induced toxicity (Bai et al., 1996). Glutamine-supplemented TPN diminished the intestinal bacterial translocation associated with TPN (Burke et al., 1989; Bai et al., 1996). In rats, parenteral glutamine decreased the TPN-induced lithogenic effects of hepatic bile (Li and Stahlgren, 1995). Clinically, glutamine-supplemented TPN decreased the incidence of sepsis in low-birth-weight infants (Neu et al., 1999) and improved intestinal function and shortened the hospital stay in adults compared with patients receiving TPN without glutamine (Morlion et al., 1998).

The decrease in drug-metabolizing enzymes produced by parenteral nutrition may be related, in part, to TPN-induced changes in intestinal function. This led us to test the hypothesis that the addition of glutamine to parenteral nutrition, sufficient to reverse the TPN-induced intestinal effects, would prevent the TPN-related depression in hepatic P450 activity. The effects of administering TPN with glutamine on hepatic CYP2C and CYP3A activities were examined in a physiological relevant, chronically catheterized rat model (Uhing and Kimura, 1995; Raftogianis et al., 1996a). Hepatic CYP2C and CYP3A activities were selectively probed by quantifying the formation rates of relevant monohydroxy metabolites of testosterone and midazolam (MDZ) in microsomes from rats receiving TPN with and without glutamine and in chow-fed surgical controls.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. Midazolam, 1'-hydroxymidazolam, and 4-hydroxymidazolam were gifts of Hoffmann-La Roche (Nutley, NJ and Basel, Switzerland). -NADPH and L-glutamine were purchased from Sigma Chemical Co. (St. Louis, MO) and used as received. Testosterone, androstenedione, and 2-, 2-, 6-, and 16-hydroxytestosterone were obtained from Steraloids, Inc. (Newport, RI). Xylazine was obtained from Butler Co. (Columbus, OH), and ketamine was obtained from the Parke-Davis Division of Warner Lambert Company (Morris Plains, NJ). HPLC-grade solvents and all other reagents were obtained from Fisher Scientific (Pittsburgh, PA).

Animal Preparation. Male Sprague-Dawley rats weighing 225 to 250 g were obtained from Harlan Bioproducts for Science(Indianapolis, IN) and allowed free access to rat chow and water. The animals were housed on shaved bedding for 5 days under a 12-h light/dark cycle before surgery. All experimental procedures conformed to the guidelines promulgated by the National Institutes of Health and Purdue University, and the study was approved by the Institutional Animal Care and Use Committee.

TPN Administration and Glutamine Supplementation. Surgically prepared animals were divided into four groups and received 1) TPN without glutamine, 2) TPN plus 0.1% (w/v) glutamine, 3) TPN plus 4.5% (w/v) glutamine, or 4) ad libitum rat chow. Animals receiving TPN were weaned onto the full-strength solution. On postop day 1, animals received 10% dextrose in lactated Ringer's solution containing 1 U/ml heparin at a constant rate of infusion of 1.0 ml/h (Harvard model 22 infusion pump; Harvard Apparatus, South Natick, MA). On postop day 2, the infusion rate was increased to 2.0 ml/h. On day 3, full-strength TPN [25% dextrose, 5% Aminosyn II (Abbott Laboratories, North Chicago, IL) in a balanced electrolyte solution (sodium, 100 mEq/l; chloride, 80 mEq/l; potassium, 8 mEq/l; magnesium and phosphate, 13 mEq/l) with vitamins (2.5 ml/l M.V.I.; Astra Pharmaceutical Products, Inc., Westborough, MA) and 1 U/ml heparin] was infused at a constant rate of 2.15 ml/h. The TPN solution contained 1.1 kcal/ml, providing the rats with approximately 220 to 250 kcal/kg/day. Aminosyn II is a glutamine-free amino acid solution. Fresh TPN solutions were prepared every 48 h and sterilized during administration using an inline 0.45-µm filter (Acrodisc; Gelman Sciences, Ann Arbor, MI). The two groups of glutamine-treated rats received TPN plus 0.1 or 4.5% (w/v) L-glutamine (L-2-aminoglutaramic acid). The three groups of rats received TPN solutions that were isocaloric but were not isonitrogenous.

Preparation of Microsomes, Midazolam and Testosterone Hydroxylase Assays. Before the preparation of hepatic microsomes, the infusion of TPN was discontinued, and rats were lightly anesthetized with carbon dioxide and immediately sacrificed by decapitation. The livers were rapidly perfused in situ with cold isotonic saline. Liver microsomes were prepared by differential centrifugation (Franklin and Estabrook, 1971) and stored at 80°C in 50 mM Tris-chloride, pH 7.4, in 250 mM sucrose until assayed for activity. The protein concentration of microsomal fractions was determined colorimetrically (Lowry et al., 1951). Microsomal CYP3A activity was assessed using the formation rates of 1'- and 4-hydroxymidazolam and 2- and 6-hydroxytestosterone. CYP2C activity was assessed from the formation rates of 2- and 16-hydroxytestosterone.

HPLC Determination of Midazolam and Testosterone. Microsomes incubated with MDZ were processed using a liquid-liquid extraction technique, as described previously (Gorski et al., 1994). Following evaporation of the solvent, the residue was reconstituted with 200 µl of mobile phase (acetonitrile, methanol, and 20 mM ammonium acetate, pH 7.3; 40:20:40) and a portion was injected onto an HPLC column. MDZ, 4-OH MDZ, 1'-OH MDZ, and desmethyldiazepam were separated using a Phenomenex Luna C₁₈ column (5 µ × 4.6 mm × 150 mm) and a Brownlee RP-18 guard column. The mobile phase was delivered at a flow rate of 1 ml/min, and the eluent was delivered to a mass spectrometer (Navigator; Thermo Finnigan, San Jose, CA). The atmospheric pressure chemical ionization probe was run in the positive ion mode with source and probe temperatures of 200°C and 550°C, respectively. MDZ and desmethyldiazepam were detected in the selected ion-recording mode at m/z 326 and 271, respectively. The m/z for the 4-OH and 1'-OH MDZ was set at 342.3. The limit of quantification was 0.25 ng/ml for MDZ and metabolites.

Analysis of Kinetic Data and Statistics. The data for each microsomal incubation represent the mean of duplicate assays. Untransformed kinetic data from the midazolam incubations were fitted to the following equation:
for a single site, nonallosteric Michaelis-Menten enzyme model by nonlinear least-squares regression (WinNonLin v 1.1; Pharsight, Mountain View, CA). The appropriateness of the fit and the weighting schemes were determined by the visual inspection of residual patterns, residual sums of squares, and precision of the parameter estimates. Midazolam kinetics were performed using microsomes from a representative individual animal from each of the treatment groups. Testosterone metabolite data are reported as mean ± S.D. The effects of TPN with and without glutamine on the hydroxytestosterone metabolite formation rates were detected by single-factor analysis of variance, and the groups responsible for an effect were identified by Tukey's multiple comparison test. Differences were judged to be due to TPN or glutamine supplementation rather than chance variation when P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Midazolam Metabolism by Rat Liver Microsomes. The total body weight and liver weight for the animals from each nutrient treatment group are summarized in Table 1. Animals typically lose 10 to 15% of preoperative body weight after surgery. TPN provided sufficient calories for the animals to regain and maintain their initial body weight over the course of the experiment. The liver weight was decreased in rats receiving TPN compared with chow-fed controls, and this reached statistical significant only in animals receiving the high-dose glutamine. The kinetic data were normalized to microsomal protein (i.e., nanomoles per minute per milligram). When analyzed per whole liver or per gram of liver, the results and conclusions were unchanged.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Kinetics of formation of 4-OH midazolam () and 1'-OH midazolam () in rat liver microsomes from a chow-fed control. A, plot of formation rate versus midazolam concentration. B, plot of the transformation of the Michaelis-Menten kinetics to the Lineweaver-Burk equation.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Kinetics of formation of 4-OH midazolam () and 1'-OH midazolam () in rat liver microsomes from a rat receiving TPN. A, plot of formation rate versus midazolam concentration. B, plot of the transformation of the Michaelis-Menten kinetics to the Lineweaver-Burk equation.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Kinetics of formation of 4-OH midazolam () and 1'-OH midazolam () in rat liver microsomes from a rat receiving TPN plus 4.5% (w/v) glutamine. A, plot of formation rate versus midazolam concentration. B, plot of the transformation of the Michaelis-Menten kinetics to the Lineweaver-Burk equation.

View this table: [in this window] [in a new window]   TABLE 2 Effect of TPN and TPN plus glutamine on the formation of 4-hydroxymidazolam and 1'-hydroxymidazolam by rat hepatic microsomes Male Sprague-Dawley rats received total parenteral nutrition for 7 days containing 0, 0.1 or 4.5% (w/v) glutamine. Hepatic microsomes were prepared as described under Materials and Methods. Duplicate incubations were performed at MDZ concentrations ranging from 1 to 1000 µM incubated with -NADPH for 1 min at 37°C in microsomes from a representative animal from each nutrient group. Kinetic parameters were determined by nonlinear least-squares regression.

cDNA-Expressed CYP3A1 and CYP3A2. The hydroxymetabolite formation rate versus substrate concentration plot and the transformed Lineweaver-Burk plots for CYP3A1 and CYP3A2 are shown in Figs. 4 and 5, respectively. cDNA-expressed CYP3A2 preferentially formed 4-OH MDZ over 1'-OH MDZ, whereas CYP3A1 enzyme formed more 1'-OH MDZ compared with 4-OH MDZ. The K_m for the formation of the 4-OH MDZ was 31.0 and 30.1 µM by CYP3A1 and CYP3A2, respectively. The K_m for the formation of the 1'-OH MDZ was 24.7 and 8.7 µM by CYP3A1 and CYP3A2, respectively. The V_max values for the formation of 4-OH MDZ by CYP3A2 and CYP3A1 were 8.43 and 1.09 nmol/min/mg, respectively. The V_max values for the formation of 1'-OH MDZ by CYP3A2 and CYP3A1 were 1.66 and 1.16 nmol/min/mg, respectively. The intrinsic clearance (V_max/K_m) for 4-OH MDZ formation by CYP3A2 and CYP3A1 was 280.1 and 35.2 µl/min/mg of protein, respectively, and for 1'-OH MDZ was 67.2 and 133.3 µl/min/mg.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Kinetics of formation of 4-OH midazolam () and 1'-OH midazolam () in cDNA-expressed CYP3A1 microsomes containing cDNA-expressed rat P450 reductase and human cytochrome b5. A, plot of formation rate versus midazolam concentration. B, plot of the transformation of the Michaelis-Menten kinetics to the Lineweaver-Burk equation.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Kinetics of formation of 4-OH midazolam () and 1'-OH midazolam () in cDNA-expressed CYP3A2 microsomes containing cDNA-expressed rat P450 reductase and human cytochrome b5. A, plot of formation rate versus midazolam concentration. B, plot of the transformation of the Michaelis-Menten kinetics to the Lineweaver-Burk equation.

Testosterone Metabolism by Rat Liver Microsomes. The formation rates for 2-OH and 16-OH testosterone (CYP2C) and 2-OH and 6-OH testosterone (CYP3A) from rats in the four nutrient treatment groups are summarized in Table 3. Compared with chow-fed controls, the CYP3A-mediated hydroxylation of testosterone was significantly decreased in all rats receiving TPN. In rats receiving TPN with 4.5% glutamine, the formation rate for 2-OH and 6-OH testosterone was significantly increased compared with rats receiving TPN alone or TPN supplemented with 0.1% glutamine. A similar pattern was observed for the formation rates of 2- and 16-OH testosterone. The formation rates of 2-OH and 16-OH testosterone were highly correlated (Fig. 6), as were the formation rates of 2-OH and 6-OH testosterone (Fig. 7). Conversely, the formation rates of 2-OH testosterone with either 2-OH (r = 0.49) or 6-OH testosterone (r = 0.61) were weakly correlated. Similarly, the formation rates of 16-OH testosterone with either 2-OH (r = 0.38) or 6-OH testosterone (r = 0.49) were only weakly correlated (graphs not shown).

View this table: [in this window] [in a new window]   TABLE 3 Effect of TPN and TPN plus glutamine on the formation rate of 2-, 2-, 6-, and 16-hydroxytestosterone in rat hepatic microsomes Male Sprague-Dawley rats received TPN for 7 days containing 0, 0.1 or 4.5% (w/v) glutamine or allowed ad lib rat chow. Hepatic microsomes were prepared as described under Materials and Methods. Duplicate incubations were performed at testosterone concentrations of 250 µM with -NADPH for 10 min at 37°C.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Relationship between the formation rate of 2-OH testosterone and 16-OH testosterone in microsomes in rats receiving TPN with 0 glutamine (, n = 7), TPN with 0.1% glutamine (, n = 8), TPN with 4.5% glutamine (, n = 6), and chow-fed controls (, n = 5). R2 = 0.809; P < 0.01.

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Relationship between the formation rate of 2-OH testosterone and 6-OH testosterone in microsomes in rats receiving TPN with 0 glutamine (, n = 7), TPN with 0.1% glutamine (, n = 8), TPN with 4.5% glutamine (, n = 6), and chow-fed controls (, n = 5). R2 = 0.749; P < 0.01.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The current study was designed to evaluate whether the addition of glutamine to TPN prevented the depression in hepatic P450 activity that is a consequence of TPN. CYP2C and CYP3A are the principal hepatic drug-metabolizing enzymes in male rats, and the formation of monohydroxy metabolites of midazolam and testosterone have been used to profile these enzyme activities.

TPN produces significant intestinal and hepatic pathologies, including depression of hepatic drug metabolism. Ross et al. (1982) first reported that 7 days of lipid-free TPN decreased rat hepatic P450 content and cytochrome b₅ concentration. These investigators subsequently showed that TPN decreased the in vivo elimination of antipyrine by 73% (Ross et al., 1983). Knodell and colleagues (1989) found that the administration of TPN to rats for 7 days significantly reduced the constitutive amounts and activities of CYP3A and CYP2C11 but not of CYP2A1 and CYP2C6. Raftogianis et al. (1996a,b) found that administration of TPN, with and without lipid calories, to rats for 10 to 14 days significantly reduced P450 activity and sulfate and glucuronide conjugation. The present study found that CYP2C activity decreased about 35% in rats receiving TPN, but in contrast to Knodell et al. (1989), this did not reach statistical significance.

Glutamine is a five-carbon amino acid with two amino moieties and is the most abundant amino acid in plasma. Glutamine is an important factor in nitrogen transport, in intermediary metabolism, and as a cosubstrate for the formation of glutathione (Labow and Souba, 2000). Glutamine is also an important source of fuel for enterocytes. Glutamine depletion can occur during such catabolic states as surgery, trauma, or sepsis. Standard TPN solutions are not formulated with glutamine because of the chemical instability of glutamine. Thus, TPN administration routinely produces glutamine depletion, leading to intestinal mucosal atrophy and subsequent gut and liver hypofunction. Several investigators have shown in animal studies that the TPN-induced changes in gut physiology, including increased bacterial translocation (Alverdy and Burke, 1992), can be prevented by administering glutamine-enriched parenteral solutions (Burke et al., 1989; O'Dwyer et al., 1989). In addition, the effects of glutamine seem to be dose-dependent. Platell et al. (1993) found that reversal of the TPN-induced small-bowel atrophy in rats required TPN concentrations of glutamine above 1.5 to 2%. At these levels, glutamine also prevented the hepatic steatosis and biliary lithogenic effects of TPN, probably by decreasing hepatic lipid accumulation (Grant and Snyder, 1988) and stimulating glucagon secretion (Li and Stahlgren, 1995), thereby decreasing the elevated portal insulin to glucagon ratio produced by continuous TPN.

CYP3A2 and CYP2C11 are the major P450 proteins found in rat liver. Both CYP3A1 and CYP3A2 are expressed in the liver; however, CYP3A2 is considered the predominant hepatic form in male rats (Park et al., 1986; Cooper et al., 1993), whereas CYP3A1 is more predominately expressed in extrahepatic tissues (Debri et al., 1995). CYP3A1 is the only CYP3A expressed in the intestine (Zhang et al., 1996). 2-OH testosterone is mainly formed by CYP3A, whereas 6-OH testosterone is formed by CYP1A1, CYP2A2, CYP2C13, and CYP3A (Waxman, 1988). The metabolite formation rates of 2-OH and 6-OH were strongly correlated suggesting that these two metabolites are formed mainly by CYP3A. MDZ is almost exclusively metabolized by CYP3A in rats. The V_max ratio of the 4-OH MDZ to 1'-OH MDZ ranged from 3.0 to 4.5 in rat microsomes. The V_max ratios of 4-OH MDZ to 1'OH MDZ for the cDNA-expressed CYP3A2 and CYP3A1 were 5.0 and 0.6, respectively, which is consistent with CYP3A2 being the predominant CYP3A expressed in rat liver. 2-OH and 16-OH testosterone are the major metabolites of CYP2C11, and the formation rates for these metabolites were also highly correlated.

Several groups have previously reported apparent K_m values for the rat hepatic microsomal metabolism of midazolam (Ghosal et al., 1996; Takedomi et al., 1998; Higashikawa et al., 1999; Yamano et al., 1999) In the present study, we found linear Lineweaver-Burk kinetics for the formation of both 4-OH and 1-OH midazolam using 1-min incubations. Ghosal et al. (1996) incubated microsomes for 15 min and detected linear kinetics for the formation of the monohydroxy metabolite and nonlinear Michaelis-Menten formation of the 1',4-dihydroxy metabolite. K_m values for the formation of 1'OH and 4-OH MDZ reported by Ghosal et al. (1996) were similar to our values using the cDNA-expressed enzymes but lower compared with microsomes from rats receiving TPN or chow-fed control. Iga and colleagues (Takedomi et al., 1998; Yamano et al., 1999) and Higashikawa et al. (1999) measured the rate of midazolam disappearance and reported a midazolam K_m in the range of 6 to 8 µM using incubation times ranging from 30 s to 5 min. Overall, the differences in K_m values may reflect differences in nonspecific protein binding, the range of substrate concentrations studied, surgical treatment, the duration of the incubation times, and/or techniques of data fitting and analyses.

In conclusion, the present study demonstrates that supplementation of TPN with glutamine can largely prevent the TPN-induced depression in hepatic P450 activity. The inclusion of 4.5% glutamine into the TPN infusion prevented the depression in CYP3A activities as characterized by 2-OH and 6-OH testosterone formation rates and 4-OH and 1'-OH MDZ kinetics. The relative lack of effect of TPN or glutamine supplementation on the V_max ratios of 4-OH MDZ to 1'-OH MDZ and the testosterone metabolite correlations suggests that the effects of glutamine is not enzyme-selective. The addition of glutamine to TPN is important for the maintenance of gut structure and function. This, in turn, prevents the hepatotoxic effects of parenteral nutrition, including the depression of hepatic drug metabolism.