Altered hepatic mitochondrial fatty acid oxidation and ketogenesis in endotoxic rats NAOSHI TAKEYAMA, YOSHITOSHI ITOH, YASUHIDE KITAZAWA, AND TAKAYA TANAKA Department of Emergency and Critical Care Medicine, Kansai Medical University, Moriguchi, Osaka 570 Japan

TAKEYAMA, NAOSHI, YOSHITOSHI ITOH, YASUHIDE KITAZAWA, AND TAKAYA TANAKA. Altered hepatic mitochondrial fatty acid oxidation and ketogenesis in endotoxic rats. Am. J.

Physiol. 259 (Endocrinol. Metab. 22): E498-E505, 1990.-Rat hepatic mitochondrial function, including oxidative phosphorylation, fatty acid oxidative capacity, kinetic parameters of carnitine palmitoyltransferase I (CPT I), and sensitivity of CPT I to malonyl-CoA inhibition were studied in vitro in isolated mitochondria following Escherichia cob lipopolysaccharide (LPS). The hepatic mitochondrial CPT I in LPStreated rats showed a lower apparent maximum velocity ( V,,,) for palmitoyl-CoA and Ki for malonyl-CoA without changes in apparent Km for palmitoyl-CoA. The rate of oxygen consumption or end-product formation of palmitoyl-L-carnitine and octanoate was not altered, but the rate of CPT I-dependent palmitoyl-CoA (plus L-carnitine) oxidation was reduced by LPS, when acetyl-CoA produced via P-oxidation was directed toward citrate. When acetyl-CoA was directed to acetoacetate, the oxygen consumption rates of palmitoyl-L-carnitine and palmitoyl-CoA (plus L-carnitine) were decreased by LPS, although mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase activity was not altered. These results indicate that hepatic mitochondria isolated from LPS-treated rats show lower ketogenic and long-chain acyl-CoA oxidative capacity than those of fasted controls, and inhibition of ketogenesis is elicited at a site distal to CPT I in addition to reduction in CPT I activity. oxygen consumption; mitochondrial P-oxidation; 3-hydroxy-3methylglutaryl-CoA synthase; carnitine palmitoyltransferase; malonyl-CoA

SEPSIS AND ENDOTOXEMIA comprise a common clinical problem with extremely high mortality. Although a number of investigations have provided valuable information regarding biochemical alterations in recent years, many aspects are still unknown, and many controversial issues remain unresolved. For example, studies of the capacity of septic patients and animals to use fat as an energy source have yielded apparently contradictory results. Several previous studies have demonstrated that hypertriglyceridemia (1, 36, 40), hypoketonemia (24), hyperinsulinemia (17, 32, 40, 41), decreased lipoprotein lipase activity in skeletal and cardiac muscle (l), and decreased carnitine concentration in muscle (3) and liver (36) occur in animals after in vivo administration of lipopolysaccharide (LPS) or live bacteria. These findings appear to support the suggestion that the oxidation of both exog-



enous and endogenous lipids is suppressed in sepsis. A study by Chen (7) of the utilization of exogenous fat emulsion in septic rats showed that sepsis depressed the utilization of exogenous fat to different degrees depending on the severity of sepsis. On the other hand, Wolfe et al. (40) proposed that the ability to oxidize endogenous long-chain fatty acids was not impaired in septic dogs and that the fatty acids in very low-density-lipoprotein can serve as an important energy source. We have previously demonstrated in rats that the hepatic activity of total carnitine palmitoyltransferase (CPT; EC decreased and that the concentration of hepatic malonyl-CoA, the first intermediate in lipogenesis to become elevated in the fed state and is known to be a potent inhibitor of CPT I (overt form of CPT) (21), rose significantly after LPS administration (36). Vary et al. (38) also found that hepatic malonyl-CoA levels were significantly increased in animals with septic abscess compared with controls or those with sterile inflammation. Both authors suggest that the entry of long-chain acyl-CoAs into the mitochondria may be inhibited after LPS administration or during infectious states, presumably resulting from an interaction between lowered CPT activity and a higher level of malonyl-CoA. Ketone bodies are produced by the liver as one of the end products of P-oxidation, and this production is closely related to the degree of fatty acid oxidation induced by different nutritional and hormonal states. Hypoketonemia is characteristically observed in sepsis (24, 38), endotoxemia (24, 36), and injury (30, 33), and an inverse relationship has been observed between the concentration of ketone bodies in . the blood and the severity of injury (33). Although the observation of the failure of the liver to enhance ketogenesis during carbohydrate deprivation is probably important, the precise mechanisms of hypoketonemia in sepsis and endotoxemia remain unresolved. The aim of this study was to clarify the mechanisms of hypoketonemia induced by administration of LPS to fasted rats and to dissect the conflicting views with regard to fatty acid oxidation. We have measured the hepatic mitochondrial oxidative rates of various lipid and nonlipid substrates by means of polarography and have also measured citrate, ,&hydroxybutyrate, and acetoacetate production from palmitoyl+carnitine in isolated mitochondria from LPS-treated rats. In addition, the

$1.50 Copyright 0 1990 the American Physiological


Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.



kinetic parameters of CPT I, the sensitivity of CPT I to malonyl-CoA inhibition, and the activity of 3-hydroxy3-methylglutaryl (HMG)-CoA synthase (EC in isolated liver mitochondria were measured. METHODS Animals and treatment. Male Wistar rats weighing -190 g were used in all experiments. Animals were kept in wire-bottomed cages under controlled temperature (22”C), with a 12:12 h light-dark photoperiod. Those rats used in the experiments were allowed free access to water but were without food for 24 h before administration of Escherichia cob LPS (0127: B8, Difco Laboratories, Detroit, MI). LPS (7 mg/kg body wt; 15% lethal dose at 24 h) dissolved in saline solution was injected intraperitoneally. Control rats were injected only with saline. After the injection of LPS, these animals were further starved for 24 h before being killed. Isolation of mitochondria. Mitochondria were isolated according to Hoppel et al. (18). Rats were killed by decapitation, arteriovenous blood was collected in heparinized tubes, and a 20% (wt/vol) liver homogenate was prepared in an ice-cold buffer (in mM: 220 mannitol, 70 sucrose, and 5 MOPS; pH 7.4) with a Potter-Elvehjem PTFE glass homogenizer. The homogenate was diluted twice with homogenizing buffer, and EDTA was added to achieve a final concentration of 2 mM. After differential centrifugation (successively 400 g and 7,000 g for 10 min at 4”C), the final mitochondrial pellet was washed twice with homogenizing buffer and then resuspended in homogenizing buffer to give a final protein concentration of 30 mg/ml. In one subgroup, portal vein blood was obtained for the measurement of plasma free fatty acids, insulin, and glucagon concentration under pentobarbital sodium (40 mg/kg body wt ip) anesthesia. Glutamate dehydrogenase (EC as the marker enzyme of the mitochondrial matrix was measured in both the initial homogenate and the final mitochondrial suspension. Mitochondrial protein per gram of liver was calculated by dividing the glutamate dehydrogenase activity per gram of liver by that of activity per milligram of mitochondrial protein. Polarographic measurements. Oxygen consumption by isolated mitochondria was measured with a Clark electrode (Yellow Springs Instruments, Yellow Springs, OH) at 30°C in a temperature-controlled chamber of &ml volume stirred with a magnetic flea. The incubation medium used was air-saturated (in mM) 130 KCl, 10 N2-hydroxyethylpiperazine-N ‘-2-ethanesulfonic acid (HEPES), 0.1 ethylene glycol-bis( @-aminoethylether) -N, N,N’,N’-tetraacetic acid (EGTA), 5 MgCl,, and 2 KPi and 2 mg/ml essentially fatty-acid free bovine serum albumin, pH 7.2 (26), containing ~1.5 mg mitochondrial protein/ml. After endogenous mitochondrial substrates were depleted by the addition of small amounts of ADP, nonlipid substrates, 10 mM succinate plus 2 PM rotenone, or 10 mM glutamate plus 5 mM L-malate were added and the rates of oxygen consumption were measured. To initiate state 3 conditions (ADP-stimulated oxidation), 400 nmol





ADP was added with 10 ~1 of the buffer. The respiratory control ratios (RCR) and ADP/O ratios were calculated from the polarographic recordings with the use of standard procedures (11). The capacity of the hepatic mitochondria to use palmitoyl groups (20 PM palmitoyl-CoA plus 2 mM Lcarnitine or 10 P.M palmitoyl-L-carnitine) as substrates for oxidation was measured by using the same respiratory medium containing 0.1 mM 2,4dinitrophenol (DNP), -1.5 mg mitochondrial protein/ml, malonyl-CoA (0 or 3 PM), and either 2.5 mM L-malate or 10 mM malonate. The oxidation of middle-chain fatty acid (0.2 mM octanoate) was measured similarly in the presence of 2.5 mM L-malate, 0.1 mM DNP, 1 mM ATP, and 2.5 pg of oligomycin/ml to inhibit the DNP stimulation of ATPase (13) Acetoacetate, P-hydroxybutyrate, and citrate production were measured at 30°C using the same respiratory medium containing, in a final volume of 1 ml, 40 PM palmitoyl-L-carnitine, 0.1 mM DNP, ~1.5 mg of mitochondrial protein/ml, and either 10 malonate or 2.5 mM L-malate, respectively (5) . After 1 min of incubation, the reaction was terminated by the addition of 100 ~1 of 70% (wt/vol) perchloric acid. The mixture was centrifuged, neutralized by 2 M K3POd, and measured spectrophotometrically by standard procedures (5). The net product formation was obtained by subtracting the amount of product formation in the absence of palmitoyl-L-carnitine. In these conditions, both acetoacetate and citrate formation were linear with incubation time for at least 2 min. The ratio of oxygen consumed to palmitoyl groups oxidized (AO/AP ratio) was determined at low concentrations (7 PM) of palmitoyl+carnitine by dividing the amount of oxygen consumed (in nanograms per atoms) by the number of nanomoles of palmitoyl+carnitine added (15). Assay of CPT I. The CPT I activity of liver mitochondria was assayed by measuring the difference in coenzyme A (CoASH) generation from malonyl-CoA with 5min incubation at 25°C in the presence and absence of L-carnitine using high-performance liquid chromatography (HPLC) (model LC-4A; Shimadzu, Kyoto, Japan) as described previously (35). The incubation mixture contained (in mM), in a final volume of 1 ml, 75 potassium chloride, 50 mannitol, 25 HEPES (pH 7.0), 0.2 EGTA, 2 potassium cyanide, and 5 dithiothreitol, as well as 1.75 mg of essentially fatty acid-free bovine serum albumin, palmitoyl-CoA (30-120 PM), 0.5 mM L-carnitine, malonyl-CoA (O-150 PM), and -200 pg of mitochondrial protein. To exclude the CoASH generation derived from palmitoyl-CoA hydrolase, and nonenzymic palmitoyl-CoA hydrolysis, the reference solution (distilled water was substituted for L-carnitine) was made in each. The CoASH generation was linear with time for up to 8 min of incubation at 25OC with 300 pg of mitochondrial protein and with protein concentrations of up to 600 pg of protein per assay mixture for 5-min incubations. The contributions of CPT II (latent form of CPT) and peroxisomal carnitine octanovltransferase were cor-

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.





rected by subtracting the activity of the unsuppressed fractions by the high malonyl-CoA concentrations (150 PM) [only CPT I is sensitive to malonyl-CoA (22)]. The sensitivity of CPT I to inhibition by malonyl-CoA was assessed by employing the graphic method of Dixon and by using 50 PM and 80 PM palmitoyl-CoA, 0.5 mM L-carnitine and malonyl-CoA concentration of 0, 0.25, 0.5,0.8, 1, 1.5,2, and 2.5 PM (35). The lines of the Dixon plot were linear below malonyl-CoA concentration of 2.5 PM for fasted control animals and 1.5 PM for LPStreated animals. Assay of HMG-CoA synthase. HMG-CoA synthase activity from rat liver mitochondria was assayed according to Miziorko (23). A mitochondrial pellet prepared as described above was suspended in a volume of 20 mM potassium phosphate buffer, pH 7.0, containing 0.1 EDTA and 0.1 mM dithiothreitol, which is equal to the number of grams of liver processed. The mitochondrial suspension was frozen in liquid nitrogen and lyophilized (Neocool Freeze Drier; Yamato, Tokyo, Japan) and was stored below -70°C until use. The lyophilized powder (10 mg) was resuspended in 10 ml of 20 mM potassium phosphate buffer, pH 7.0, containing 0.1 mM EDTA, 0.1 mM dithiothreitol, 20 PM acetoacetyl-CoA, and 0.1 mM phenylmethylsulfonyl fluoride, and the homogenate was centrifuged for 15 min at 15,000 g at 4°C. The supernatant was used as the enzyme solution. Enzyme activity was assayed spectrophotometrically by measuring the initial rate of acetyl-CoA-dependent disappearance of acetoacetyl-CoA at 300 nm (UV-2100; Shimadzu, Kyoto, Japan). A molar extinction coefficient of 13,600 cm-’ for acetoacetyl-CoA at pH 8.2 was used. A unit (U) of enzymatic activity was defined as the amount of enzyme necessary to convert 1 pmol of substrate into HMG-CoA per minute. General analytical methods. The concentrations of hepatic protein and of plasma free fatty acids, acetoacetate, P-hydroxybutyrate, glucose, and triacylglycerol were assayed spectrophotometrically as described previously (36). Plasma insulin and glucagon concentration were determined by radioimmunoassay (40). Chemicals. Palmitoyl-L-carnitine, CoASH, other CoA derivatives, and bovine serum albumin fraction V (fatty acid-free) were purchased from Sigma (St. Louis, MO). The HPLC columns used, a 250 X 4.0-mm ID Lichrosorb RP-18 (analytical column; 5 pm particle size), and a 45.0 x 4.0-mm ID Zorbax ODS (guard column; 45 pm particle size), were purchased from Wako Pure Chemical Industries (Tokyo, Japan). L-Carnitine was kindly donated by Earth Chemical (Osaka, Japan). All other reagents were of the highest grade commercially available. Statistical analysis. The statistical significance of differences were assessed with Student’s unpaired t test. Variance of the mean was expressed as the standard deviation (SD). RESULTS General data. LPS administration to starved rats increased the liver weight (5.21 t 0.84 g for 10 fasted control rats vs. 7.83 t 1.45 g for 11 LPS-treated rats; P



< 0.01). This was probably due to an increase in the water content of the livers, because the dry weights of the livers were not significantly different between the two groups (data not shown). The amount of mitochondrial protein per gram liver was significantly lower in LPS-treated (70.05 t 9.48 mg) than in fasted control animals (108.45 t 15.39; P < 0.01); however, the amount of mitochondrial protein per total liver was unaffected [538.9 t 60.5 mg for fasted control rats vs. 551.4 t 53.1for LPS-treated rats; not significant (NW. Because the amount of mitochondrial protein per total liver was unaffected after LPS administration, the results obtained on the basis of mitochondrial protein should reflect the results expressed per total liver. Thus, the data of oxygen consumption and enzyme activities were expressed only relative to mitochondrial protein. Blood analysis. The effects of LPS administration to fasted rats on the plasma concentrations of glucose, free fatty acids, triacylglycerol, total ketone bodies (p-hydroxybutyrate plus acetoacetate), insulin, and glucagon are shown in Table 1. The mean total concentration of plasma ketone bodies was decreased fivefold, and plasma triacylglycerol concentration was increased twofold in the rats treated with LPS compared with controls. The portal level of free fatty acids and systemic plasma glucose level were not altered. The portal concentrations of insulin and glucagon were increased in LPS-treated compared with control rats with the insulin-to-glucagon molar ratio being reduced because of the much more dramatic increase in portal glucagon. Oxidative phosphorylation. Rates of oxygen consumption for nonlipid substrates of hepatic mitochondria isolated from LPS-treated and fasted control rats were measured (Table 2). The substrates studied were the citric acid cycle intermediates succinate in a rotenoneblocked system (inhibiting reversed electron transport) or an amino acid glutamate that enters the citric acid cycle at a-ketoglutarate. Glutamate is NAD-linked and supplies electrons to energy coupling site I of the respiratory chain. Succinate is FAD-linked and enters the electron transport pathway after site I. State 4 oxidative rates and ADP/O ratio were similar in both groups for either substrate. Hepatic mitochondria isolated from LPS-treated rats showed an increased RCR for either substrate because of an increase in state 3 oxidative rates. Oxidation of various lipid substrates. Uncoupled oxidation of CPT I-independent palmitoyl+carnitine and product formation in LPS-treated and fasted control liver mitochondria was measured under two different conditions. In the presence of L-malate, which is a donor of oxaloacetate, acetyl-CoA is directed to the citric acid cycle and the end product of P-oxidation should be citrate (15). In this condition, the AO/AP ratio was not altered in LPS-treated hepatic mitochondria and was similar to the theoretical value of 22 (Table 3). The detected end product in the presence of L-malate was only citrate, whereas acetoacetate and /3-hydroxybutyrate were not detected (Table 3). In the presence of malonate, which is an inhibitor of succinate dehydrogenase (EC,

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.




1. Effects of LPS administration Total Ketone Bodies, pmol/ml




of metabolites

on plasma concentrations

Glucose, mg/lOO ml



Triacylglycerol, mg/lOO ml

Free Fatty Acid, meq/l

Insulin, dJ/ml

Glucagon, pg/ml

139.3t18.5 0.618kO.273 47.8zk5.1 10.4k3.6 44.2+: 12.9 Control 2.41kO.39 0.711IkO.187 g&8&18.8* 16.8+4.9t 331.8284.6” LPS 0.51zko.14* 138.0zk12.3 Values are means k SD from 10 to 12 animals. Measurements were made on plasma obtained 24 h after treatment with lipopolysaccharide (LPS; 7 mg/kg body wt) or saline (control). * P < 0.01, t P < 0.05 compared with control animals.


2. Oxidative phosphorylation Mitochondrial

in liver mitochondria


Succinate + rotenone Control LPS

State 4 28.02t2.38 26.32k4.12

and fasted control rats

isolated from LPS-treated State 3


96.32t5.55 112.25&16.05*


3.57t0.42 4.29&0.41-f-

1.69k0.16 1.72k0.12

Glutamate + L-malate Control 15.17t1.81 69.19k8.44 4.63k0.58 2.57kO.16 13.34k1.97 76.79&6.42* 5.98=t0.74t 2.67k0.22 LPS Values are means t SD from 8 to 13 animals. Liver mitochondria were isolated and incubated at 30°C. Oxygen consumption was measured in mitochondrial suspensions at 1.5 mg protein/ml with 10 mM succinate (in presence of 2 PM rotenone) or 10 glutamate + 5 mM L-malate as substrates, in the absence (state 4) or presence (state 3) of 200 pM ADP. State 4 and state 3 values, nmol0. min-’ .mg protein-‘. * P < 0.05, t P < 0.01 compared with control animals. RCR, respiratory control ratio.

3. Palmitoyk-carnitine oxidation and fasted control liver mitochondria


Mitochondrial Sources

Oxygen Consumption, nmol0. min-’ . mg protein-*

in presence of L-ma&e

AO/AP, natoms O/nmol

or malonate in LPS-treated

Acetoacetate Production, nmol min-’ . mg protein-’


@Hydroxybutyrate Production, nmol . min-’ mg protein-’

Citrate Production, nmol . min-’ . mg protein-’


L-Malate Control 92.38k6.34 20.76t1.42 ND ND 10.29t1.83 LPS 95.25k9.79 22.00t0.97 ND ND 11.71k1.43 Malonate Control 50.66t7.02 13.83k0.69 12.1221.72 ND ND LPS 36.29&5.38* 13.75kO.82 6.70t0.92” ND ND Values are means k SD from 8 to 10 animals. Liver mitochondria were isolated and incubated as in Table 2 in presence of 0.1 mM 2,4dinitrophenol. Oxidation of palmitoyl-L-carnitine was determined in the presence of either 10 mM malonate or 2.5 mM L-malate. Citrate, * P < 0.01 acetoacetate, and P-hydroxybutyrate production were calculated as described in METHODS by using 40 PM palmitoyl+carnitine. compared with control rats. ND, not detectable.

acetyl-CoA is not available for citrate formation and the end product of P-oxidation should be acetoacetate (15). In this condition, the AO/AP ratio was not altered in LPS-treated hepatic mitochondria, and the data were similar to the theoretical value of 14 (Table 3). Only acetoacetate was detected as an end product in the presence of malonate, whereas citrate and ,&hydroxybutyrate were not detected (Table 3). Thus under these conditions the acyl-groups transferred to mitochondria were quantitatively oxidized to acetoacetate or citrate, and the oxygen consumption caused by oxidation of substrates is therefore a direct measure of the flux through ,&oxidation in mitochondria isolated from both LPS-treated and fasted control animals, as suggested by Garland et al. (15) In the presence of L-malate, the capacity of hepatic mitochondria obtained from LPS-treated rats to oxidize palmitoyl-L-carnitine was not altered, as assessed by the rates of oxygen consumption and citrate formation (Table 3). In the presence of malonate, the rates of oxygen consumption and acetoacetate formation were significantly reduced in LPS-treated mitochondria (Table 3). The ability to oxidize octanoate was unaffected bv LPS administration, whereas CPT I-dependent pal-

mitoyl-CoA oxidation was significantly reduced in LPStreated mitochondria whether the end product of fatty acid oxidation is citrate or acetoacetate (Table 4). Effect of malonyl-CoA on palmitoyl-CoA oxidation. The rate of oxygen consumption for palmitoyl-CoA plus Lcarnitine was measured in mitochondria obtained from LPS-treated and fasted control rats with or without 4. Octanoate and palmitoyl-CoA oxidation in presence of L-ma&e or malonate in LPS-treated and fasted control liver mitochondria


Oxygen Consumption, nmol0 min-’ mg protein-’ l

Mitochondrial Sources


Palmitoyl-CoA Octanoate



Control 27.78k2.65 58.68t4.89 31.35k2.38 LPS 28.67t4.63 53.37zk8.15” 24,43&3.26? Values are means t SD from 8 to 13 animals. The oxidation of 20 PM palmitoyl-CoA plus 2 mM L-carnitine was determined in the presence of 0.1 mM 2,4-dinitrophenol and either 10 malonate or 2.5 mM L-malate. Octanoate (0.2 mM) oxidation was determined in the presence of 2.5 mM L-malate, 1 mM ATP, 2.5 pg of oligomycin/ml, and 0.1 mM &&dinitrophenol. * P c 0.05, t P < 0.01 compared with control animals.

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.





malonyl-CoA (Fig. 1). In mitochondria from LPS-treated rats, the oxidative capacity from palmitoyl-CoA plus Lcarnitine was nearly twofold lower in the presence of malonyl-CoA than in the absence of malonyl-CoA, regardless of the end product formed. In mitochondria from fasted control rats, the oxidative capacity from palmitoyl-CoA plus L-carnitine was nearly 30% lower in the presence of malonyl-CoA than in the absence of malonylCoA, regardless of the end product formed. In the presence of malonyl-CoA, the difference in palmitoyl-CoA (plus L-carnitine) oxidative capacity between LPStreated and fasted control liver mitochondria was expanded. Relationship of mitochondrial enzymes to fatty acid oxidation. The kinetics and the malonyl-CoA inhibition

of CPT I in mitochondria isolated from LPS-treated and fasted control rats are shown in Table 5. Because in CPT I assays using albumin the actual concentration of the substrate (the fraction of palmitoylCoA not bound to albumin) cannot be determined, the values of both the Michaelis constant (Km) and the maximum velocity ( Vmax) for palmitoyl-CoA and the dissociation constant for the CPT I-malonyl-CoA complex (Ki) are only apparent values. However, valid comparison among the values still can be made because they were assayed under the same experimental conditions. LPS administration in vivo lowered the apparent Vmax by 30% without significant alteration of the apparent K, for palmitoyl-CoA. The apparent Ki values are 5.4-fold lower in LPS-treated rats than in fasted control rats. This result, obtained from a 48-h fasted control rat, is consistent with our previous data (35). The specific activity of HMG-CoA synthase in lyophmalate

7 60 .i g


if i; *F





FIG. 1. Effect of malonyl-CoA on palmitoyl-CoA oxidation in the presence of L-malate or malonate in LPS-treated and fasted liver mitochondria. Oxygen consumption of isolated mitochondria was de-


as described

in Table

4, with

or without

3 mM


Values are means t SD from 8 to 10 animals. * P < 0.05., g. P < 0.01 compared with control animals.


5. Kinetic parameters and malonyl-CoA sensitivity of CPT I and HMG-CoA synthase activities in liver mitochondria from LPS-treated and fasted control TABLE

CPT I Group

Max, nmolmin-’ . mg protein-’

Km for PalmitoylCoA, PM

Ki for MalonylCoA PM

HMG-CoA Synthase, mU/mg protein

Control 8.53t0.90 82.64t10.21 2.04t0.45 34.77t4.47 LPS 5.01*1.05* 76.48k8.71 0.38tO. 14* 32.09t7.11 Values are means & SD from 8 to 10 animals. CPT I activity was assayed as described in METHODS, using 30-120 PM palmitoyl-CoA, 0.5 mM L-carnitine, and 1.75 mg/ml albumin (pH 7.0) at 25°C in a final volume of 1 ml. Ki value was calculated from Dixon plots using 50 or 80 PM palmitoyl-CoA and 0.5 mM L-carnitine. Malonyl-CoA concentration varied from 0 to 2.5 PM. HMG-CoA synthase activity was measured as described in METHODS, in lyophilized liver mitochondria. * P < 0.01 compared with control rats.

ilized mitochondria Q

did not change following LPS (Table

DISCUSS1oN Hepatic mitochondrial function, including oxidative phosphorylation, has been extensively examined in endotoxic and septic animals with inconsistent results (14, 31). In this study, isolated mitochondria from LPS-treated rats were shown to be fully functional with regard to oxidative phosphorylation and respiratory chain capacity, as demonstrated by unaltered ADP/O ratio and enhanced RCR. An increased RCR probably represents an adaptive response of mitochondria in terms of increased efficiency of oxidative function, as suggested by Fry et al. (14) in murine peritonitis. The rate of liver ketogenesis is a function of some intrahepatic factors as well as the supply of fatty acids to the liver, which is determined by the concentration of portal and arterial fatty acids and by the circulatory rate. However, even when the amounts of fatty acids supplied to the liver are equal, ketone bodies production in fed rats is lower than in starved rats (25). Besides, a more recent report has demonstrated that high physiological levels of insulin are capable of restricting hepatic ketogenesis independently of plasma fatty acid concentration in humans (19). In this experiment a contribution of extrahepatic regulation of ketogenesis to the fivefold decrease in plasma ketone bodies observed in LPStreated rats is doubtful, because the concentration of portal fatty acids seen in this study was unaltered, and other investigators have reported variable data on the plasm a concen .tration of fatty acids in endotoxemia and sepsis (1, 24, 40). Hence it appears that intrahepatic factors may be more important. It is generally agreed that CPT I appears to play an important role in regulating hepatic mitochondrial longchain fatty acid oxidation and ketogenesis (13, 21). This enzyme may be of particular importance under conditions where concentrations of malonvl-CoA are increased (9). The activity of CPT I in vivo Gould be modulated by three factors, as follows: 1) the synthesis rate of CPT I protein (4, 29); 2) the changes in sensitivity of CPT I to malonyl-CoA inhibition (6, 29); and 3) the content of

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.



(22) and long-chain acyl-CoA (29). Mamalonyl-CoA lonyl-CoA, a potent and probably a physiological inhibitor of CPT I, fluctuates in parallel with the rate of lipogenesis. At the high rates of lipogenesis observed in the fed state, increased malonyl-CoA concentrations permit the diversion of long-chain acyl-CoA into esterification rather than oxidation, as a consequence of CPT I inhibition. The former two factors would be concerned with long-term regulation of CPT I activity (5, 28), whereas short-term regulation would be performed by the latter factor (16). The changes in the kinetic parameters of CPT I induced by LPS administration were similar to those observed in insulin-treated diabetic animals (9). Several studies have shown that hepatic CPT I activity decreased in states characterized by high insulin-to-glucagon ratios. Guzman and Geelen (16) demonstrated that insulin, epidermal growth factor, vasopressin, and the phorbol ester inhibit CPT I activity in vitro, whereas glucagon renders the enzyme more active when assayed in digitonin permeabilized hepatocytes. In contrast, in vitro exposure to glucagon (2) failed to affect the activity and sensitivity of CPT I to malonyl-CoA when permeabilized hepatocytes from adult rats was used. In more recent experiments, hepatic CPT activity, CPT mRNA, and transcription rates are increased two- to six-fold in BB Wistar spontaneously diabetic rats and return to control values on insulin administration in vivo (4). Although LPS-treated rats show the decrease in the plasma insulin/glucagon molar ratio, the possible effect of these hormones on the changes in CPT I activity after LPS cannot be denied. The possibility exists that the plasma insulin and glucagon levels are not proportional to this postreceptor action, since such hormonal resistance arises in septic animals (32, 41). The hypersensitivity of CPT I to malonyl-CoA inhibition in endotoxic animals was ascertained by the observed fivefold decrease in Ki values for malonyl-CoA. The decrease in Ki values indicates that the affinity of the enzyme for malonyl-CoA was increased. The direct inhibitory effect of malonyl-CoA on longchain acyl-CoA oxidation and the hypersensitivity of CPT I to malonyl-CoA inhibition induced by LPS administration also was confirmed by polarographic measurements when palmitoyl-CoA (plus L-carnitine) was used as a substrate regardless of the end product formed. The elevation of hepatic malonyl-CoA content in rapidly frozen samples has been previously reported to occur after LPS (36) or intraperitoneal administration of a fecal-agar pellet (38). On the other hand, we also demonstrated that the hepatic level of long-chain acyl-CoA was progressively decreased after LPS (36). MalonylCoA inhibits CPT I competitively with respect to the long-chain acyl-CoA substrate (29). Therefore, when the concentration of long-chain acyl-CoA decreases, the inhibition by malonyl-CoA will become more potent. Thus we inferred that a reduced concentration of long-chain acyl-CoA after LPS accentuates the inhibition of CPT I by malonyl-CoA in vivo much more than that observed in this in vitro condition. These metabolic changes, including the characteristics





of CPT I kinetics, indicate that the entry of long-chain acyl-CoAs into mitochondria was restricted at the CPT I barrier; consequently, hepatic fatty acid oxidation and ketogenesis would be decreased after LPS. The inhibition of long-chain acyl-CoA oxidation by LPS administration at the site of CPT I was also confirmed by in vitro polarograph .lC measurement of isolated mitochondrial respiration. When acetyl-CoA produced by P-oxidation was directed to citrate, the oxidation of palmitoyl-CoA plus L-carnitine, which requires CPT Idependent enzymatic transfer to palmitoyl-L-carnitine before oxidation occurs, was reduced in mitochondria isolated from LPS-treated rats. However, the oxidation of CPT I-independent palmitoyl-L-carnitine and octanoate were similar to that in the fasted control animals. The medium-chain fatty acids, such as octanoate, are not dependent on CPT I and are freely permeable to the mitochondrial matrix for P-oxidation. The intactness of both octanoate and palmitoyl-L-carnitine oxidation when acetyl-CoA was directed to the citric acid cycle indicates an additional important point. It appears probable that the activity of specific enzymes of P-oxidation per se, the CPT II activity, and the intramitochondrial fatty acyl-CoA synthetase activity (EC were unaffected by LPS administration. Thus these factors do not influence the failure of physiological ketogenic adaptation to occur during fasting associated with endotoxemia. However, the regulation of the ketogenic pathway is multisite, and consequently, the inhibition only of CPT I activity is not enough to explain the marked hypoketonemia observed in this study. It is likely that the partition of acetyl-CoA produced via ,&oxidation between synthesis of citrate catalyzed by citrate synthase (EC and flux into the HMGCoA cycle is the factor that determines intramitochondrial regulation of ketogenesis (10, 20, 27, 30). Previous study on the effect of partial hepatectomy and acute surgical stress on hepatic ketogenesis have demonstrated that decreased ketonemia compared with sham-operated controls is also observed after administration of medium-chain fatty acids (30). The authors concluded that decreased ketogenesis results from increased diversion of acetyl-CoA to citric acid cycle for complete oxidation in association with reduced CPT I activity. The present study demonstrated that the oxidation of palmitoyl-L-carnitine, as well as that of palmitoyl-CoA plus L-carnitine, was reduced after LPS administration when the end product of ,&oxidation was directed to acetoacetate, although the oxidation of palmitoyl-L-carnitine was not altered in the presence of L-malate. Because the oxidation of palmitoyl-L-carnitine is CPT Iindependent, the difference in palmitoyl-L-carnitine oxidation in the presence of either L-malate or malonate indicates the existence of intramitochondrial factors (distal to the CPT I level) that control ketogenesis in LPS-treated rats. This result is good agreement with the conclusion of the study of Schofield et al. (30). The restriction of acetoacetate formation from acetyl-CoA and increased diversion to the citric acid cycle may be explained by an increased hepatic energy requirement

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.





such as increased rates of gluconeogenesis, which was commonly observed in endotoxemia (17). The precise regulatory mechanisms of the partitioning of intramitochondrial acetyl-CoA are not fully elucidated. Decaux et al. (10) reported that the decreased ketogenesis at weaning is caused by reduced activity of HMG-CoA synthesis, which is involved in HMG-CoA cycle, in addition to reduced activity of CPT I. In our study, the specific activity of HMG-CoA synthase measured in vitro in isolated mitochondria was not altered after LPS administration, and the activity was considerably higher than that of CPT I. This finding belies a possible role of HMG-CoA synthase in the hypoketogenesis observed in endotoxic animals. However, recent reports presented by Lowe and Tubbs (20) have postulated that succinyl-CoA, which reduces the activity of hepatic mitochondrial HMG-CoA synthase as a consequence of the enzyme catalyzing its own succinylation, may regulate the ketogenic flux through the HMG-CoA cycle in vivo (27). Although we have not measured intramitochondrial succinyl-CoA content, it is possible that an inhibitor(s) such as succinyl-CoA may participate in the in vivo regulation of HMG-CoA synthase activity, and that the actual activity in vivo may be much lower than that measured in vitro. It is not possible at present to make a definite statement as to the mechanisms that are responsible for changes in malonyl-CoA sensitivity and the Vmaxof CPT I or hypoketogenesis after LPS. Apart from glucagon and insulin, other hormones such as catecholamine and cortisol, and tumor necrosis factor (TNF)/cachectin may be related to these changes. Many studies have shown that TNF/cachectin, a protein produced by monocytes and macrophages, is implicated as a mediator of the lethal effects of LPS (37) and the metabolic effects of tissue injury (34). Feingold et al. (12) showed that TNF/ cachectin administration to rats in vivo elicits hypertriglyceridemia as a consequence of enhanced lipogenesis by the liver independently of insulin concentration. Although no direct evidence has yet been made available, the possibility exists that this cytokine elicits antiketogenie action. We conclude that the hepatic mitochondria isolated from LPS-treated rats show less ketogenic and longchain acyl-CoA oxidative capacity than those of fasted control animals. The mechanisms for this can be explained as a consequence of reduced CPT I activity. In addition to this, intramitochondrial factors may play some part in the decreased rate of ketogenesis. The hypoketonemia induced by LPS administration may, at least in part, result from reduced ketogenesis by the liver. The present study does not exclude the possibility that decreased ketonemia after LPS administration results from accelerated extrahepatic utilization of ketone bodies. It has been reported, however, that the rate of ketone body clearance is unaffected after induction of pneumococcal sepsis (39) or surgical stress (30) under conditions where marked hypoketonemia was observed. We are indebted to the staff of our department for helpful advice and kind assistance. Address for reprint reouests: Dept. of Emergencv and Critical Care


Medicine, Kansai Medical Univ., Fumizono-cho 570, Japan.




Received 23 October 1989; accepted in final form 30 May 1990. REFERENCES G. J., C. B. CORLL, AND R. R. MARTINEZ. Triacylglycerol kinetics in endotoxic rats with suppressed lipoprotein lipase activity. Am. J. Physiol. 253 (Endocrinol. Metub. 16): E59-E64, 1987. Use of a selectively permeabilized 2. BOON, M. R., AND V. A. ZAMMIT. isolated rat hepatocyte preparation to study changes in the properties of overt carnitine palmitoyltransferase activity in situ. 1. BAGBY,

Biochem. 3. BORDER,

J. 249: 645-652,1988. J. R., G. P. BURNS,

C. RUMPH, AND W. G. SCHENK, JR. Carnitine levels in severe infection and starvation: a possible key to the prolonged catabolic state. Surgery St. Louis 68: 175-179,

1970. 4. BRADY,

L. J., AND P. S. BRADY. Regulation of carnitine palmitoyltransferase synthesis in spontaneously diabetic BB Wistar rats.

Diabetes 5. BRADY,

38: 65-69,

Biophys. 7. CHEN,



L. J., AND C. L. HOPPEL. Hepatic mitochondrial function in lean and obese Zucker rats. Am. J. Physiol. 245 (Endocrinol. Metab. 8): E239-E245, 1983. J. The effect of fasting on the activity of liver carnitine 6. BREMER, palmitoyltransferase and its inhibition by malonyl-CoA. Biochim. 665: 62%631,198l.

W. J. Utilization of exogenous fat emulsion (intralipid) in septic rats. J. Parenter. Enterczl Nutr. 8: 14-17, 1984. 8. COOK, G. A., AND M. S. GAMBLE. Regulation of carnitine palmitoyltransferase by insulin results in decreased activity and decreased apparent Ki values for malonyl-CoA. J. Biol. Gem. 262: 2050-2055,1987. 9. COOK, G. A.,

D. A. OTTO, AND N. W. CORNELL. Differential inhibition of ketogenesis by malonyl-CoA in mitochondria from fed and starved rats. Biochem. J. 192: 955-958, 1980. J. F., D. ROBIN, P. ROBIN, P. FERRE, AND J. GIRARD. 10. DECAUX, Intramitochondrial factors controlling hepatic fatty acid oxidation at weaning in the rat. FEBS Lett. 232: 156-158,1988. R. W. Mitochondrial respiratory control and the 11. ESTABROOK, polarographic measurement of ADP:O ratios. Methods Enzymol. 10: 41-47,1967. K. R., M. SOUED, I. STAPRANS, 12. FEINGOLD, DONAHUE, B. HUANG, A. H. MOSER, R. GULLI,


M. E.


Effect of tumor necrosis factor (TNF) on lipid metabolism in the diabetic rat: evidence that inhibition of adipose tissue lipoprotein lipase activity is not required for TNF-induced hyperlipidemia. J. Clin. Inuest. 83: 1116-1121, 1989. 13. FRITZ, I. B., AND K. T. N. YUE. Long-chain

carnitine acyltransferase and the role of acylcarnitine derivatives in the catalytic increase of fatty acid oxidation induced by carnitine. J. Lipid Res. 4: 279-

288,1963. AND L. 14. FRY, D. E., B. B. SILVER, R. D. RINK, L. R. VANARSDALL, M. FLINT, JR. Hepatic cellular hypoxia in murine peritonitis. Surgery St. Louis 85: 652-661,1979. 15. GARLAND, P. B., D. SHEPHERD, D. G. NICHOLLS, D. W. YATES, AND P. A. LIGHT. Interactions between fatty acid oxidation and the tricarboxylic acid cycle. In: Citric Acid Cycle, edited by John

M. Lowenstein. New York: Dekker, 1969, p. 163-212. M., AND M. J. H. GEELEN. Short-term regulation of carnitine palmitoyltransferase activity in isolated rat hepatocytes.


Biochem. Biophys. Res. Commun. 17. HARGROVE, D. M., G. J. BAGBY,

151: 781-787,1988.

C. H. LANG, AND J. J. SPITZER. Adrenergic blockade prevents endotoxin-induced increases in glucose metabolism. Am. J. Physiol. 255 (Endocrinol. Metab. 18): E629-E635,1988. 18. HOPPEL, C., J. P. DIMARCO, AND B. TANDLER. Riboflavin and rat hepatic cell structure and function: mitochondrial oxidative metabolism in deficiency states. J. Biol. Chem. 254: 4164-4170, 1979. 19. KELLER, U., P. P. G. GERBER, AND W. STAUFFACHER. Fatty acidindependent inhibition of hepatic ketone body production by insulin in humans. Am. J. Physiol. 254 (Endocrinol. Metab, 17): E694-E699,1988. 20. LOWE, D. M., AND P. K. TUBBS. Succinylation and inactivation of 3-hydroxy-3-methylglutaryl-CoA synthase by succinyl-CoA and its

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.



possible relevance to the control of ketogenesis. Biochem. J. 232: 37-42,1985. 21. MCGARRY,

J. D., G. F. LEATHERMAN, AND D. W. FOSTER. Carnitine palmitoyltransferase I: the site of inhibition of hepatic fatty acid oxidation by malonyl-CoA. J. Biol. Chem. 253: 4128-4136,

1978. 22. MCGARRY,

23. 24.



J. D., S. E. MILLS, C. S. LONG, AND D. W. FOSTER. Observations on the affinity for carnitine, and malonyl-CoA sensitivity, of carnitine palmitoyltransferase I in animal and human tissues: demonstration of the presence of malonyl-CoA in nonhepatic tissues of the rat. Biochem. J. 214: 21-28, 1983. MIZIORKO, H. M. 3-Hydroxy-3-methylglutaryl-CoA synthase from chicken liver. Methods Enzymol. 110: 19-26, 1985. NEUFELD, H. A., J. G. PACE, M. V. KAMINSKI, D. T. GEORGE, P. B. JAHRLING, R. W. WANNEMACHER, JR., AND W. R. BEISEL. A probable endocrine basis for the depression of ketone bodies during infectious or inflammatory state in rats. Endocrinology 107: 596601,198O. ONTKO, J. A. Metabolism of free fatty acids in isolated liver cells: factors affecting the partition between esterification and oxidation. J. Biol. Chem. 247: 1788-1800,1972. OSMUNDSEN, H., AND K. BJ~RNSTAD. Inhibitory effects of some long-chain unsaturated fatty acids on mitochondrial P-oxidation: effects of streptozotocin-induced diabetes on mitochondrial @oxidation of polyunsaturated fatty acids. Biochem. J. 230: 329-337,


carnitine palmitoyltransferase I. Biochem. J. 241: 475-481, 1987. W., T. K. D. GUPTA, G. S. Moss, AND L. M. NYHUS. Effect of endotoxemia on liver cell mitochondria in man. Ann. Surg. 171: 875-882,197O. 32. SHANGRAW, R. E., F. JAHOOR, H. MIYOSHI, W. A. NEFF, C. A. STUART, D. N. HERNDON, AND R. R. WOLFE. Differentiation between septic and postburn insulin resistance. Metabolism 38: 983-989,1989. 33. SMITH, R.,

D. J. FULLER, J. H. WEDGE, D. H. WILLIAMSON, AND K. G. G. M. ALBERTI. Initial effect of injury on ketone bodies and other blood metabolites. Lancet 1: l-3, 1975. H. F., JR., R. S. WARREN, M. JEEVANANDAM, J. L. 34. STARNES, GABRILOVE, W. LARCHIAN, H. F. OEITGEN, AND M. F. BRENNAN. Tumor necrosis factor and the acute metabolic response to tissue 1988. injury in man. J. Clin. Inuest. 82: 1321-1325, 35. TAKEYAMA, N., N. MATSUO, D. TAKAGI, AND T. TANAKA. Determination of overt carnitine palmitoyltransferase by reversed-phase high-performance liquid chromatography. J. Chromatogr. 491: 6976, 1989. 36. TAKEYAMA, TANAKA.



1988. 28. SAGGERSON, E. D., AND J. J. WRIGHT.


224: 201-206,1984. SAGGERSON, E.


M. I. BIRD, C. A. CARPENTER, K. A. WINTER, Cycloheximide blocks changes in rat liver carnitine palmitoyltransferase I activity in starvation. Biochem. J.

D., AND C. A. CARPENTER. Effects of fasting and malonyl-CoA on the kinetics of carnitine palmitoyltransferase and carnitine octanoyltransferase in intact rat liver mitochondria. FEBS I&t. 132: 166-168,198l. SCHOFIELD, P. S., T. J. FRENCH, AND M. C. SUGDEN. Ketonebody metabolism after surgical stress or partial hepatectomy: evidence for decreased ketogenesis and a site of control distal to



1985. 27. QUANT,

P. A., P. K. TUBBS, AND M. D. BRAND. Mannoheptulose and glucagon treatment of fed rats stimulates 3-hydroxy-3-methylglutaryl-CoA synthase activity in rat liver mitochondria by desuccinylation of the enzyme. Biochem. Sot. Trans. 16: 633-634,



N., D. TAKAGI, N. MATSUO, Y. KITAZAWA, AND T. Altered hepatic fatty acid metabolism in endotoxicosis: effect of L-carnitine on survival. Am. J. Physiol. 256 (Endocrinol. Metab. 19): E31-E38, 1989. TRACEY, K. J., B. BEUTLER, S. F. LOWRY, J. MERRYWEATHER, S. WOLPE, I. W. MILSARK, R. J. HARIRI, T. J. FAHEY III, A. ZENTELLA, J. D. ALBERT, G. T. SHIRES, AND A. CERAMI. Shock and tissue injury induced by recombinant human cachectin. Science Wash. DC 234: 470-474,1986. VARY, T. C., J. H. SIEGEL, T. NAKATANI, T. SATO, AND H. AOYAMA. A biochemical basis for depressed ketogenesis in sepsis. J. Trauma 26: 419-425,1986. WANNEMACHER, R. W., JR., J. G. PACE, F. A. BEALL, R. E. DINTERMAN, V. J. PETRELLA, AND H. A. NEUFELD. Role of the liver in regulation of ketone body production during sepsis. J. Clin. Invest. 64: 1565-1572,1979. WOLFE, R. R., J. H. F. SHAW, AND M. J. DURKOT. Effect of sepsis on VLDL kinetics: responses in basal state and during glucose infusion. Am. J. Physiol. 248 (Endocrinol. Metab. 11): E732-E740, 1985. ZENSER, T. V., F. R. DERUBERTIS, D. T. GEORGE, AND E. J. RAYFIELD. Infection-induced hyperglucagonemia and altered hepatic response to glucagon in the rat. Am. J. Physiol. 227: 12991305,1974.

Downloaded from by ${individualUser.givenNames} ${individualUser.surname} ( on September 13, 2018. Copyright © 1990 American Physiological Society. All rights reserved.

Altered hepatic mitochondrial fatty acid oxidation and ketogenesis in endotoxic rats.

Rat hepatic mitochondrial function, including oxidative phosphorylation, fatty acid oxidative capacity, kinetic parameters of carnitine palmitoyltrans...
2MB Sizes 0 Downloads 0 Views