1580

Effect of 3-Thiadicarboxylic Acid on LipidMetabolism in Experimental Nephrosis

Ayman AL-Shurbaji, Jon Skorve, Rolf K. Berge, Mats Rudling,Ingemar Bjorkhem, Lars Berglund

The effect of the sulfur-substituted fatty acid analogue 1,10 6i?(carboxymethylthio)decane, also known as3-thiadicarboxylic acid, on puromycin aminonucleoside-induced nephrotic hyperlipidemia was studied inrats. Treatment with 3-thiadicarboxylic acid (250 mg/kg) for 5 days reduced plasma levels of triglyceridesfrom 5.8 to 2.7 mmol/L and cholesterol from 11.0 to 7.7 mmol/L. This was accounted for by decreases invery-low-density lipoprotein triglycerides, very-low-density lipoprotein cholesterol, and low-density lipo-protein cholesterol, without any major changes in the composition of plasma lipoproteins. The activitiesof two enzymes involved in fatty acid synthesis (ATP:citrate lyase and fatty acid synthetase) were inhibitedby 3-thiadicarboxylic acid treatment, whereas acetyl-coenzyme A carboxylase activity was unchanged. Incontrast, treatment with the sulfur-substituted fatty acid analogue induced the peroxisomal /3-oxidationof fatty acids ninefold and the mitochondrial /3-oxidation by 54% to 73%, depending on the substrate used.This was accompanied by a 26% reduction in hepatic triglyceride secretion rate. The hepatic phosphati-date phosphohydrolase activity was unchanged. 3-Thiadicarboxylic acid treatment suppressed the activityof the rate-limiting enzyme in cholesterol biosynthesis, 3-hydroxy-3-methylglutaryl-coenzyme A reduc-tase, by 58%, whereas hepatic LDL receptor expression was unaltered. The activities of lipoprotein lipaseand hepatic lipase were unchanged by treatment. These results demonstrated that treatment with3-thiadicarboxylic acid ameliorates hyperlipidemia in experimental nephrosis primarily by decreasing theoverproduction of very-low-density lipoprotein present. The data also indicate that hepatic very-low-density lipoprotein synthesis and secretion is strongly influenced by the availability of the fatty acidsubstrate under the same hyperlipidemic conditions. (Arterioscler Thromb. 1993;13:1580-1586.)

KEY WORDS • triacylglycerol • 3-hydroxy-3-methylglutaryl-coenzyme A reductase • hypolipidemicdrugs • lipoprotein synthesis • cholesterol • /3-oxidation

Hyperlipidemia is a consistent feature of thenephrotic syndrome.1-2 Patients with nephro-sis usually have elevated plasma levels of both

cholesterol and triglycerides and might therefore be athigher risk for developing premature atherosclerosisleading to ischemic heart disease and progressive renalfailure.36 Because of the growing recognition of thisrisk, several studies have assessed the efficacy of differ-ent lipid-lowering drugs in nephrotic hyperlipidemia inhumans as well as in experimental animal models.79

Recently, a new group of hypolipidemic agents hasemerged, namely the sulfur-substituted non-/3-oxidiz-able fatty acid analogues.10 These compounds havebeen shown to reduce plasma triglyceride and choles-terol levels in normolipidemic rats.10-11 We have dem-onstrated12 that the lipid-lowering properties of theseagents in normal rats are mediated by effects on hepaticvery-low-density lipoprotein (VLDL) production. Since

Received March 18, 1993; revision accepted July 9, 1993.From the Departments of Clinical Chemistry (A.A.-S., I.B.,

L.B.) and Medicine (M.R.) and the Center for Nutrition andToxicology, Novum (M.R.), Karolinska Institutet, Huddinge Uni-versity Hospital, Huddinge, Sweden, and the Laboratory of Clin-ical Biochemistry (J.S., R.K.B.), University of Bergen, HaukelandHospital, Bergen, Norway.

Correspondence to Lars Berglund, MD, Department of ClinicalChemistry, Huddinge University Hospital, S-14186 Huddinge,Sweden.

increased lipoprotein synthesis and secretion has beensuggested as a major underlying mechanism in ne-phrotic hyperlipidemia,1318 we investigated the effect ofthe sulfur-substituted fatty acid analogue 3-thiadicar-boxylic acid on lipid metabolism in the nephrotic hyper-lipidemic rat.

MethodsAnimals and Animal Treatment

Male Sprague-Dawley rats weighing 200 to 250 g werehoused individually in metal cages and maintained on a12-hour light-dark cycle. After 1 week of acclimatiza-tion, nephrosis was induced by a single intraperitonealinjection of puromycin aminonucleoside (100 mg/kgbody wt; Sigma Chemical Co, St Louis, Mo). Nephrosiswas confirmed by measuring urinary albumin loss (>250mg/24 h as determined by an immunonephelometricprocedure; Nordic Immunology, Tilburg, The Nether-lands) and plasma lipid levels 8 days after injection. Theanimals were treated with 3-thiadicarboxylic acid fromday 9 through day 14 after the induction of nephrosis.3-Thiadicarboxylic acid was suspended in 0.5% sodiumcarboxymethylcellulose (CMC) and administered bygavage in a daily dose of 250 mg/kg body wt. Controlrats received only CMC. All animals had free access tofood and water throughout the experiment. There wasno significant difference in food consumption or weight

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

AL-Shurbaji et al 3-Thiadicarboxylic Acid in Nephrotic Hyperlipidemia 1581

gain between the two experimental groups. On day 15after injection, blood sampling was performed by heartpuncture under neuroleptic anesthesia (Hypnorm, Jans-sen Pharmaceutica, Beerse, Belgium), and the animalswere subsequently killed. For lipoprotein lipase andhepatic lipase assays a separate group of rats wasintravenously injected with heparin 1000 U/kg body wt 8to 10 minutes before blood sampling.

Enzyme AssaysLiver tissue from individual rats was homogenized in 4

volumes of ice-cold sucrose buffer (0.3 mol/L) containing50 mmol/L tris(hydroxymethyl)aminomethane-HCl, pH7.4,10 mmol/L EDTA, 50 mmol/L NaCl, and 10 mmol/Ldithiothreitol. Cytosolic, mitochondrial, microsomal, andperoxisomal fractions were prepared by centrifugation asdescribed elsewhere.19-20 The following microsomal en-zyme activities were assayed as described previously:glycerol 3-phosphate acyltransferase21-22; diacylglycerolacyltransferase23; 3 -hydroxy-3 -methylglutaryl-coenzymeA (HMG-CoA) reductase24; cholesterol 7a-hydroxy-lase25; and acyl-CoAxholesterol acyltransferase.26

ATPxitrate lyase,27 acetyl-CoA carboxylase,28 andfatty acid synthetase12-29 were assayed in the cytosolicfraction as described previously. For assay of phos-phatidate phosphohydrolase activity, a cytosolic frac-tion was precipitated by 40% ammonium sulfate, andthe enzyme activity was assayed as previously de-scribed.30-31 Mitochondrial and peroxisomal /3-oxida-tion as well as fatty acyl-CoA oxidase activity weremeasured by the same procedures as detailed else-where.3234 Heparin-releasable lipoprotein lipase andhepatic lipase activities were determined by themethod of Nilsson-Ehle and Schotz.35 In this assay,one unit of enzyme activity corresponds to the releaseof 1 /xmol fatty acid from triolein per minute at 37°C.

Plasma and Hepatic Lipid LevelsLipids were extracted from 20% liver homogenates

using methanol/chloroform (1:2, vol/vol). Triglyceridesand cholesterol were measured in plasma and lipidextracts by using commercially available enzymatic as-says (Boehringer Mannheim, Mannheim, FRG). Plasmafree fatty acids (FFAs) were determined by an enzy-matic colorimetric method (NEFAC, Wako ChemicalsGmbH, Neuss, FRG) as described.36-37

Hepatic VLDL Triglyceride SecretionThe VLDL triglyceride secretion rate was estimated

by measuring the accumulation of triglyceride in plasmaafter a single intravenous injection of Triton WR 1339as described before.38-39

Plasma Lipoprotein AnalysisLipoproteins were determined quantitatively by using

a combination of ultracentrifugation and precipitationas described in detail elsewhere.40-41 Briefly, plasmasamples were centrifuged at a density of 1.006 g/mL for18 hours at 35 000 rpm in a Centrikon T-2060 ultracen-trifuge (Contron Roche, Zurich, Switzerland) equippedwith a 45.6 Ti rotor. Triglycerides and cholesterol weremeasured in both the infranatant (low-density lipopro-tein [LDL]+high-density lipoprotein [HDL]) and thefloating fraction (VLDL). One portion of the infrana-tant was treated with phosphotungstic acid to precipi-

tate apolipoprotein B-containing lipoproteins, and theresulting supernatant was assayed for cholesterol andtriglyceride content. To analyze the composition of eachlipoprotein fraction, the infranatant fractions obtainedafter the initial centrifugation were subjected to furthersequential ultracentrifugation, yielding LDL (1.025 <d< 1.063) and HDL (1.063<rf<1.21) fractions. The iso-lated lipoprotein fractions were then subjected to a"wash" by a repeated ultracentrifugation at the upperdensity level. Before analysis, the lipoprotein fractionswere dialyzed against 0.15 mol/L NaCl containing 0.3mmol/L EDTA. Total cholesterol, free cholesterol, andtriglycerides were determined enzymatically as men-tioned above. Protein was measured by the Lowryprocedure using bovine serum albumin as a standard.42

Phospholipids were analyzed by a modification of themethod described by Bartlett.43

Ligand BlottingLDL receptor expression was measured by a ligand

blot assay as described.44 Briefly, hepatic membraneswere prepared from pooled frozen rat liver samples, andmembrane proteins were separated by sodium dodecylsulfate-polyacrylamide gel electrophoresis, transferredto nitrocellulose filters, and incubated with 125I-labeledrabbit /3-VLDL. After autoradiography of dried filters,LDL receptor bands were cut out, and radioactivity wasmeasured in a gamma counter.

Statistical AnalysisData were evaluated by Student's t test with the level

of statistical significance set at /><.05.

ResultsInduction of nephrosis by puromycin aminonucleo-

side resulted in a marked hyperlipidemia, with increasesin plasma levels of both triglycerides and cholesterol(Table 1). The increase in triglyceride levels was mainlyaccounted for by an increase in the VLDL fraction,whereas cholesterol levels increased in all lipoproteinfractions, albeit most notably in LDL and HDL. Treat-ment of nephrotic rats with 3-thiadicarboxylic aciddecreased plasma triglyceride and cholesterol levels by53% and 30%, respectively (Table 1). This was accom-panied by corresponding reductions of VLDL triglycer-ides (59%), VLDL cholesterol (45%), and LDL choles-terol (39%) (Table 1). HDL cholesterol levels, however,were not affected. Despite its substantial effects onVLDL and LDL lipid levels, treatment with 3-thiadi-carboxylic acid did not influence the composition ofthese particles to any large extent, apart from a relativeincrease in VLDL protein content (Table 2).

Nephrotic rats display an overproduction of VLDLtriglycerides that is indicated by a higher entry rate oftriglycerides into the circulation.45 Since 3-thiadicarbox-ylic acid decreases VLDL triglyceride production innormal rats,12 it was of interest to investigate whetherthis was also the case under the present hyperlipidemicconditions. For this purpose, rats were injected withTriton WR 1339, a detergent that inhibits the action oflipoprotein lipase on triglyceride-rich lipoproteins. Asshown in Fig 1, the accumulation of triglycerides inplasma, which reflects the entry rate of newly synthe-sized hepatic triglycerides into the circulation, wasreduced by 26% in 3-thiadicarboxylic acid-treated rats

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

1582 Arteriosclerosis and Thrombosis Vol 13, No 11 November 1993

TABLE 1. Effect of 3-Thiadicarboxylic Acid on Plasmaand Lipoprotein Lipid Levels in Nephrotic Rats

= 30 T

Non-treatedNephrotic Rats

LipidNormal

RatsNon- 3-Thiadicarboxylic

treated Acid Treated

Plasma

Triglycerides 1.2+0.3 5.8±1.7* 2.7±1.0t

Cholesterol 2.5±0.8 11.0±1.8* 7.7+1.9t

VLDL

Triglycerides 0.55±0.09 5.46±1.08* 2.22±1.01t

Cholesterol 0.21 ±0.06 1.32+0.17* 0.73±0.17t

LDL

Triglycerides 0.18±0.03 0.30±0.13 0.27±0.14

Cholesterol 0.87±0.21 5.16±0.57* 3.17±1.33t

HDL

Triglycerides 0.13±0.03 0.18±0.12 0.10±0.00

Cholesterol 1.37+0.11 4.72±0.73* 4.03±0.93

VLDL indicates very-low-density lipoprotein; LDL, low-densitylipoprotein; and HDL, high-density lipoprotein.

Values for nephrotic rats represent mean+SD for six rats ineach group. Values for control rats represent mean±SD for sixpreparations of pooled plasma from three individual rats in eachpreparation.

*P<.05 vs normal rats.tP<.05 vs nontreated nephrotic rats.

(3.92±0.24 mmol/L per hour versus 5.27±0.38 mmol/Lper hour in controls, P<.05). In parallel, the livercontent of triglyceride but not of cholesterol was re-duced by 3-thiadicarboxylic acid treatment (Table 3).

Treatment with 3-thiadicarboxylic acid stimulatedthe mitochondrial fatty acid /3-oxidation by 54% and73% as measured with palmitoyl-CoA and palmitoyl-L-carnitine as substrates, respectively (Table 4). Also, theperoxisomal /3-oxidation was markedly stimulated (byninefold). This was reflected in a pronounced increasein the activity of peroxisomal acyl-CoA oxidase (Table4). Of the enzymes involved in fatty acid synthesis,

Time after Triton WR 1339 injection (h)FIG 1. Line graph showing effect of 3-thiadicarboxylicacid (TD) on plasma triglyceride entry rate after a singleintravenous injection of Triton WR 1339. The values areexpressed as mean±SD for six rats in each experimentalgroup. • Indicates nontreated rats and A, 3-thiadicarbox-ylic acid-treated rats. To convert millimoles per liter tomilligrams per deciliter, multiply by 88.5.

ATPxitrate lyase and fatty acid synthetase were signif-icantly inhibited by 3-thiadicarboxylic acid treatment(Table 4). Acetyl-CoA carboxylase activity was notaffected. Since the carboxylase is known to be therate-limiting enzyme in fatty acid synthesis, the presentfindings argue against retarded fatty acid synthesis as amajor underlying mechanism for the hypotriglyceri-demic effect observed. Administration of 3-thiadicar-boxylic acid, which decreases plasma FFA levels innormal rats,12 had no effect on FFA in nephrotichyperlipidemic rats (0.45±0.06 mmol/L and 0.41 ±0.06mmol/L in control and treated rats, respectively). Of theenzyme activities involved in triglyceride biosynthesis,glycerol 3-phosphate acyltransferase and diacylglycerolacyltransferase were significantly stimulated by 3-thiadi-carboxylic acid treatment (Table 5). However, in con-

TABLE 2. Effect of 3-Thiadicarboxylic Acid on Lipoprotein Composition in Nephrotic Rats

Lipid

VLDL

Nontreated

TD treated

LDL

Nontreated

TD treated

HDL

Nontreated

TD treated

Triglycerides

66.2

57.4

22.8

22.8

1.4

1.1

UnesterifiedCholesterol

6.3

7.3

6.8

8.3

2.9

3.3

CholesterolEsters

2.5

2.5

23.9

20.1

24.8

24.1

Phospholipids

15.6

18.5

24.2

25.2

31.2

30.7

Protein

9.4

14.3

22.3

23.6

39.7

40.8

VLDL indicates very-low-density lipoprotein; TD, 3-thiadicarboxylic acid; LDL, low-density lipoprotein; and HDL,high-density lipoprotein.

Results represent mean values from two different experiments and are given as percent by weight. In eachexperiment, plasma from six rats was pooled and the respective lipoprotein fractions were isolated as described in"Methods."

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

AL-Shurbaji et al 3-Thiadicarboxylic Acid in Nephrotic Hyperlipidemia 1583

TABLE 3. Effect of 3-Thiadicarboxylic Acid on HepaticLipid Levels in Nephrotic Rats non-treated TD-treatedHepatic Lipid Levels,/imol/g

NontreatedRats

3-ThiadicarboxylicAcid-Treated Rats

Triglycerides

Cholesterol

3.6±0.6

4.5±0.5

2.6±0.3*3.9+0.4

yg proteinper lane

Values represent mean±SD for six rats in each group.*P<.05.

trast to the findings in normal rats, the hepatic cytosolicphosphatidate phosphohydrolase activity was unaf-fected by treatment (Table 5). As shown in Table 1,administration of 3-thiadicarboxylic acid to nephroticrats reduced plasma, VLDL, and LDL cholesterol lev-els. This was accompanied by an inhibition of HMG-CoA reductase activity (58%; Table 6). Cholesterol7a-hydroxylase and acyl-CoAxholesterol acyltransfer-ase activities were depressed by 34% and 28%, respec-tively. Despite the inhibition of HMG-CoA reductaseactivity, the expression of the hepatic LDL receptor wasnot affected (Fig 2).

In view of earlier reports that nephrosis-inducedhyperlipidemia is associated with defective clearance ofVLDL,4546 we investigated whether 3-thiadicarboxylicacid treatment affected the clearance potentiality ofVLDL. As shown in Table 5, lipoprotein lipase andhepatic lipase activities were not significantly influencedby drug treatment.

Discussion

Treatment of nephrotic rats with the sulfur-substi-tuted non-/3-oxidizable fatty acid analogue 3-thiadicar-boxylic acid resulted in a substantial reduction of mark-edly elevated plasma lipoprotein levels and a favorableincrease in the HDL/LDL ratio. These hypolipidemicproperties were mainly due to an effect on the increasedhepatic VLDL synthesis and secretion, an abnormality

200

cutout cpm per1 20-kDa-band

3647

2176|

1272

3424

2170

|l 289 |

FIG 2. Ligand blot of nephrotic control (non-treated) ratsand 3-thiadicarboxylic acid-treated (TD-treated) ne-phrotic rats. Samples of liver tissue from six animals ofeach group were pooled and hepatic membranes wereprepared as described in "Methods." The indicatedamounts of membrane were separated on 6% sodiumdodecyl sulfate-polyacrylamide gel electrophoresis fol-lowed by electrotransfer to a nitrocellulose membrane.The membrane was incubated with 125l-labeled /3-very-low-density lipoprotein (5 /xg/mL). After washing, the filterwas exposed at -70°C for 4.5 hours. The given cutoutcounts per minute values were obtained by counting thegamma radioactivity of excised 120-kDa bands of thefilter. Mean filter background (1605 cpm) was obtained bycounting the radioactivity of four pieces of filter with thesame area from the same membrane; these values weresubtracted from the data presented.

TABLE 4. Effect of 3-Thiadicarboxylic Acid on Fatty Acid Oxidation and Synthesis inNephrotic Rats

Fatty Acid Oxidation andSynthesis

Oxidation

Mitochondrial /3-oxidation,nmol/min per milligram protein

Palmitoyl-L-carnitine as substrate

Palmitoyl-CoA as substrate

Peroxisomal /3-oxidation, nmol/minper milligram protein

Acyl-CoA oxidase, nmol/min permilligram protein

Synthesis

ATP-citrate lyase, nmol/min permilligram protein

Acetyl-CoA carboxylase, nmol/minper milligram protein

Fatty acid synthetase, nmol/minper milligram protein

NontreatedRats

3.4+0.4

2.4+1.3

0.9±0.3

5.6±0.9

16.7±3.6

8.0+2 2

0.23+0.03

3-ThiadicarboxylicAcid-Treated Rats

5.9+1.1*

3.7±1.3*

8.7±1.9*

44.3±8.6*

12.2±1.9*

7.1 ±1.4

0.11+0.04*

CoA indicates coenzyme A. Values represent mean±SD for six rats in each group.*P<.05.

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

1584 Arteriosclerosis and Thrombosis Vol 13, No 11 November 1993

TABLE 5. Effect of 3-Thiadicarboxylic Acid on Key Enzymes in TriglycerideBiosynthesis and Degradation

Enzymes

Glycerol 3-phosphate acyltransferase,nmol/min per milligram protein

Microsomal fraction

Mitochondrial fraction

Diacylglycerol acyltransferase,nmol/min per milligram protein

Cytosollc phosphatidatephosphohydrolase, nmol/min permilligram protein

Lipoproteln llpase, U/L

Hepatic lipase, U/L

NontreatedRats

1.0±0.4

0.6±0.1

3.6±1.0

1.0±0.2

147±29

100±15

3-ThiadlcarboxyllcAcid-Treated Rats

8.4±0.5*

1.2±0.3*

5.4±0.9*

0.9±0.2

184+35

114±13

Values represent mean±SD for six rats in each group.*P<.05.

that has been suggested as an important causative factorin nephrotic hyperlipidemia.1416-47'48 As shown in Fig 1,3-thiadicarboxylic acid treatment decreased hepaticVLDL triglyceride secretion by 26%. Interestingly, thisdecrease was of the same magnitude as the increase inthe VLDL secretion rate reported in experimentalnephrosis,45 indicating that drug treatment rectified theVLDL overproduction present in the nephrotic syn-drome. It is noteworthy that the specific mechanismsunderlying the increase in VLDL synthesis present innephrosis are still largely unknown. Further studies areneeded to elucidate this issue to determine whichdisturbances are specifically corrected by the drug. Thedecrease in VLDL production was obviously insufficientto normalize plasma lipid levels. As shown in Table 1, aconsiderable degree of hyperlipidemia was still presentin 3-thiadicarboxylic acid-treated animals, lending fur-ther evidence to the contention that nephrosis-inducedhyperlipidemia is probably multifactorial.49

Treatment with 3-thiadicarboxylic acid influencedlipid metabolism in the hyperlipidemic rat in a similarway as in normal rats.1112 The most striking effect was amarked stimulation of fatty acid oxidation in bothperoxisomes and mitochondria (Table 4). The stimula-tion of fatty acid oxidation indicated that the druginterferes with the availability of the fatty acid sub-strate. This emphasizes the importance of this factor asa major determinant of triglyceride synthetic rate in thehyperlipidemic state. However, the role of phosphatid-ate phosphohydrolase as a regulatory enzyme under thepresent hyperlipidemic conditions is less obvious. Thisenzyme activity has been shown to be correlated toplasma and liver triglyceride levels and to vary inparallel with changes in plasma triglyceride levels.5053

In normolipidemic rats, the enzyme activity decreasesin parallel with the triglyceride reduction produced by3-thiadicarboxylic acid treatment.12 However, in ne-phrotic rats, 3-thiadicarboxylic acid treatment did notaffect this enzyme activity (Table 5). The reason for thediscrepancy observed between the findings in the nor-molipidemic and the hyperlipidemic situations is notclear at present. It is conceivable that in the nephroticrat, under conditions of deranged lipid metabolism andexcessive plasma lipid levels, a different kind of regula-

tion of the enzyme activity might be operative. Inaddition to phosphatidate phosphohydrolase, diacyl-glycerol acyltransferase54-57 and glycerol 3-phosphateacyltransferase58 are implicated as being of regulatoryimportance in hepatic triglyceride biosynthesis. Thepresent observation that both activities were stimulatedby 3-thiadicarboxylic acid treatment indicates that thetriglyceride-lowering effect of this drug was not medi-ated through either enzyme activity. This finding wouldseem to argue against a rate-limiting role of glycerol3-phosphate acyltransferase or diacylglycerol acyltrans-ferase under the present conditions.

Hypercholesterolemia of the nephrotic syndromemight in part be due to enhanced cholesterol biosynthe-sis,59 which would provide a rationale for the use ofdrugs that inhibit cholesterogenesis to treat nephrotichyperlipidemia.9 Indeed, HMG-CoA reductase inhibi-tors are effective in reducing plasma lipid levels underthese conditions.7"960-61 In the present study, the hypo-cholesterolemic effect of 3-thiadicarboxylic acid wasassociated with a significant inhibition of HMG-CoAreductase activity (Table 6). This suggests that 3-thiadi-carboxylic acid treatment might decrease cholesterolbiosynthesis, thereby contributing to the decrease inVLDL secretion, as discussed above. Whether the ef-fects on HMG-CoA reductase activity and cholesterolsynthesis are secondary to retarded triglyceride produc-tion remains to be established. The finding that 3-thi-adicarboxylic acid treatment also affected other en-zymes in cholesterol metabolism (cholesterol 7a-hydroxylase and acyl-CoAxholesterol acyltransferase)indicated that the drug might affect cholesterol homeo-stasis on different levels. It should be noted that in-creased cholesterol synthesis might not be the onlymechanism underlying hypercholesterolemia in nephro-sis. Defective LDL clearance via the receptor pathwayas well as altered renal metabolism of mevalonate havealso been suggested as possible mechanisms.18'62 Ourfinding that the hepatic LDL receptor activity was notinduced by 3-thiadicarboxylic acid treatment (Fig 2)indicates that the cholesterol-lowering effect of thisdrug in nephrosis is probably not mediated by an effecton the LDL receptor. The hepatic mRNA levels for theLDL receptor and HMG-CoA reductase seem to be

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

AL-Shurbaji et al 3-Thiadicarboxylic Acid in Nephrotic Hyperlipidemia 1585

TABLE 6. Effect of 3-Thiadicarboxylic Acid on the Rate-Limiting Enzymes inCholesterol Metabolism

Enzymes

HMG-CoA reductase,pmol/min per milligram protein

Cholesterol 7a-hydroxylase, nmol/minper milligram protein

ACAT, pmol/min per milligram protein

NontreatedRats

2122+1178

42±10

319±36

3-ThiadicarboxylicAcid-Treated Rats

880±416*

27 ±7*

231 ±41*

HMG-CoA indicates 3-hydroxy-3-methylglutaryl-coenzyme A and ACAT, acyl-CoA:cholesterolacyltransferase. Values represent mean±SD for six rats in each group.

*P<.05.

regulated in a coordinate but curvilinear pattern, imply-ing that changes in receptor expression could be elicitedonly when synthesis of the reductase is severely affect-ed.63 Therefore, a decrease of the reductase activity onthe order of 50%, as in the present study, does notnecessarily result in detectable changes in the hepaticLDL receptor expression.

The clearance defect present in nephrotic hyperlip-idemia is ascribed to several mechanisms.46 Althoughpossible effects of 3-thiadicarboxylic acid treatment onVLDL clearance were not specifically addressed in thepresent study, several observations indicated that thehypolipidemic properties of 3-thiadicarboxylic acid areprobably not mediated by any major effect on VLDLcatabolism. First, no effect on LDL receptor activity wasseen, as discussed above. Second, treatment did notaffect heparin-releasable lipoprotein lipase or hepaticlipase activities (Table 5). The lack of a substantialeffect of 3-thiadicarboxylic acid on VLDL triglycerideclearance might explain why plasma lipid levels were notnormalized despite the decrease in VLDL secretion.However, it cannot be ruled out that 3-thiadicarboxylicacid treatment might to some extent improve VLDLcatabolism. Since plasma VLDL triglyceride levels weredecreased by 3-thiadicarboxylic acid treatment, lipopro-tein lipase would be less saturated, and the efficiency oflipolysis and VLDL removal would be enhanced.

In conclusion, we have for the first time demonstratedthat treatment with the sulfur-substituted non-/3-oxidiz-able fatty acid analogue 3-thiadicarboxylic acid reducesboth plasma triglycerides and cholesterol in a hyperlip-idemic state in the rat. The triglyceride-lowering effectin the current nephrotic model was mainly achieved bystimulating fatty acid oxidation, thereby reducing he-patic triglyceride synthesis and output. The hypocholes-terolemic properties were, at least in part, mediated byan inhibition of the regulatory enzyme in cholesterolbiosynthesis. In other words, similar mechanisms seemto be operative in both normolipidemic and nephrotichyperlipidemic rats. The therapeutic potential of theseso-far-experimental lipid-lowering agents in hyperlipid-emia remains to be established.

Acknowledgments

This work was supported by The Swedish Medical ResearchCouncil (Projects 3141 and 10349); Nordisk Insulinfond;Swedish Hoechst, Hans and Loo Ostermans Fond; GoljesFoundation; Svenska Margarinindustrins Forening for Nar-ingsfysiologisk Forskning; The Norwegian Council on Cardio-vascular Disease (NCC); The Norwegian Research Council forScience and Humanities (NAVF); Norges Diabetesforbunds

Forskningsfond; Familien Blix Forskningsfond; ODD FellowVitenskapelige Forskningsfond; and The Norwegian CancerSociety (DNKF). The authors are grateful to Ms AnitaLofgren, Ms Ulla Andersson, Ms Gunvor Nilsson, Mr SveinKriiger, and Ms Anne R. Alvestad for excellent technicalassistance. The manuscript preparation by Ms Vivianne Hannaand Ms Ingrid Tornblom is gratefully acknowledged.

References

l. Baxter JH. Hyperlipoproteinemia in nephrosis. Arch Intern Med.1962;109:742-747.

2. Chopra JS, Mallick NP, Stone MC. Hyperlipoproteinemia innephrotic syndrome. Lancet. 1971;1:317-321.

3. Porro GB, Branchessi M. Ischemic heart disease complicatingnephrotic syndrome. Lancet. 1969;2:804.

4. Berlyne GM, Mallick NP. Ischemic heart disease as a complicationof nephrotic syndrome. Lancet. 1969;2:399-400.

5. Alexander JH, Scharpel GJ, Edwards KDG. Increased incidenceof coronary heart disease associated with combined elevation oftriglyceride and cholesterol concentrations in nephrotic syndromein man. Med J Aust. 1974;2:119-122.

6. Kasiske BL, O'Donnell MP, Keane WF. Lipids and progressiverenal injury. Semin Nephrol. 1989;9:354-369.

7. Rabelink AJ, Hene RJ, Erkelens DW, Joles JA, Koomans HA.Effects of simvastatin and cholestyramine on lipoprotein profile inhyperlipidemia of nephrotic syndrome. Lancet. 1988;1:1335-1338.

8. Hirano T, Komuro F, Furukawa S, Nagano S, Takahashi T. Effect ofpravastatin sodium, a new inhibitor of 3-hydroxy-3-methylglutarylcoenzyme A reductase, on very-low-density lipoprotein compositionand kinetics in hyperlipidemia associated with experimentalnephrosis. Metabolism. 1990;39:605-609.

9. Vega GL, Grundy SM. Lovastatin therapy in nephrotic hyperlip-idemia: effects on lipoprotein metabolism. Kidney Int. 1988;33:1160-1168.

10. Aarsland A, Aarsaether N, Bremer J, Berge RK. Alkylthioaceticacids (3-thia fatty acids) as non-/J-oxidizable fatty acid analogues: anew group of hypolipidemic drugs, III: dissociation of cholesterol- andtriglyceride-lowering effects and the induction of peroxisomal^oxidation. J Lipid Res. 1989;30:1711-1718.Skorve J, Asiedu D, Rustan AC, Drevon CA, AL-Shurbaji A,Berge RK. Regulation of fatty acid oxidation and triglyceride andphospholipid metabolism by hypolipidemic sulfur-substituted fattyacid analogues. J Lipid Res. 1990;31:1627-1635.

12. Skorve J, AL-Shurbaji A, Asiedu D, Bjorkhem I, Berglund L,Berge RK. On the mechanism of the hypolipidemic effect of sulfur-substituted hexadecanedioic acid (3-thiadicarboxylic acid). J LipidRes. 1993;34:1177-1185.

13. Marsh JB. Lipoprotein metabolism in experimental nephrosis.J Lipid Res. 1984;25:1619-1623.

14. Shafrir E, Brenner T. Lipoprotein lipid and protein synthesis inexperimental nephrosis and plasmapheresis, I: studies in rat invivo. Lipids. 1979;14:695-702.

15. Brenner T, Shafrir E. Lipoprotein lipid and protein synthesis inexperimental nephrosis and plasmapheresis, II: perfused rat liver.Lipids. 1980;15:637-643.

16. Kaysen GA. Hyperlipidemia of the nephrotic syndrome. KidneyInt. 1991;39(suppl 31):S-8-S-15.

17. McKenzie IF, Nestel PJ. Studies on the turnover of triglycerideand esterified cholesterol in subjects with the nephrotic syndrome./ Clin Invest. 1968;47:1685-1695.

11

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

1586 Arteriosclerosis and Thrombosis Vol 13, No 11 November 1993

18. Warwick GL, Caslake MJ, Boulton-Jones JM, Dagen M, PackardCJ, Shepherd J Low density lipoprotein metabolism in thenephrotic syndrome. Metabolism. 1990;39:178-192.

19. Berge RK, Aarsland A, Kryvi H, Bremer J, Aarsaether N. Alkyl-thioactic acids (3-thia fatty acids): a new group of non-/3-oxidizableperoxisome-inducing fatty acid analogues, I: a study on thestructural requirements for proliferation of peroxisomes and mito-chondria in rat liver. Biochun Biophys Acta. 1989; 1004:345-356.

20. AL-Shurbaji A, Larsson-Backstrom C, Berglund L, Eggertsen G,Bjorkhem I. Effect of n-3 fatty acids on the key enzymes involvedin cholesterol and triglyceride turnover in rat liver. Lipids. 1991;26:385-389.

21. Bates EJ, Saggerson DA. A selective decrease in mitochondrialglycerol phosphate acyltransferase activity in livers fromstreptozotocin-diabetic rats. FEBS Lett. 1977;84:229-232.

22. Daae LNW, Bremer J. The acylation of glycerol phosphate in ratliver: a new assay procedure for glycerophosphate acylation in ratliver, studies on its subcellular and submitochondrial localizationand determination of the reaction products. Biochim Biophys Acta.1970;210:92-104.

23. Bell RM, Coleman RM. Enzymes of glycerolipid synthesis ineucaryotes. Annu Rev Biochem. 1980;49:459-487.

24. Brown MS, Goldstein JL, Dietschy JM. Active and inactive formsof 3-hydroxy-3-methylglutaryl coenzyme A reductase in the liverof the rat: comparison with the rate of cholesterol synthesis indifferent physiological states. J Biol Chem. 1979;254:5144-5149.

25. Bjorkhem 1, Kallner A. Hepatic 7a-hydroxylation of cholesterol inascorbate-deficient and ascorbate-supplemented guinea pigs.J Lipid Res. 1976;17:360-365.

26. Field FJ, Albright E, Methur S. Inhibition of acyl coenzyme A:cho-lesterol acyltransferase activity by PD 128042: effect on cholesterolmetabolism and secretion in CaCo-2 cells. Lipids 1991;26:l-8.

27. Rose-Kahn G, Bar-Tana J. Inhibition of lipid synthesis by B,B'-tetramethyl-substituted Cl4-C22, a,w-dicarboxylic acids in culturedrat hepatocytes. J Biol Chem. ~1989;260:8411-8415.

28. Tanabe T, Nakanishi S, Hashimoto T, Nikowa JI, Numa S. AcetylCoA carboxylase from rat liver. Methods Enzymol. 1981;71:5-16.

29. Roncari DA. Fatty acid synthetase from human liver. MethodsEnzymol. 1981:71:73-79.

30. Berglund L, Bjorkhem 1, Einarsson K. Apparent phosphorylation-dephosphorylation of soluble phosphatidic acid phosphatase in ratliver. Biochem Biophys Res Commun. 1982;105:288-295.

31. Humble E, Berglund L. Stimulation and inhibition of rat livercytosolic phosphatidate phosphohydrolase by various phospho-lipids. J Lipid Res. 1991;32:1869-1872.

32. Berge RK, Nilsson A, Hus0y AM. Rapid stimulation of liverpalmitoyl-CoA synthetase, carnitine palmitoyltransferase and glycero-phosphate acyltransferase compared to peroxisomal /3-oxidation andpalmitoyl-CoA hydrolase in rats fed high-fat diets. Biochim BiophysActa. 1988;960:417-426.

33. Berge RK, Flatmark T, Christiansen EN. Effect of a high-fat dietwith partially hydrogenated fish oil on long-chain fatty acid metab-olizing enzymes in subcellular fractions of rat liver. Arch BiochemBiophys. 1987:252:269-276.

34. Asiedu D, Aarsland A, Skorve J, Svardal AM, Berge RK. Fattyacid metabolism in liver of rats treated with hypolipidemic sulphur-substituted fatty acid analogues. Biochim Biophys Acta. 1990; 1044:211-221.

35. Nilsson-Ehle P, Schotz MC. A stable, radioactive substrateemulsion for assay of lipoprotein hpase. J Lipid Res. 1976:17:536-541.

36. Shimizu S, Yasui K, Tani Y, Yamada H. Acyl-CoA oxidase fromCandida tropicalis. Biochem Biophys Res Commun 1979;91:108-113.

37. Mulder C, Schouten JA, Popp-Snijders C. Determination of freefatty acids: a comparative study of the enzymatic versus the gaschromatographic and colonmetric method. J Clin Chem Biochem.1983;21:823-827.

38. Otway S, Robinson DS. A non-ionic detergent (Triton WR 1339)to determine rates of triglyceride entry into the circulation of therat under different physiological conditions. J Physiol. 1967;190:321-323.

39. AL-Shurbaji A, Berglund L, Bjorkhem I. The effect of Acipimoxon triacylglycerol metabolism in rat. Scand J Clin Lab Invest. 1990;50:203-208.

40. Carlson K. Lipoprotein fractionation. J Clin Pathol. 1973;26:32-37.41. Lopez-Virella MF, Stone P, Ellis S, Colwell JA. Cholesterol deter-

mination in high-density lipoproteins separated by three differentmethods. Clin Chem. 1977:23:882-884.

42 Lowry OH, Rosebrough MJ, Farr AL, Randall RJ. Protein mea-surement with the Folin phenol reagent. J Biol Chem. 1951; 193.265-275.

43. Bartlett G. Phosphorus assay in column chromatography. J BiolChem. 1958;234:466-468.

44. Rudling M, Norstedt G, Olivecrona H, Reihner E, Gustafsson J-A,Angelin B. Importance of growth hormone for the induction ofhepatic low density lipoprotein receptors. Proc Natl Acad SciUSA. 1992;89:6983-6987.

45. Furukawa S, Hirano T, Mamo JCL. Nagano S, Takahashi T. Cat-abolic defect of triglyceride is associated with abnormal very-low-density lipoprotein in experimental nephrosis. Metabolism 1990;39:101-107.

46. Garber DW, Gottlieb BA, Marsh JB, Sparks CE. Catabolism ofvery low density lipoproteins in experimental nephrosis. J ClmInvest. 1984;74:1375-1383.

47. Marsh JB, Sparks CE. Hepatic secretion of lipoproteins in the ratand the effect of experimental nephrosis. J Clin Invest. 1979;64:1229-1237.

48. Calandra S, Gherardi F, Fainaru M, Guaitani A, Bartosek I.Secretion of lipoproteins, apolipoprotein A-I and apolipoprotein Eby isolated and perfused liver of rat with experimental nephroticsyndrome. Biochim Biophys Acta. 1981;665:331-338.

49. Joven J, Villabona C, Vilella E, Masana L, Alberti R, Valies M.Abnormalities of lipoprotein metabolism in patients with thenephrotic syndrome. N EnglJ Med. 1990;323:579-584.

50. Brindley DN. Phosphatidate phosphohydrolase activity in the liver.In: Brindley DN, ed Phosphatidate Phosphohydrolase. Boca Raton,Fla: CRC Press; 1988;l:l-77. CRC Series in Enzyme Biology.

51. Savolainen MJ, Lehtonen MA, Ruokonen A, Hassinen IE.Postnatal development and sex difference in hepatic phosphatidatephosphohydrolase activity in the rat. Metabolism. 1981;30:706-711

52. Jamdar SC, Moon M, Bow S, Fallon HJ. Hepatic lipid metabolism:age-related changes in triglyceride metabolism. J Lipid Res. 1978;19:763-770.

53. Halminski MA, Marsh JB, Harrison EH. Differential effects offishoil, safflower and palm oil on fatty acid oxidation and glycerolipidsynthesis in rat liver. J Nutr. 1991; 12:1554-1561.

54. Goldberg DM, Roomi MW, Yu A, Roncari DAK. Triacylglycerolmetabolism in the phenobarbital treated rat. Biochem J. 1981;196:337-346.

55. Strum-Odin R, Adkins-Findke B, Blake DL, Phinney SD, ClarkeSD. Modification of fatty acid composition of membrane phos-pholipid in hepatocyte monolayer with n-3, n-6 and n-9 fatty acidsand its relationship to triacylglycerol production. Biochim BiophysActa. 1987;921:378-391.

56. Mayorek N, Bar-Tana J. Inhibition of diacylglycerol acyltrans-ferase by 2-bromooctanoate in cultured rat hepatocytes. J BiolChem 1985;260:6528-6532.

57. Mayorek N, Grinstein I, Bar-Tana J. Triacylglycerol synthesis incultured rat hepatocytes: the rate-limiting role of diacylglycerolacyltransferase. EurJ Biochem. 1987;182:395-400.

58. Stals HK, Mannaerts GP, Declercq PE. Factors influencing tri-acylglycerol synthesis in permeabilized rat hepatocytes. Biochem J.1992;283-719-725.

59. Golper TA, Feingold KR, Fulford MH. Siperstein M-D. The roleof circulating mevalonate in nephrotic hypercholesterolemia in therat. J Lipid Res. 1986;27:1044-1051.

60 Warwick GL, Packard CJ, Murray L, Grierson D, Stewart JP,Shepherd J, Boulton-Jones M. Effect of simvastatin on plasma lipidand lipoprotein concentrations and low-density lipoproteinmetabolism in the nephrotic syndrome. Clin Sci. 1992:82:701-708.

61. Grundy SM. Management of hyperlipidemia in kidney disease.Kidney Int. 1990;37:847-853.

62. Golper TA, Swartz SH. Impaired renal mevalonate metabolism innephrotic syndrome: a stimulus for increased hepatic cholestero-genesis independent of GFR and hypoalbuminemia. Metabolism.1982;31:471-476.

63. Rudling M. Hepatic mRNA levels for the LDL receptor andHMG-CoA reductase show coordinate regulation in vivo. J LipidRes. 1992;33:493-501.

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from

A al-Shurbaji, J Skorve, R K Berge, M Rudling, I Björkhem and L BerglundEffect of 3-thiadicarboxylic acid on lipid metabolism in experimental nephrosis.

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 1993 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleArteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/01.ATV.13.11.15801993;13:1580-1586Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/13/11/1580World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

document.Permissions and Rights Question and AnswerFurther information about this process is available in theis being requested is located, click Request Permissions in the middle column of the Web page under Services.Clearance Center, not the Editorial Office. Once the online version of the published article for which permission

can be obtained via RightsLink, a service of the CopyrightArteriosclerosis, Thrombosis, and Vascular Biology Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on June 19, 2018http://atvb.ahajournals.org/

Dow

nloaded from