The Journal of Nutrition Vol. 128 No. 12 December 1998, pp. 2427-2434

Mitochondrial Lysine Uptake Limits Hepatic Lysine Oxidation in Rats Fed Diets Containing 5, 20 or 60% Casein^1,2

Kenneth P. Blemings^*, , Thomas D. Crenshaw^*, and Norlin J. Benevenga^{*, , 3}

Departments of ^* Department of Animal Sciences, University of Wisconsin Madison, Madison, Wisconsin 53706, USA and  Department of Nutritional Sciences, University of Wisconsin Madison, Madison, Wisconsin 53706, USA

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Sixty male Sprague-Dawley rats were randomly allotted to receive diets containing 5, 20 or 60% casein. Rats had access to the diet only during the initial 8 h of the daily 12-h dark period. Hepatic mitochondrial lysine uptake, lysine -ketoglutarate reductase (LKR) and saccharopine dehydrogenase (SacD) activities, and in vitro lysine oxidation (LOX) were measured 0, 6, 12, 18 and 24 h after the start of the dark period. Diurnal variation of mitochondrial lysine uptake was not detected (P > 0.10) although uptake varied 3-fold over 24 h. Mitochondrial lysine uptake was greater (P < 0.05) for rats fed diets containing 60% casein than for rats fed diets containing 5% casein. Diurnal variation of LKR was detected (P < 0.05) in rats fed diets containing 20 and 60% casein. Diurnal variation of SacD was detected (P < 0.05) in rats fed diets containing 60% casein. Increased casein consumption resulted in increased LKR and SacD activities (4- to 5-fold; P < 0.05). Diurnal variation of LOX was detected in rats fed diets containing 20 and 60% casein (P < 0.05). Increasing the casein concentration in the diet from 5 to 60% resulted in a 7-fold increase in LOX (P < 0.05). To make rate comparisons, LKR and SacD activities and LOX were predicted from a range of substrate concentrations (0.1 to 5.0 mmol/L). Overall, LKR and SacD were 6-107 times that of LOX, suggesting that, in liver, mitochondrial lysine uptake limits LOX.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Lysine is frequently the first limiting amino acid in human diets relatively high in grain [Bobwell et al. 1981, Food and Agriculture Organization 1985, National Research Council (NRC) 1989]. Therefore, improving efficiency of lysine use for protein synthesis offers benefits to human health. Lysine is also typically the first limiting amino acid in the usual corn and soybean meal diets of pigs (NRC 1988). Economic benefits from an increase in the efficiency of lysine use for protein synthesis in pigs in concert with a decrease in total excreted nitrogen offer additional incentives for understanding regulation of lysine oxidation (LOX).⁴

Little is known about regulation of lysine oxidative flux. Rat studies demonstrated that LOX occurs predominantly (94 to 100%) in liver (Miller 1962, Noda and Ichihara 1976), with a small fraction occurring in nervous tissue (Chang 1978a and 1978b). Reports on the role of kidneys in lysine catabolism are conflicting since some suggest little or no role (0 to 6%) (Miller 1962, Noda and Ichihara 1976) while others suggest a greater role (70%) (Forsberg and Austic 1986).

Increasing protein consumption by rats elevates activities of liver amino acid degrading and urea cycle enzymes (Anderson et al. 1968, Harper 1965, Schimke 1962). Several reports suggest enzymes involved in lysine degradation, lysine -ketoglutarate reductase (LKR) (Blemings et al. 1990, Chu and Hegsted 1976, Muramatsu et al. 1984) and saccharopine dehydrogenase (SacD) (Blemings et al. 1990) activities change in response to dietary protein intake as well. Moreover, increasing protein consumption increases liver homogenate LOX (Blemings et al. 1992, Soliman and Harper 1971).

Early reports suggested that the mitochondrion was involved in LOX (Higashino et al. 1965). Our previous work bore this out by demonstrating that the initial enzymes of saccharopine-dependent, lysine catabolism in rat liver are housed exclusively in the mitochondrial matrix (Blemings et al. 1994). The unique subcellular site of lysine catabolism raised the possibility that entrance into the mitochondrial matrix limits LOX.

Dietary studies such as those referenced above often neglect changes in activity and metabolism that occur during a 24-h light/dark cycle. These rhythmic changes in activity and metabolism are commonly referred to as diurnal variation. Evaluating diurnal variation is important for design of future experiments; if large fluctuations occur throughout a day, they may impair interpretation of treatment effects (Fuller and Snoddy 1968). No reports that specifically addressed diurnal variation of mitochondrial lysine uptake, LKR and SacD activities, or LOX were found. Therefore, the objective of this study was to determine the effect of diets containing 5, 20 and 60% casein on in vitro mitochondrial lysine uptake, homogenate LKR and SacD activities, and homogenate LOX at different times throughout a 24-h period. By making all four measurements in the same rat, one may compare rates and determine which step may be limiting for LOX.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Radiochemicals

Water and lysine.  [1-¹⁴C]L-lysine (a generous gift from Purina Mills Inc., St. Louis, MO) and ³H₂O were from American Radiolabeled Chemical Incorporated (St. Louis, MO). [U-¹⁴C]L-lysine was from NEN (Wilmington, DE). The radiochemical purity of both lysine tracers was checked via high performance liquid chromatography with a Beckman Ultrasphere C-18 column (5 µm packing in 4.6 mm × 25 cm column) with a 25% methanol (v/v%), 1 mmol/L sodium decylsulfonate, pH 3 mobile phase flowing at 0.6 mL/min, and detection with an in-line radioactivity detector (Flo 1-Radiomatic Instruments, Tampa, FL). No radiochemical impurities were detected in either lysine tracer.

Sucrose.  [U-¹⁴C]-sucrose was from NEN and was checked for radiochemical purity on a Bio-Rad (Richmond, CA) HPX-87P column with an 80°C water mobile phase flowing at 1 mL/min. Detection was by an in-line radioactivity detector. Glucose and fructose contaminants were detected in older sucrose preparations. Contaminated sucrose was first lyophilized and then treated with 2 units of hexokinase (from baker's yeast type C-300; Sigma Chemical Co., St. Louis, MO) in pH 8 buffer (50 mmol/L Tris, 13.3 mmol/L MgCl₂, 16.5 mmol/L ATP) to phosphorylate the glucose and fructose in the presence of 0.1 mmol/L nonradioactive sucrose. After 15 min, the solution pH was lowered with 1 mol/L HCl to about 2, and the mixture passed over a Dowex 1 (Cl form; Sigma) column in water. The sucrose passed through the column while the phosphorylated forms of glucose and fructose were retained. When the sucrose that passed through the Dowex 1 column was rechecked for radiochemical purity, no impurities were detected.

Chemicals.  Sodium decylsulfonate was obtained from Eastman Kodak (Rochester, NY). Silicone oils were from William F. Nye Incorporated (New Bedford, MA). Enzyme-grade sucrose was obtained from Schwarz-Mann (Orangeburg, NY). Perchloric acid was purchased from MCB Reagents (Cincinnati, OH). Magnesium chloride was obtained from Columbus Chemical Industries (Columbus, WI). Potassium chloride was from Mallinckrodt Incorporated (Paris, KY). Methanol, EDTA, potassium phosphate and monoethanolamine were acquired from Fisher (Fair Lawn, NJ). L-lysine HCl and methylcellosolve were procured from U.S. Biochemical (Cleveland, OH). Triton X-100 was purchased from J.T. Baker (Phillipsburg, NJ). All other reagents were from Sigma.

Animals and diets.  Sixty male Sprague-Dawley rats (Harlan Sprague-Dawley, Indianapolis, IN), weighing 75 to 100 g, were housed individually in wire-mesh cages in rooms maintained at ~24°C. Rats had access to their diets only during the initial 8 h of the 12-h dark period. Two rooms with opposite light cycles were used so that the dark cycle for some rats occurred during working hours. Based on treatment needs, rats were randomly allotted to diet and room. Feed intakes and body weights were recorded daily. Rats received diets containing 5, 20 or 60% casein (Table 1). Three rats from each group were killed by decapitation at 0, 6, 12, 18 or 24 h after the start of the dark cycle. To meet the experimental design requirements, 45 of the 60 rats were randomly selected for use in the in vitro lysine metabolism measurements. Animal care and conduct of experiments were approved by the University of Wisconsin-Madison, College of Agriculture and Life Sciences Animal Care Committee and were in accord with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NRC 1985).

Preparation of liver fractions.  Liver homogenate and mitochondria were prepared as previously described (Blemings et al. 1994). The respiratory control ratios (Greenawalt 1974) of mitochondria prepared in this manner were greater than or equal to 4 (data not shown), indicating that the mitochondria were intact.

Lysine uptake by rat liver mitochondria

Uptake protocol.  Centrifugal filtration was used to assess lysine uptake (LaNoue and Schoolwerth 1979, Palmieri and Klingenberg 1979). In a 1.8-mL microfuge tube, 0.7 mL of incubation media (density 1.00 kg/L) was layered on top of 0.5 mL of a silicone oil preparation (density 1.06 kg/L), which was itself layered on top of 0.15 mL of a buffer trap (density 1.13 kg/L). The silicone oil preparation was made by mixing 45 mL of DC 550 oil (125 centistokes) and 6.5 mL of DC 200 oil (1.5 centistokes) to obtain a viscosity which would allow recovery of rat mitochondria at the g force used. The buffer trap was a 1:1 mixture of homogenization buffer and glycerol.

Mitochondrial lysine uptake was measured at 37°C. The microfuge tubes containing the three layers were kept in a 37°C water bath (Model 50, Precision Scientific; Chicago, IL) prior to use. The microcentrifuges (Eppendorf 5415C; Brinkman Instruments Inc, Westbury, NY) were housed in a chamber constructed of 2-cm thick foil-faced foam insulation. The chamber (65 × 71 × 68 cm) was maintained at 37°C by a suspended 750-watt strip heater (35 × 6.4 × 0.95 cm) (Chromalox, Pittsburgh, PA) attached to a Rex C-10 Temperature Controller (RKC Syscon, Japan) which was also attached to a 700 JU thermocouple (Chromalox, Pittsburgh, PA) housed in the chamber.

Uptake was initiated by the addition of 50 µL of mitochondria (10 g protein/L based on a modified Biuret reaction; Layne 1957). Mitochondria were preincubated (5 min) in incubation media without lysine. Mitochondrial matrix lysine concentrations were assessed at 10, 15, 20, 30, 60, 120, 180 and 240 s. Complete recovery of mitochondria in the buffer trap was established in preliminary studies (data not shown).

Incubation media.  The incubation media contained mannitol, 120 mmol/L; sucrose, 70 mmol/L; HEPES, 10 mmol/L; PIPES, 10 mmol/L; succinate, 10 mmol/L; KCl, 50 mmol/L; rotenone, 1 µmol/L as a 0.3 mmol/L ethanolic solution; L-lysine, 1.07 mmol/L; in a 5 mmol/L potassium phosphate buffer pH 7.0. To obtain estimates of the rate of lysine uptake and the concentration of lysine in the mitochondrial matrix water, the incubation media (0.7 mL) contained either: i) ³H₂O (29.2 kBq) to estimate total water and [U-¹⁴C]L-lysine (2.92 kBq) to measure total lysine or ii) ³H₂O (29.2 kBq) to estimate total water and [U-¹⁴C] sucrose (2.92 kBq) to estimate extra matrix water since sucrose does not enter the mitochondrial matrix. Measurement of extra matrix water space is essential since much of the lysine (70%) recovered in the buffer trap is not in the matrix. The extra-matrix lysine is subtracted from the total lysine to obtain an estimate of matrix lysine. The matrix lysine concentration is calculated by relating matrix lysine to matrix water. Matrix lysine concentrations as a function of time were fit as described below for uptake rates, except matrix lysine was scaled to matrix water instead of mg of mitochondrial protein. This enabled us to estimate matrix lysine concentration as a function of time.

Calculation of mitochondrial lysine uptake rates.  Mitochondrial matrix lysine per mg of mitochondrial protein as a function of time was fit to a single exponential curve using a nonlinear, curve-fitting procedure (the Marquadt iterative procedure) in SAS (Cary, NC) with the model in Equation 1: where [lys]_t is the matrix lysine per mg of mitochondrial protein at time t, [lys] is the asymptomatic matrix lysine per mg of mitochondrial protein at infinite time, k is a fractional rate constant, and C is a constant (the lysine per mg of mitochondrial protein at zero time as determined by the model). After the initial parameters were fit, a derivative (with respect to time) of Equation 1 (Eq. 2) was used to calculate the initial rate of mitochondrial lysine uptake/mg of mitochondrial protein. The initial rate of mitochondrial lysine uptake was calculated from the parameter estimates and solution of the differentiated equation with time equal to 0. The coefficient of variation (CV) for this assay was ~25%.

A common concern with the centrifugal filtration method is metabolism of substrate after transport that would affect initial rate estimates. In preliminary studies (data not shown) to determine if this was a problem, mitochondria were lysed with detergent after the transport assay. By using Dowex-1 chromatography, we determined a step gradient to separate lysine, saccharopine and -aminoadipate. The radioactivity of the nonlysine eluates did not increase as a function of time, indicating metabolism of lysine during the short incubations was not confounding uptake estimates.

LKR (EC 1.5.1.8).  LKR activity was measured at 37°C as previously described (Blemings et al. 1994). LKR activity was determined in each homogenate and mitochondrial preparation in order to calculate mitochondrial recovery that varied from 60 to 70% in this work. By knowing the mitochondrial recovery, all measurements could be scaled back to a whole animal basis. The CV for this assay was less than 5%.

SacD (EC 1.5.1.9).  Homogenate SacD activity was measured at 37°C as previously described (Blemings et al. 1994). The CV for this assay was less than 5%.

LOX in rat liver homogenates.  LOX was determined in sleeve style (Wheaton, red rubber) stoppered glass t-tubes (7 × 1.2 cm) which were connected to scintillation vials containing 1 mL of a 1:2 mixture of ethanolamine:methylcellosolve to trap carbon dioxide. The assay system is similar to that used by others (Dancis et al. 1969, Soliman & Harper 1971). Incubations were started with 0.1 mL of liver homogenate. The pH of the assay system was 7, and concentration of reagents was: L-lysine, 1 mmol/L; -ketoglutarate, 15 mmol/L; HEPES, 10 mmol/L; MgCl₂, 3 mmol/L; EDTA, 0.2 mmol/L; NADPH, 0.5 mmol/L; sucrose, 39 mmol/L; mannitol, 157 mmol/L in a volume of 0.4 mL. Incubations also included 1.67 kBq of [1-¹⁴C]L-lysine. Recovery of ¹⁴CO₂ from lysine catabolism was time-linear over the 30-min incubation.

T-tubes were kept on ice until they were placed in a shaking water bath maintained at 37°C; the rate of shaking was 150/min. A preincubation time of 5 min was used before homogenate was added to start the incubation. The purpose of the preincubation was to allow the contents to equilibrate to 37°C. A heat-killed blank (10 min in boiling water) was used to correct for nonenzymatic production of ¹⁴CO₂. Incubations were terminated by injecting 0.25 mL of 35% perchloric acid into the incubation media through a rubber septum covering the t-tube. After acid injection, t-tubes remained in the bath for 2 h with shaking. Preliminary data indicated complete recovery of ¹⁴CO₂ by 2 h. After 2 h, t-tubes were disassembled, 13 mL of Biosafe (Research Products Inc., Mt. Prospect, IL) scintillation fluid was added to the trapping solution in the scintillation vial and the sample counts determined (Beckman LS 3801, Beckman Instruments, Fullerton, CA). The CV for this assay was less than 5%. Sufficient radioactivity was accumulated so that the error in counting was 2%.

Statistical analysis.  Three rats were killed at each time point. Analysis of variance (ANOVA) procedures appropriate for a randomized block design were used to test for diurnal variation of mitochondrial lysine uptake, LKR and SacD activities, and LOX. The general linear models procedure (GLM) of SAS (version 6.07) was used to analyze the model in which time and diet were considered as main effects. Because of significant diet by time interactions, one-way ANOVA procedures were used to fit time within a diet for each measured variable. Diurnal variation was considered significant when ANOVA resulted in a significant F-value during the 24-h period within a diet. Additionally the effect of room and diet on food intake, relative food intake (g food eaten/g body weight) and body weight were modeled with GLM. Average daily gain was estimated from the slope of the regression of body weight against day for each diet. In the event of a significant F-value, Tukey's test was used to identify differences. Differences were considered significant at P < 0.05.

	RESULTS

Abstract Introduction Methods Results Discussion References

Food intake, relative food intake and average daily gain.  Rats were given 9 d to adjust to the feed-and-light schedule. Rats were killed between 9 and 21 d after initiation of diet treatment. Within a diet, no significant effect of room on feed intake, relative feed intake or body weight was detected. Because no room effect was detected, data were pooled across rooms. The average daily food intake, average relative daily food intake and average daily gain for d 9 to 21 are shown in Table 2. The average daily gain is similar to reported values for rats treated in this manner (Peters and Harper 1985).

Lysine uptake by rat liver mitochondria.  Representative uptake data and predicted curves for rats consuming diets containing 5, 20 or 60% casein are shown for 1 mmol/L lysine in Figure 1. Lysine uptake by liver mitochondria was rapid and generally complete by 1 min. Using the matrix lysine concentration data, we determined that the model accurately predicted the matrix lysine concentration. Moreover, the model predicted the matrix lysine concentration at t = 0. Pooled across all rats, the matrix lysine concentration (mean ± SD) from 180-240 s was 3.5 ± 0.26 mmol/L (observed), 3.63 ± 0.17 (predicted) and lysine concentration at t = 0 was 1.41 ± 0.38 mmol/L (predicted). Our attempts to determine the K_m and V_max for lysine uptake were unsuccessful. Individual lysine uptake estimates utilizing the centrifugal filtration technique, especially at higher lysine concentrations, are not sufficiently repeatable to obtain a standard value. No effect (P > 0.10) of dietary protein concentration could be detected on matrix lysine concentration. This 3.5-fold (1 vs. 3.5 mmol/L) concentration of lysine into the mitochondrion is similar to the 3.2- to 6.4-fold reported for skeletal muscle (Bergstrom et al. 1974, Forsberg and Austic 1986) and 2.6-fold for kidney (Forsberg and Austic 1986) cells relative to the extracellular environment.

View larger version (20K): [in this window] [in a new window]   Fig 1. Mitochondrial matrix lysine concentration. Mitochondrial matrix lysine concentration was calculated from duplicate values measured at 10, 15, 20, 30, 60, 120, 180 and 240 s. The figure shows the mitochondrial matrix lysine concentration as a function of time for rat liver mitochondria prepared from single rats fed a 5 (A), 20 (B) or 60% (C) casein diet: () represent observed values and () represents the predicted values based on the model as described in the Materials and Methods section, A, Y = 2.84 (1  e0.0272x) + 0.293; B, Y = 2.06 (1  e0.0096x) + 0.993, C, Y = 5.06 (1  e0.0282x) + 0.231.

Mitochondrial lysine uptake was measured at 6-h intervals over a 24-h period (Fig. 2). No significant differences within a diet were detected, likely due to the large variation of the data within a time point which, in part, is due to the difficulty of obtaining the initial rate (i.e., pipetting and mixing of the mitochondria in the assay medium, the short time between mixing and sampling and the fit of the model). However, pooled across time, lysine uptake was greater (P < 0.05) in rats fed diets containing 60% casein compared to rats fed diets containing 5% casein. Averaged over the entire 24-h period, mitochondrial lysine uptake of rats fed diets containing 5, 20 and 60% casein was 0.19, 0.47 and 0.71 mmol · d¹ · 100 g body wt¹, respectively (Table 3).

View larger version (23K): [in this window] [in a new window]   Fig 2. Mitochondrial lysine uptake measured at 6-h intervals over a 24-h period for rats fed diets containing 5, 20 or 60% casein. Initial rates of mitochondrial lysine uptake were estimated from the duplicate values obtained at 10, 15, 20, 30, 60, 120, 180 and 240 s, using the curve-fitting routine described in the Materials and Methods section. The horizontal bars represent the light/dark period and diet access. The data are expressed as the mean mmol · d1 · 100 g body wt1 (BW), and the pooled SEM is shown for reference. Each point represents the mean of three rats. No significant time effect (P > 0.05) was detected in response to diet.

View this table: [in this window] [in a new window]   Table 3. Rates of mitochondrial lysine uptake, homogenate lysine -ketoglutarate reductase (LKR) and saccharopine dehydrogenase (SacD) activities and lysine oxidation (LOX)

LKR and SacD activities.  The LKR and SacD activities measured at 6-h intervals over a 24-h period are shown in Figures 3 and 4, respectively. Significant diurnal variation of LKR activity was detected in rats fed diets containing 20% and 60% casein. The V_max LKR activities (i.e., not scaled to 1 mmol/L substrate as shown in Table 3) averaged over the entire 24-h period of rats fed diets containing 5, 20 and 60% casein was 6.54, 11.16 and 34.20 mmol · d¹ · 100 g body weight¹, respectively. Significant diurnal variation of SacD activity was detected only in rats fed diets containing 60% casein. The V_max SacD activities (i.e., not scaled to 1 mmol/L substrate as shown in Table 3) averaged over the entire 24-h period of rats fed diets containing 5, 20 and 60% casein were 5.65, 6.86 and 21.60 mmol · d¹ · 100 g body weight¹, respectively.

View larger version (21K): [in this window] [in a new window]   Fig 3. Lysine -ketoglutarate reductase activity (LKR) measured at 6-h intervals over a 24-h period for rats fed diets containing 5, 20 or 60% casein. LKR was assayed as described in the Materials and Methods section. The horizontal bars represent the light/dark period and diet access. The data are expressed as the mean mmol · d1 · 100 g body wt1 (BW) and the pooled SEM is shown for reference. Each point represents the mean of three rats. A significant (P < 0.05) time effect (diurnal variation) was detected for rats fed diets containing 20 or 60% casein.

View larger version (20K): [in this window] [in a new window]   Fig 4. Saccharopine dehydrogenase activity (SacD) measured at 6-h intervals over a 24-h period for rats fed diets containing 5, 20 or 60% casein. SacD was assayed as described in the Materials and Methods section. The horizontal bars represent the light/dark period and diet access. The data are expressed as the mean mmol · d1 · 100 g body wt1 (BW) and the pooled SEM is shown for reference. Each point represents the mean of three rats. A significant (P < 0.05) time effect (diurnal variation) was detected for rats fed diets containing 60% casein.

View larger version (23K): [in this window] [in a new window]   Fig 5. Lysine oxidation (LOX) measured at 6-h intervals over a 24-h period for rats fed diets containing 5, 20 or 60% casein. LOX was assayed as described in the Materials and Methods section. The horizontal bars represent the light/dark period and diet access. The data are expressed as the mean mmol · d1 · 100 g body wt1 (BW) and the pooled SEM is shown for reference. Each point represents the mean of three rats. A significant (P < 0.05) time effect (diurnal variation) was detected for rats fed diets containing 20 and 60% casein.

View this table: [in this window] [in a new window]   Table 4. Comparison of the rates of mitochondrial lysine uptake, homogenate lysine -ketoglutarate reductase activity (LKR), saccharopine dehydrogenase activity (SacD), and lysine oxidation (LOX) across a range of substrate concentrations1

LOX.  LOX was measured in liver homogenates since preliminary experiments showed that homogenates oxidized about twice as much lysine per unit of LKR activity as isolated mitochondria. In preliminary experiments, the addition of other cofactors (NAD, PALP or CoA) did not stimulate LOX in homogenates (data not shown).

LOX measured at 6-h intervals over a 24-h period is shown in Figure 5. Diurnal variation of LOX was detected in rats fed diets containing 20 and 60% casein. A significant effect of dietary casein level on LOX was detected as previously reported (Blemings et al. 1992, Soliman and Harper 1971). Averaged over the entire 24-h period, LOX was 0.07, 0.24 and 0.55 mmol · d¹ · 100 g body weight¹ for rats fed diets containing 5, 20 and 60% casein, respectively (Table 3). Numerically, the oxidation rate was one-half to two-thirds of the uptake rate, and the difference is reflected by an accumulation of lysine in the mitochondrial matrix over time (Fig. 1). Even though lysine uptake and oxidation rates increased 3- to 7-fold with dietary casein concentration, the rates of uptake and oxidation at each dietary casein concentration are not different (P > 0.05; Table 3). Thus, as lysine uptake made substrate available to LKR in the matrix and as the matrix lysine concentration increased from an initial predicted concentration of 1.4 to a steady-state concentration of 3.5 mmol/L, the oxidation rate could increase depending on the availability of enzyme. The steady-state matrix lysine concentration of 3.5 mmol/L from 180 to 240 s across dietary treatments sustained a 5-fold increase in oxidation, LKR and SacD activities in an environment with an enzyme activity potential that was at least 10 or more times that of uptake. This is consistent with the notion that a balance of uptake and oxidation occurs at 3.5 mmol/L lysine and thus, uptake (transport) limits LOX.

In another effort to determine which steps might be rate limiting for lysine catabolism, mitochondrial lysine uptake, LKR and SacD activities and LOX rates were compared. This comparison was possible after both LKR and SacD activities were scaled to 1 mmol/L substrate (lysine and saccharopine) concentration by use of the Michaelis-Menten equation. The K_m of LKR for lysine used was 5 mmol/L based on findings in this system (Blemings, unpublished results). This K_m is higher than the reported value for the purified enzyme of 1.5 mmol/L (Hutzler and Dancis 1975). The K_m of saccharopine for SacD used was 1.15 mmol/L (Fjellstedt and Robinson 1975). The results of the calculations are presented in Table 3. The activities of LKR and SacD vary from 12 to 107 times that of LOX. The observed LOX was about one-half the uptake rate. However, given the standard error of the uptake measurements, uptake and oxidation again appear not to be different. These calculations are consistent with mitochondrial uptake of lysine being the rate-limiting step in liver LOX.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Diurnal variation.  Animals develop cyclic patterns in physical activities such as feed intake (Richter 1927) and physiological responses (Garlick et al. 1980, Peret et al. 1981, Potter et al. 1968), a fact that has been well known but underappreciated for some time. An extensively studied example of diurnal variation is tyrosine-aminotransferase (EC 2.6.1.5), which undergoes a 6-fold range in activity during the course of a day (Fuller and Snoddy 1968). Serine dehydratase (EC 4.2.1.3, Potter et al. 1968) and argininosuccinate synthetase (EC 6.3.4.5, Kato et al. 1978) undergo 2-fold variation throughout a day. Likewise, in this current report, ~2-fold range in LKR and SacD activities occurred over a 24-h period. Although not significant (P > 0.10), mitochondrial uptake of lysine did change ~3-fold during the day. The diurnal variation of lysine uptake is similar in magnitude with the diurnal variation in hepatic uptake of the system L model amino acid, cycloleucine (Baril and Potter 1968). An appreciation of the impact of diurnal variation is of critical importance in experimental design and in assessing treatment effects. These rhythmic variations could confound the effect of a treatment and lead to decision errors.

LOX.  Although the work of Miller (1962) and Noda and Ichihara (1976) suggest lysine is oxidized only in liver, their work does not eliminate the possibility that lysine is converted to -aminoadipate or -ketoadipate in extrahepatic tissues and later oxidized by the liver since these metabolites contain all the lysine carbon. Indeed, as much as 45% of the total body LKR activity may be in the kidney in the rat (Muramatsu et al. 1984). If the conversion of lysine to -aminoadipate occurs in tissues other than liver, one might expect an increased plasma -aminoadipate concentration when animals consume diets high in lysine. Young pigs fed a diet with 3.45% added lysine had increased -aminoadipic acid concentrations in plasma, liver, kidney and muscle of 12-, 3-, 7- and 5-fold, respectively (Edmonds and Baker 1987), suggesting increased production of/or decreased degradation of -aminoadipate. The liver was not likely the source of much -aminoadipate since -aminoadipate aminotransferase activity is 50-fold greater than the LKR and SacD activities (Scislowski et al. 1994), implying the increased concentration of -aminoadipate reported by Edmonds and Baker (1987) is a result of the production of -aminoadipate in extrahepatic tissues. The contribution of extrahepatic lysine degradation to whole-body lysine catabolism is required before a working model of lysine metabolism can be developed.

Lysine uptake by rat liver mitochondria.  Little work has been done to measure the uptake of lysine by mitochondria. Logically, mitochondria should contain uptake mechanisms for all amino acids that they themselves cannot synthesize, although, conceivably, some fraction of precursor amino acids arises from intramitochondrial protein degradation (Ferdinand et al. 1973, Nicoletti et al. 1977). The uptake of a number of amino acids by mitochondria has been reported (Cybulski and Fisher 1977, Torres et al. 1993). The observation that mitochondria of rats used in this study concentrated lysine is in agreement with King and Diwan (1973), but not in agreement with others (Gamble & Lehninger 1973, Hommes et al. 1983). The reason for the discrepancy is unclear but may be related to mitochondrial integrity. The respiratory control ratio of the mitochondria used in this current report was greater than four, suggesting that they were intact. The mechanism of mitochondrial lysine uptake is also unclear, although, a hydrogen ion antiporter was implied (King and Diwan 1973). This is reasonable since ornithine, which is structurally similar to lysine and like lysine (Blemings et al. 1994) is degraded in the mitochondrial matrix (Gamble and Lehninger 1973, Ip et al. 1974), is transported in this manner (McGivan et al. 1977).

Assuming that lysine taken up by the mitochondrion is not perfectly channeled to LKR or to a tRNA, one would expect an endogenous lysine pool in the mitochondrial matrix. An endogenous mitochondrial lysine pool contradicts the findings of Ferdinand et al. (1973) who were unable to detect lysine in freshly isolated mitochondria, although lysine was detected after 1 h, presumably from proteolysis of mitochondrial proteins. The model used in the current studies (see the Materials and Methods section) predicted positive C-values (matrix lysine at 0 time) 43 times out of 45 estimates, which is consistent with a matrix lysine concentration at the start of the experiment that is not zero. The average C-value was nearly 1.4 mmol/L.

Are the lysine uptake and oxidation rates less than the potential metabolism by LKR and SacD?  Some reasonable objections could be raised with regard to the comparisons made in Table 3. One possible objection is that cosubstrate concentrations (NADPH, NAD, PALP and CoA) do not mimic the in vivo state. Another possible objection is the use of 1 mmol/L substrate (lysine and saccharopine). However, a review of 10 reports from 1965 to 1993 revealed an average liver lysine concentration of 1.8 mmol/L with a range of 1.6 to 2.4 mmol/L.

Finally, the results presented in Figure 1 should be considered. In all cases, a concentration of 1 mmol/L lysine was used, and in all cases, a final mitochondrial matrix lysine concentration ~3.5 mmol/L was seen even as uptake and enzyme activities increased ~5-fold in response to increased dietary protein (Table 3). While concentrations other than 1 mmol/L lysine may have been used, we think it unlikely that other metabolic relationships would have become evident.

Transport as a limiting step in lysine catabolism.  In Table 4, the rates of lysine uptake, LKR and SacD activities, and LOX are compared using the results obtained from rats fed diets containing 5, 20 or 60% casein. Results obtained with 1 mmol/L lysine were scaled using the Michaelis-Menten equation. Lysine uptake could not be scaled because no estimate for V_max and K_m was available. To identify potential limiting steps, the ratio of LKR or SacD to LOX was calculated for each set of values. In all cases at substrate levels that should cover those which might be physiologically relevant (0.25, 0.5, 1.0, 3.5 or 5.0 mmol/L), the activities of enzymes are from 6- to 100-fold that of the predicted oxidation rate.

Our previous findings that liver LKR and SacD are exclusively housed in the mitochondrial matrix (Blemings et al. 1994) and the inherited defect in LOX while the activities of LKR and SacD are within normal ranges (Oyanagi et al. 1986) are consistent with a requirement for transport of lysine into the mitochondria as a prerequisite for its catabolism. Based on the results presented in Table 3, in concert with the discussion of these results, we suggest that the uptake of lysine into the mitochondrion limits the rate at which it is oxidized in rat liver. Lysine is not unique in this aspect since ornithine, a dibasic amino acid whose degrading enzymes are also restricted to the mitochondrial matrix (Gamble and Lehninger 1973, Ip et al. 1974), also was shown to have a transport limitation for degradation (Cohen et al. 1987, McGivan et al. 1977, Zollner 1984). Results of the present investigation and the data of others strongly suggest that transport can be an important control point in the catabolism of ornithine and lysine.

	FOOTNOTES

Manuscript received 1 April 1998. Initial reviews completed 7 August 1998. Revision accepted 31 August 1998.

	ACKNOWLEDGMENTS

The technical assistance of Linda Haas and Sonja Jensen is greatly appreciated.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Anderson H. L., Benevenga N. J., Harper A. E. Associations among food and protein intake, serine dehydratase, and plasma amino acids. Am. J. Physiol. 1968; 214:1008-1013[Free Full Text]
• Baril E. F., Potter V. R. Systematic oscillations of amino acid transport in liver from rats adapted to controlled feeding schedules. J. Nutr. 1968; 95:228-237
• Bergstrom J., Furst P., Noree L. O., Vinnars E. Intracellular free amino acid concentration in human muscle tissue. J. Appl. Physiol. 1974; 36:693-697[Free Full Text]
• Blemings K. P., Crenshaw T. D., Benevenga N. J. Kinetic parameters of [U-¹⁴C-]L-lysine oxidation in liver homogenates from rats adapted to 5% or 60% casein diets. FASEB J. 1992; 6:A1943
• Blemings K. P., Crenshaw T. D., Swick R. W., Benevenga N. J. Response of rat hepatic mitochondrial lysine oxidation (Lys. ox.), lysine -ketoglutarate reductase (LKGR) and saccharopine dehydrogenase (SacD) to 5%, 18%, and 60% casein diets. FASEB J. 1990; 4:A919
• Blemings K. P., Crenshaw T. D., Swick R. W., Benevenga N. J. Lysine -ketoglutarate reductase and saccharopine dehydrogenase are found exclusively in the mitochondrial matrix in rat liver. J. Nutr. 1994; 124:1215-1221
• Bobwell, C. E., Adkins, J. S. & Hopkins, D. T. (eds.) (1981) Protein Quality in Humans: Assessment and In Vitro Estimation, pp. 118-147. Avi Publishing Co., Westport, CT.
• Chang Y. F. Lysine metabolism in the rat brain: Blood-brain barrier transport, formation of pipecolic acid and human hyperpipecolatemia. J. Neurochem. 1978a; 30:355-360[Medline]
• Chang Y. F. Lysine metabolism in the rat brain: The pipecolic acid-forming pathway. J. Neurochem. 1978b; 30:347-354[Medline]
• Chu S., Hegsted D. M. Adaptive responses of lysine and threonine degrading enzymes in adult rats. J. Nutr. 1976; 106:1089-1096
• Cohen N. S., Cheung C. W., Rajiman L. Channeling of extramitochondrial ornithine to matrix ornithine transcarbamylase. J. Biol. Chem. 1987; 262:203-208[Abstract/Free Full Text]
• Cybulski R. L., Fisher R. R. Mitochondrial neutral amino acid transport: evidence for a carrier mediated mechanism. Biochem. 1977; 16:5116-5120[Medline]
• Dancis J., Hutzler J., Cox R. P., Woody N. C. Familial hyperlysinemia with lysine-ketoglutarate reductase insufficiency. J. Clin. Invest. 1969; 48:1447-1452
• Edmonds, M. S. & Baker, D. H. (1987) Failure of excess dietary lysine to antagonize arginine in young pigs. J. Nutr: 117: 1396-1401.
• Ferdinand W., Bartley W., Broomhead V. Amino acid production in isolated rat liver mitochondria. Biochem. J. 1973; 134:431-436[Medline]
• Fjellstedt T. A., Robinson J. C. Properties of partially purified saccharopine dehydrogenase from human placenta. Arch. Biochem. Biophys. 1975; 171:191-196[Medline]
• Food and Agriculture Organization, World Health Organization, & United Nations University (1985) Energy and Protein Requirements. WHO Technical Report Series 724, WHO, Geneva.
• Forsberg N. E., Austic R. E. Effect of dietary and extracellular potassium on lysine metabolism in the rat. Nutr. Res. 1986; 6:191-202
• Fuller R. W., Snoddy H. D. Feeding schedule alteration of daily rhythm in tyrosine alpha-ketoglutarate transaminase of rat liver. Science 1968; 159:738[Abstract/Free Full Text]
• Gahl M. J., Benevenga N. J., Crenshaw T. D. Concurrent measurement of protein synthesis and lysine catabolism in adult female rats induced to grow at different rates. J. Nutr. 1998; 128:1503-1511[Abstract/Free Full Text]
• Gamble J. G., Lehninger A. L. Transport of ornithine and citrulline across the mitochondrial membrane. J. Biol. Chem. 1973; 248:610-618[Abstract/Free Full Text]
• Garlick P. J., Clugston G. A., Swick R. W., Waterlow J. C. Diurnal pattern of protein and energy metabolism in man. Am. J. Clin. Nutr. 1980; 33:1983-1986[Abstract/Free Full Text]
• Greenawalt, J. W. (1974) The isolation of outer and inner mitochondrial membranes. In Methods in Enzymology Fleischer, S. and Packer, L. (eds.) vol. 31, pp. 310-323. Academic Press, NY.
• Harper A. E. Effect of variations in protein intake on enzymes of amino acid metabolism. Can. J. Biochem. 1965; 43:1589-1603[Medline]
• Higashino K., Tsukada K., Lieberman I. Saccharopine, a product of lysine breakdown by mammalian liver. Biochem. Biophys. Res. Comm. 1965; 20:285-290[Medline]
• Hommes F. A., Kitchings L., Eller A. G. The uptake of ornithine and lysine by rat liver mitochondria. Biochem. Med. 1983; 30:313-321[Medline]
• Hutzler J., Dancis J. Lysine-ketoglutarate reductase in human tissues. Biochim. Biophys. Acta 1975; 377:42-51[Medline]
• Ip M. M., Chee P. Y., Swick R. W. Turnover of hepatic mitochondrial ornithine aminotransferase and cytochrome oxidase using ¹⁴C carbonate as a tracer. Biochim. Biophys. Acta. 1974; 354:29-38[Medline]
• Kato H., Mizutani-Funahashi M., Shiosaka S., Nakagawa H. Circadian rhythms of urea formation and argininosuccinate synthetase activity in rat liver. J. Nutr. 1978; 108:1071-1077
• King M. J., Diwan J. J. Permeability of rat liver mitochondria to leucine, tyrosine, -aminoisobutyric acid, and lysine. Arch. Biochem. Biophys. 1973; 59:166-173
• LaNoue K. F., Schoolwerth A. C. Metabolite transport in mitochondria. Ann. Rev. Biochem. 1979; 48:871-922[Medline]
• Layne, E. (1957) Biuret Method. In: Methods in Enzymology (Colowick, S. D. and Kaplan, N. O., eds.) vol. 3, pp. 450-451. Academic Press, NY.
• McGivan J. D., Bradford N. M., Beavis A. D. Factors influencing the activity of ornithine aminotransferase in isolated rat liver mitochondria. Biochem. J. 1977; 162:147-156[Medline]
• Miller, L. L. (1962) The role of liver and the non-hepatic tissues in the regulation of free amino acid levels in the blood. In: Amino Acid Pools-Distribution, Formation, and Function of Free Amino Acids (Holden, J.T., ed.) pp. 708-721. Elsevier, NY.
• Muramatsu K., Takada R., Uwa K. Adaptive response of liver and kidney lysine -ketoglutarate reductase and lysine oxidation in rats fed graded levels of dietary lysine and casein. Agric. Biol. Chem. 1984; 48:703-711
• National Research Council (1985) Guide for the care and use of laboratory animals, Publication no. 85-23 (rev), National Institutes of Health, Bethesda, MD.
• National Research Council (1988) Nutrient requirements of swine, p. 48, National Academy Press, Washington, D.C.
• National Research Council (1989) Recommended Dietary Allowances, 10th ed, pp. 70-71, National Academy Press. Washington, D.C.
• Nicolletti M., Guerri C., Grisolia S. Turnover of carbamyl-phosphate synthase, of outer mitochondrial enzymes, and of rat tissues. Eur. J. Biochem. 1977; 75:583-592[Medline]
• Noda C., Ichihara A. Control of ketogenesis from amino acids IV. Tissue specificity in oxidation of leucine, tyrosine, and lysine. J. Biochem. (Tokyo) 1976; 80:1159-1164[Abstract/Free Full Text]
• Oyanagi K., Aoyama T., Tsuchiyama A., Nakao T., Uetsuji N., Wagatsuma K., Tsugawa S. A new type of hyperlysinemia due to a transport defect of lysine into mitochondria. J. Inher. Metab. Dis. 1986; 9:313-316[Medline]
• Palmieri, F. & Klingenberg, M. (1979) Direct methods for measuring metabolite transport and distribution in mitochondria. In: Methods in Enzymology LVI (Fleischer, S. & Packer, L., eds.) pp. 279-301. Academic Press, NY.
• Peret J., Foustock S., Chanez M., Bois-Joyeux B., Assan R. Plasma glucagon and insulin concentrations and hepatic phosphoenolpyruvate carboxykinase and pyruvate kinase activities during and upon adaptation of rats to a high protein diet. J. Nutr. 1981; 111:1173-1184
• Peters J. C., Harper A. E. Adaptation of rats to diets containing different levels of protein: effects on food intake, plasma and brain amino acid concentrations and brain neurotransmitter metabolism. J. Nutr. 1985; 115:382-398
• Potter V. R., Baril E. F., Watanabe M., Whittle E. D. Systematic oscillations in metabolic functions in liver from rats adapted to controlled feeding schedules. Fed. Proc. 1968; 27:1238-1245[Medline]
• Richter C. P. Animal behavior and internal drives. Qt. Rev. Biol. 1927; 2:307-343
• Schimke R. T. Adaptive characteristics of urea cycle enzymes in the rat. J. Biol. Chem. 1962; 237:459-468[Free Full Text]
• Scislowski P. W. D., Foster A. R., Fuller M. F. Regulation of oxidative degradation of L-lysine in rat liver mitochondria. Biochem. J. 1994; 300:887-891
• Soliman A., Harper A. E. Effect of protein content of diet on lysine oxidation by the rat. Biochim. Biophys. Acta 1971; 244:146-154[Medline]
• Torres N., Tovar A. R., Harper A. E. Metabolism of valine in rat skeletal muscle mitochondria. J. Nutr. Biochem. 1993; 4:681-689
• Zollner H. Ornithine uptake by isolated hepatocytes and distribution within the cell. Int. J. Biochem. 1984; 16:681-685[Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences