Biochimica et Biophysics Acta, 1124( 1992) 71-79 0 1992 Elsevier Science Publishers B.V. All rights reserved

BBALIP

71 0005-2760/92/$05.00

53841

Formation and degradation of dicarboxylic acids in relation to alterations in fatty acid oxidation in rats Per Br@bech Mortensen Departinetzt of ~edicirze A, Section of ~a.str~)etzt~r~~lo~ attd Hepatology, Rigs~l~spitai~t~ Unir,ersityof Copenhagen, Copenhagen ~Den~lark~ (Received

Key words:

Dicarboxylic

acid; Omega

oxidation:

19 August

19911

Beta oxidation; Ketosis; (Rat peroxisome)

Reye’s syndrome;

Valproate;

(Rat mitochondria);

Dicarboxylic acids are excreted in urine when fatty acid oxidation is increased (ketosis) or inhibited (defects in &oxidation) and in Reye’s syndrome. w-Hydroxylation and w-oxidation of C,-C,, fatty acids were measured by mass spectrometry in rat liver microsomes and homogenates, and /?-oxidation of the dicarboxylic acids in liver homogenates and isolated mitochondria and peroxisomes. Medium-chain fatty acids formed large amounts of medium-chain dicarbo~lic acids, which were easily P-oxidized both in vitro and in vivo, in contrast to the long-chain Cam-dicarbo~lic acid, which was toxic to starved rats. Increment of fatty acid oxidation in rats by starvation or diabetes increased C,: C,, dicarboxylic acid ratio in rats fed medium-chain triacylglycerols, and increased short-chain dicarboxylic acid excretion in urine in rats fed medium-chain dicarboxylic acids. Valproate, which inhibits fatty acid oxidation and may induce Reye like syndromes, caused the pattern of C,-C,,-dicarboxylic aciduria seen in P-oxidation defects, but only in starved rats. It is suggested, that the origin of urinary short-chain dicarboxylic acids is w-oxidized medium-chain fatty acids, which after peroxisomal /3-oxidation accumulate as C,-C,-dicarboxylic acids. C,,-C,,-dicarboxylic acids were also metabolized in the mitochondria, but did not accumulate as C,-Cs-dicarboxylic acids, indicating that /3-oxidation was completed beyond the level of adipyl CoA.

Introduction

C~-C~-dicarbo~lic (adipic, DCA6 and suberic, DCA8) acids (DCAs) are excreted in urine during starvation and diabetic ketosis where fatty acid oxidation is increased [1,2]. Dicarboxylic aciduria is, on the other hand, also associated to defects in the p-oxidation of fatty acids, which may present as attacks of Reye like syndromes often precipitated by fasting. Typical clinical findings are lethargy, hypoglycemia, acidosis, elevated free fatty acids, reduced ketogenesis, hepatic steatosis and glycogen depletion [3-71. The patogenesis of Reye’s syndrome is unknown, but an impairment of mitochondrial function is probably important (81. Long-chain DCAs have been demonstrated to alter mitochondrial function [93 and accumulate in the serum from patients with Reye’s syndromes [lo]. The physiological importance of DCA formation is unknown, but it has been suggested that DCAs may

represent a gluconeogenic pathway from fatty acids by their ~-oxidation to succinyl CoA [ll-131. The present study investigated the chain length dependency of the o-oxidation of fatty acids (monocarboxylic acids, MCAs) and the P-oxidation of DCAs in vivo and in vitro in rats, and focused on the influence of an increased (starvation and diabetes) and inhibited (valproate) fatty acid oxidation on DCA formatioii and toxicity. Materials

and methods

Chemicals

I-‘4~-Iabelled fatty acids were purchased from Amersham (Bu~kingbamshire, U.K.) and New England Nuclear (Boston, U.S.A.). [1-‘4C]Glutaric acid was from ICN Pharmaceuticals (Irvine, U.S.A.). Sodium valproate was from Orion (Finland). Other chemicals were obtained as described before [13]. Animals and in &lo experiments

Correspondence: Rigshospitalet,

P.B. Mortensen, Department of Medicine A 2151, Blegdamsvej 9, DK-2100 Copenhagen 0 Denmark.

Female rats of Wistar strain weighing 200-250 g were placed in individual cages designed for urine

72 collection. Stock diet has been described before [13]. Food was withdrawn from starved rats 48 h prior to start of experiments. Diabetic rats were injected 100 mg streptozotocin/kg body weight 48 h before investigations took place. Unstable diabetes was ensured by severe glucosuria and ketonuria (Table III). None of the rats were treated with insulin. Valproate (500 mg/kg of sodium valproate) or water (controls) were given to fed or 72 h starved rats by stomach intubation. In experiments with radioisotopes, 250 PCi “C fatty acid of chain length C,-C,, was given together with 1 mmol of the correspondent non-radioactive fatty acid (sodium salt; pH 8) in an aqueous solution or suspension by stomach intubation at 9 a.m. and urine was collected for 24 h. Investigations of toxicity were performed by administration of 1 mmol of dicarboxylic acid (sodium salt; pH 8) by stomach intubation. Each chain length (C,C,,) was tested on four fed and four 48 h starved rats, and the condition of the rats was evaluated 24 h later (good, coma or dead). Analysis Dicarboxylic acids were extracted from acidified urine with ethylacetate and silylated with bistrimethylsilyltrifluoroacetamide/l% trimethylchlorosilane before gas chromatographic/mass spectrometric determination [14]. The mass spectrometer was run in the selected ion monitoring mode and the ions were A415: Adipic acid 275.1 m/z, suberic acid 303.1 m/z, sebacic acid 331 .l m/z, dodecanedioic acid 359.1 m/z, and diethylglutaric acid (internal standard) 3 17.1 m/z. Radioactive isotopes were measured by radiogas chromatography with [‘“Clglutaric acid as internal standard [151. w-Hydroxyla tion in isolated rat liter microsomes The biosynthesis of w-hydroxyfatty acids in isolated rat liver microsomes was measured essentially as described before [16]. Livers were chilled immediately after death and 20% homogenates (100 mmol/l phosphate buffer; pH 7.5) were prepared using a PotterElvehjelm homogenizer. Microsomes were prepared by two centrifugations (105000 x g for 60 min; 0-5”C), and washed in between with phosphate buffer before resuspension in buffer. Assays were incubated 30 min at 37°C containing 0.3 mg microsomal protein (3.0 ml; pH 7.5), NADPH 1.7 mmol/l, phosphate 100 mmol/l, KC1 200 mmol/l and C,-C,2 fatty acids in appropriate concentrations (0.1-20 mmol/l) for K, determinations. The reaction was stopped with barium hydroxide and the trimethylsilylated w-hydroxyfatty acids monitored by selected ion monitoring CM+-- 15): 6-hydroxyhexanoic acid 261.1 m/z, 8-hydroxyoctanoic acid 289.1 m/z, IO-hydroxydecanoic acid 333.1 m/z, 12-hydroxy-

dodecanoic acid 361.1 m/z. The were as described before [14,16].

GC/MS

conditions

o-Oxidation in 10000 X g rat lilvr homogenates The complete w-oxidation of fatty acids to the corresponding dicarboxylic acids was assayed in the postmitochondrial fraction. Immediately after death the liver was chilled to 0-4°C in a potassium phosphate buffer (100 mmol/l; pH 7.51, homogenized and centrifugated at 10000 x g for 30 min. The supernatant was frozen to - 20°C until use. Assays contained 1.3 ml supernatant (approx. 30 mg protein), NADPH 1.4 mmol/l, NADf 1.2 mmol/l, and fatty acids of chain length C,-C,, in concentrations of 1.2 or 4.8 mmol/l. The reaction was stopped by addition of saturated barium hydroxide, hydrolyzed 18 h at 110°C and neutralized by sulphuric acid and carbon dioxide. Barium sulphate and proteins were removed by 2000 x g centrifugation and dicarboxylic acids determined in the supernatant solution as in urine. /I-Oxidation in 600 X g rat lirler homogenates The P-oxidation of dicarboxylic acids in freshly isolated postnuclear fractions (600 x g> of rat liver homogenates was performed in assays (approx. 35 mg protein/2.7 ml; pH 7.5) containing KC1 100 mmol/l, Hepes 20 mmol/l, MgCl, 10 mmol/l, EDTA 1 mmol/l, KH,PO, 5 mmol/l, sucrose 25 mmol/l, ATP 5 mmol/l, CoASH 0.1 mmol/l, L-carnitine 2 mmol/l, NAD 1 mmol/l, cytochrome c 10 pug/ml, dithiothreitol 12 mmol/l, 0.01% Triton X-100 100 ~1, and dicarboxylic acids 1 mmol/l of chain length as tabulated. Termination and dicarboxylic acid measurements were as described above. @Oxidation in rat lil’er mitochondria and peroxisomes Mitochondria and peroxisomes from livers of ciofibrate-treated rats were separated by a combination of differential and sucrose density gradient centrifugation as described previously [ 171. Separation was ensured by the marker enzymes, cytochrome c oxidase and urate oxidase, respectively [ 181. Dicarboxylic acids (0.1 mmol/l) were incubated at 37°C with 200 ~1 organelle suspension (peroxisomes, approx. 3 mg/ml protein; mitochondria, approx. 1 mg/ml protein) in a reaction mixture (1.2 ml, pH 7.4) containing KCI 90 mmol/l, MgCI, 10 mmol/l, KH,PO, 25 mmol/l, sucrose 40 mmol/l, ATP 10 mmol/l, CoASH 5 mmol/l, NAD 5 mmol/l, L-carnitine 1.5 mmol/l, flourocitrate 25 bmol/l and dithiothreitol 5 mmol/l. Termination, hydrolysis and the measurement of dicarboxylic acids were as described above. Statistics Analysis of variance (one-way ANOVA), the test of least significant differences, rank test for paired and

73 TABLE

unpaired samples, and t-test for unpaired samples were used to evaluate. data.

I

Apparent K, (mean + S.E.) for the ~teracf~un of C,-Ct,-monocarboxylic acids (MCAs) with the hepatic w-hydro~lation systems in isolated micosomes from four rats

Results

Substrate

Product

Hexanoic acid (MCAG) Octanoic acid (MCAN) Decanoic acid (MCAlO) Dodecanoic acid (MCAI 2)

6-Hydroxyhexanoic acid 8Hydoxyoctanoic acid IO-Hydroxydecanoic acid 12-Hydroxydodecanoic acid

o-Oxidation of monocarboxylic acids (MCAS)

Table I illustrates that medium-chain MCAlO and MCA12 is easily o-hydroxylized in isolated rat liver microsomes as indicated by Km-values in the ,umol/l range, compared to the modest affinity and much larger K,-values in the mmol/l range for MCA8 and MCA6.

TABLE

0.2

_irO.Ol

If

w-Oxidation of C,-C,,-monocarboxylic homogenates Monocarboxylic

None Hexanoic acid Octanoie acid Decanoic acid Laurie acid

TABLE

8.8 k 3.0 8.2 &-1.0 0.003 + 0.001

acids added

acids (MCAs) to C,-C,,-dicarboxylic

Dicarboxylic

to assays (mmol/lJ

0 4.8 4.8 1.2 1.2

(MCA6) (MCA8) (MCAlO) (MCAl2)

acids (DCAs) in the 10000 x g supernatant fraction of rat liL%er

acids produced

in assays (pmol/l;

N = 4; mean

+ S.E.)

adipic (DCA6)

suberic (DCAS)

sebacic (DCAIOJ

dodecanedioic (DCA12)

5$1 9+0 7*0 4io 4+0

2*0 1+0 9+0 1+0 1*0

l&O llt0 l&O 191+6 1*0

1+0 1*0 1+0 2+0 71+5

III

Radioactive recovery (o/00) in 3-hydroxybutyric and C, -Cl,-dicarboxylic acids (DCAS) in urine from unstarved (A), starved (B) and diabetic ketotic (C) rats administered 250 pCi C, -C,,-[1-‘4Clmonocarboxylic acids (MCAS) by stomach intubation t4C-MCAs to rats

3-hydroxybutyric

MCA6 MCAS MCAlO MCA12 MCA14 MCA16

TABLE

A

B

0.4 1.3 0 0

1.5 9.7 0.5 0.1

Adipic

acid C

9.9 1.7 4.8 1.0

no MCT: fed or diabetic of the rats

Fed No MCT Starved

A

B

0.3 0.8 0.4 0

0.5 0.7 1.0 0.5

Suberic C

2.0 0.7 0.2 0

Sebacic

acid (DCAS)

A

B

0 0.3 0.3 0

0 0.2 0.4 0.3

C

0.6 0.1 0 0

acid (DCAIO)

A

B

0 0 0.5 0

0 0 0.4 0

C

0.2 0 0 0

IV

Urinary excretion of C,-C,,-dicarboxylic triacylglycerol (MCT)

Condition (N=4)

acid (DCA6)

ketotic

Diabetic ketotic No MCIP (ANOVA)

acids ~DCAS) and 3-hydroxybutyric acid in fed, starved and diabetic rats administered 2 ml medium-chain

rats not given medium-chain IJrinary

excretion

triacylgfycerol.

(mmol/mmol

creatine;

P: one-way

ANOVA

comparing

fed, starved

Mean S.E.)

adipic (DCM)

suberic (DCASJ

sebacic (DCAIO)

0.16 + 0.09 < 0.20

0.56 f 0.09 < 0.01

0.17 f 0.09 < 0.01

0.27 kO.07 < 0.005

4.7 kO.7

1.3 kO.2

< 0.01

*14

0.31+ 0.04

rats given MCX.

ratio

hydro~bu~ric acid

51 30

and diabetic

0.21

DCA6:DCAlO

*to.07

2*1

8*2

0.86f0.18 < 0.13

0.14+0.05 < 0.02

0.006 rt 0.002 < 0.001

138f2

< 0.02

< 0.02

< 0.05

< 0.001

74 TABLE

V

~-Oxidation of CT,-C,,-dicarboxylic acids in vi1.o ec,aluated by the urinary excretion of C,-C,,,administered C, - C,, -dicarhoxylic acids by stomach intuhation

dicarhoxylic acids in fed and starrled rats

* P, signed rank test for paired (one-way ANOVA).

dicarboxylic

samples.

Dicarboxylic acids given to rats (I mmol; N=4)

None (controls) Suberic acid (DCAX) Sebacic acid (DCAIO) Dodecanedioic acid (DCAl2) Tetradecanedioic acid (DCAl4) Hexadecanedioic acid (DCA16)

* * P. analysis

Urinary

of variance

excretion

cornpairing

of dicarboxylic

suberic

fed

starved

7i 1 x5* 22 1400 * 330 1700?290 1500 & 330 70+ IS

7+ 1 l60+ 40 2500+ 330 7600k 1100 2300+ 440 toxic

< 0.01

4*

mean

acid (DCA8)

I

acids of different

7*

acid(DCAl0)

fed 2

46Ok 7s 2600 k 600 470* 150 toxic

starved

2+

I

< 0.001

3f

140+34 28+ 6 x* 2

< 0.001 < 0.001

730 * 74 x4* 13 toxic < 0.01

< 0.01

< 0.01

VI

Dicarboxylic acids added (1 mmol/l: N = 4)

to assays

None Suberic acid (DCAH) Sebacic acid (DCAlO) Dodecanedioic acid (DCA12) Tetradecanedioic acid (DCA14) Hexadecanedioic acid (DCA16)

acids (DCAS) in the fresh isolated postnuclear fractions of rat liver homogenates Dicarboxylic adipic (DCAh) 4*0 15*1 16+1 23 + 1 4+0 3+0

1

< 0.00 I

CC,-C,,,) triacylglycerol to create a substantial dicarboxylic aciduria. Starvation and diabetic ketosis increased DCA6 excretion compared to fed rats, and ketosis almost eliminated DCAlO excretion even though the rats were given medium-chain triacylglycerol. Therefore, the degree of ketosis (and rate of p-oxidation) correlated to ratios of excreted DCA6: DCAlO. The chain length dependency of the p-oxidation of DCAs had, both in vivo (Table V) and in vitro (Table VI), a maximum at DCA12, with decreasing p-oxidability for shorter and longer DCAs. Again the excretions of short-chain DCAs were higher in starved compared to fed rats (Table V). DCA16 appeared to be a poor substrate for /?oxidation both in vivo and in vitro, and feeding of 1 mmol to four 48 h starved rats was fatal to two and resulted in coma in the remaining two within 24 h. In contrast, fed rats given DCA16 and starved or fed rats receiving the same dose of DCA of shorter (C,-C,,) chain length were unaffected and fully awake.

P-Oxidation of dicarboxylic acids (DCAS) Table IV shows the relationship between the activity of the lipid metabolism and the pattern of DCA excretion in rats, which were fed 2 ml (1.6 g> medium-chain

~-Oxidation of C,-C,,-dicarboxylic

chain length

+S.E.) sebacic

starved

340+ 46 610 of-100 190* 46 19* 5

The same pattern was seen when the complete woxidation was measured in the postmitochondrial (10000 X g> fraction (Table II). The addition of 1.2 mmol/l of MCAlO and MCA12 to the assay resulted in a substantial production of DCAlO and DCA12. In contrast, 1.2 mmol/l (not shown) and 4.8 mmol/l of MCA6 and MCA8 were converted to small amounts of DCA6 and DCA8 only. The radioactive recoveries in urinary excreted C,C,,,-DCAs were highest in rats administered radioactive medium-chain MCAs (Table III). MCAlO gave the maximal recoveries, which were further increased by starvation and diabetic ketosis. MCAs of lower and especially higher (C,,-C,,) chain length resulted in comparable lower radioactive recoveries of short-chain DCAs in the urine.

TABLE

creatinine;

fed

< 0.001

* P chain length dependence

administered

acids (~mol/mmol

adipic acid (DCA6)

* P fed vs. starved

groups

acids produced

(pmol/l;

suberic (DCAB) 2*0 28 + 1 50*1 3*0 250

mean

f SE.) sebacic (DCAlO)

dodecanedioic (DCA12)

1+0

2*0

160+4 4+0 1 *o

29+0 5+0

75 ~itoc~o~dria~ boxy&c acids

and ~~roxiso~ai

~-oxidation

of dicar-

Fig. 1 illustrates the P-oxidation of C,-C,,-DCAs in isolated mitochondria and peroxisomes. From the initial addition of 100 pmol/l, DCA12 decreased to 48 and 27 pmol/l after 5 min (P < O.OOOl),and to 39 and 8 ,umol/l after 10 min of incubation in mitochondria and peroxisomes, respectively. The metabolization was obviously caused by a P-oxidation of DCA12 to DCAlO,

DCA8 and DCA6, which all increased during the incubation (Fig. 1, left column). The first enzyme in the P-oxidation sequence is known to be rate Limiting, and accumulation of par& oxidized inte~edia~ DCAs is probably unlikely. Based on this assumption, production of succinyl CoA by a further p-oxidation of adipyl-CoA is indicated if the total amount of C,-C,,-DCAs decreases with time of incubation.

C,2 -dicarboxylic acid

Cro-dicarboxylic acid 90

o mitochondria . peroxisomes

70 50 30

ca-dicar~xyiic

acid

30

C, ~dicar~xyii~ acid

-1

I

50

50

I 1

30 1-/- I 20 -1IO

I

110 \

clo+%+%

-1

!

E&q-;

-I

60 70 60

incubation Fig. 1. &Oxidation of C,-C12-dicarboxylic

-1

1‘ I!\

"** -f fl

time (minutes)

acids fpmol/l rt SD.) in isolated mitochoffdria and ~roxisomes. **P