Directory UMM :Data Elmu:jurnal:A:Atherosclerosis:Vol152.Issue1.Sep2000:

(1)

Effects of dietary fat amount and saturation on the regulation of

hepatic mRNA and plasma apolipoprotein A-I in rats

Lucı´a Calleja

_{, Marı´a Carmen Trallero}

_{, Carmen Carrizosa}

_,

Marı´a Teresa Me´ndez

_{, Evangelina Palacios-Alaiz}

_{, Jesu´s Osada}

_*

a_{Departamento de Bioquı´mica y Biologı´a Molecular y Celular,}_{Facultad de Veterinaria,}_Uni₆_{ersidad de Zaragoza,}_{Miguel Ser}₆_et,₁₇₇_,

E-50013 Zaragoza, Spain

b_{Instituto de Bioquı´mica,}₍_{Centro mixto CSIC-UCM}₎_,_{Facultad de Farmacia,}_{Ciudad Uni}₆_ersitaria,_Madrid, _Spain

Received 22 December 1998; received in revised form 27 September 1999; accepted 3 November 1999

Abstract

The effects of the amount of dietary fat and saturation together with cholesterol both on hepatic apolipoprotein A-I gene mRNA levels and on plasma levels of this apolipoprotein were studied in male rats. To achieve these goals, seven groups of male Wistar rats were established: control group (n=5) consuming chow diet; cholesterol group (n=4) fed on a chow diet containing 0.1% (w/w) cholesterol; coco group (n=5) fed on a chow diet containing 0.1% (w/w) cholesterol and 40% coconut oil; corn group (n=5) fed on a chow diet containing 0.1% (w/w) cholesterol and 40% corn oil; and three olive groups consuming a chow diet containing 0.1% (w/w) cholesterol and percentages of 5 (n=5), 10 (n=4) and 40% (n=5), respectively, of olive oil. Animals were kept on these diets for 2 months and then sacrificed for lipoprotein, apolipoprotein and hepatic mRNA analysis. Dietary cholesterol by itself was hypercholesterolemic when compared to chow diet, an effect that was mainly due to an increase in LDL-cholesterol. Corn oil had a hypocholesterolemic action, whether compared to chow or to cholesterol diet, due to a reduction in HDL-cholesterol as well as LDL-cholesterol. HDL-cholesterol levels of 40% olive oil diet were lower than those corresponding to coconut oil and higher than those found in corn oil diet. When compared to control or cholesterol diets, plasma apoA-I concentration appeared significantly increased in coconut and 40% olive oil diets. Coconut oil or corn oil diets did not induce any significant change in apoA-I mRNA compared to control or cholesterol diets. Compared to cholesterol diet, 40 and 10% olive oil diets induced a significant increase in the expression of this message. A positive and significant (r=0.97, PB0.01) correlation between plasma apolipoprotein A-I concentration and its hepatic mRNA, was observed when the amount of dietary olive oil was 40% (w/w). A significant negative (r= −0.97,PB0.01) correlation was found in the corn oil group and no significant association was observed in the remaining groups. Based on the increased plasma levels in coconut oil and in high percentage olive oil diets, and the differences between these two diets for mRNA expression, it can be concluded that different fatty acid containing diets regulate apolipoprotein A-I through different mechanisms, and these mechanisms could be modulated by the fat intake. © 2000 Elsevier Science Ireland Ltd. All rights reserved.

Keywords:Apolipoprotein A-I; Dietary fat; Lipoproteins; Monounsaturated fatty acids; Saturated fatty acids; Polyunsaturated fatty acids www.elsevier.com/locate/atherosclerosis

1. Introduction

Epidemiological studies have demonstrated that high-density lipoprotein (HDL) cholesterol is a negative risk factor for coronary artery disease (CAD) indepen-dent of other prognostic factors [1 – 4]. Clinical trials

have also proved that increases in HDL-cholesterol are associated with decreases in CAD events, and with atherosclerotic regression [5]. While low HDL-choles-terol levels may be the result of genetic disorders, other environmental factors, such as physical inactivity, cigarette smoking, dietary composition, hormonal status, body mass index and alcohol consumption, are the primary responsible factors for the majority of changes observed in the general population. Pharmaco-logical agents such as niacin, fibrates and statins are also involved in HDL control [6].

* Corresponding author. Tel.: +34-976-761-644; fax: + 34-976-761-612.

E-mail address:[email protected] (J. Osada).

(2)

Dietary composition as an important environmental component influencing HDL-cholesterol levels is rather complex. Thus carbohydrate-rich diets are associated with lower HDL levels [6]. On the other hand, the type of supplied fat also modulates the latter parameter, in the sense that saturated, as compared to polyunsatu-rated, fat consumption induces an increase in HDL-cholesterol levels in humans and in other animal models [7,8]. The effect of monounsaturated diets is still con-troversial. While classical studies described a neutral role of monounsaturated fatty acids in determining plasma lipids [9], recent studies have shown the effect of monounsaturated fatty acids (oleic acid) in lowering serum levels of low-density lipoprotein cholesterol with-out modification of HDL-cholesterol levels when sub-stituted for saturated fatty acids in human and non-human primates [10 – 14].

Apolipoprotein A-I (apoA-I) is the main protein component of high density lipoproteins, and is involved in their structure and metabolism. Therefore, the syn-thesis and secretion of apoA-I as well as its catabolism may be mechanisms regulating plasma HDL levels (re-viewed in Refs. [15] and [16]). The mechanisms whereby different fatty acid and (or) cholesterol regulate plasma apoA-I protein are the focus of current research in the field. Thus, studies carried out in primates have shown that, in the presence of cholesterol, dietary saturated and polyunsaturated fatty acids modulate apoA-I pro-duction by transcriptional regulation involving mainly liver apoA-I mRNA, and that the catabolic rate of HDL apoA-I is also involved [17 – 20]. In rodents, the increase in apoA-I by saturated fat is produced by posttranscriptional mechanisms, not shared by dietary cholesterol [21]. Once again, few studies have addressed the influence of monounsaturated fatty acid containing diets on apoA-I regulation. In non-human primates, Fox et al. [22] described an increase in hepatic apoA-I mRNA with diets enriched in the latter type of fatty acids. Likewise, we have previously shown that rats fed on a diet containing 40% (w/w) monounsaturated fat and 0.1% cholesterol showed induced hepatic expres-sion of apoA-I mRNA [23]. The aim of the present study was to investigate whether the latter effect is influenced by the amount of fat provided and the contribution of hepatic mRNA expression on plasma apolipoprotein concentration.

2. Material and methods 2.1. Materials

Standard rat chow was provided by Pascual (Barcelona, Spain). Cholesterol was obtained from Sigma (Madrid, Spain). Starch was provided by Co-fares (Madrid). Olive oil was obtained from Mora

(Alcolea), coconut oil was obtained from Cofares and corn oil from Koipe (Andujar). Molecular biology reagents were purchased from Boehringer Mannheim (Barcelona, Spain) and Sigma. Restriction enzymes were purchased from Pharmacia and Promega (Barcelona, Spain). [a-32P]dCTP was obtained from

New England Nuclear (Madrid, Spain). 2.2. Animals and diets

Inbred male Wistar rats (200 – 250 g, aged 2 months) were kept under different experimental conditions. A detailed description of diets is shown in Table 1. The groups were as follows. Group 1: control animals (five rats) were fed on a standard commercial diet. Group 2: cholesterol group (four rats), commercial diet was sup-plemented by 0.1% (w/w) cholesterol. Groups 3 – 5 (five rats): commercial diet was supplemented by 0.1% (w/w) cholesterol and 40% (w/w) fat. All diets contained 1% starch as agglutinant. Three different types of fat were used: coconut oil, corn oil and olive oil. A wide range of polyunsaturated to saturated fatty acid ratio (P/S) was covered (0.03 – 3.8) and a wide range of monoun-saturated fatty acids was also utilized (8.2 – 74.6%). Groups 6 and 7: commercial diet was supplemented by 0.1% (w/w) cholesterol and 5% (five rats) or 10% (w/w) (five rats) olive oil, respectively. Total fat from diets was extracted (Soxhlet) and fatty acids analyzed using gas-liquid chromatography [24]. All groups were fed ad libitum for 2 months. Before killing, rats were kept on a fasting diet with free access to water for 18 h. Animals used in this study were handled and killed, strictly observing criteria from the European Union for care and use of animal laboratory in research.

2.3. Plasma chemistry

After the experimental period, animals were sac-rificed by Avertin (2,2,2-tribromoethanol) injection (Aldrich, Madrid, Spain). The abdominal cavity was opened and intestinal mass removed to visualize the abdominal aorta. Blood samples drawn from the ab-dominal aorta were collected in 0.1% EDTA tubes. Plasma was separated at 2500 rpm for 20 min at 4°C. Phenylmethyl sulfonylfluoride (PMSF, 2×10−4 _M), andN-ethylmaleimide (NEM, 1.25 mg/ml, Sigma) were added to plasma as proteolytic and lecithin:cholesterol acyltransferase inhibitors. Cholesterol and triglycerides were measured by enzymatic methods [25,26]. Assays were standardized by using Cardiolipid (Sigma) as cer-tified quality control.

2.4. Antibody generation and apolipoprotein A-I assay Purified rat apolipoprotein A-I was obtained by elec-trophoresis of isolated HDL in a 4-22.5%

(3)

polyacry-lamide gel. Gel fragment corresponding to this protein was electroeluted in 0.2 M Tris/acetate (pH 7.4), 1% SDS and 100 mM dithiotreitol. Purified protein concen-tration was assayed using Biorad Protein Assay. A total of 4mg of protein diluted in 200ml of saline phosphate

buffer (PBS) (pH 7.4) was mixed with 300 ml of

com-plete Freund adjuvant. This mixture was subcuta-neously injected into rabbits. A second boost was given a month later and animals bled after another 20 days. Purified rabbit anti-rat apolipoprotein A-I inmunoglob-ulins G were used to estimate apolipoprotein concentra-tion in a chemiluminescent ELISPOT assay normalized with pure rat apolipoprotein A-I. This assay showed a 5% of interassay variation coefficient.

2.5. RNA preparation

A liver biopsy was flash frozen in liquid nitrogen and stored at −70°C until use. Liver fragments were ho-mogenized in guanidinium isothiocyanate and applied to a cushion of cesium chloride as described by Kingston [27].

2.6. Preparation of DNA probes

The following probes were used: a 0.7-kb EcoRI fragment of rat apoA-I cDNA [28] (generously pro-vided by Dr Lawrence Chan, Baylor College, Houston)

and, as an internal standard, an insert of mouse GPDH (Ambion, Madrid, Spain). DNA inserts were purified by gel electrophoresis. Purified DNA probes were la-beled with [a-32P]dCTP by a random primer protocol

[29] using Rediprime from Amersham-Pharmacia (Madrid, Spain).

2.7. Analysis of mRNA 2.7.1. Northern analysis

Denatured RNA (5 mg) was run in a formaldehyde

agarose gel, blotted into nylon membranes, hybridized to a radio labeled probe, washed and exposed to a film following current procedures [30].

2.7.2. Slot-blot analysis

Several amounts of denatured total RNA (1 – 10 mg)

were applied to nylon filters using a slot-blot apparatus (Biorad, Madrid, Spain). Films were scanned by laser densitometry using an LKB 2202 densitometer and an LKB 2400 gel scan XL software. mRNA abundance was measured in absorbance units and normalized to pg of apoA-I mRNA/mg of total RNA using an in vitro

transcript of apoA-I generated by using Riboprime (Promega). Blots were stripped and reprobed with GPDH, exposed to film and scanned as described. The relative abundance of apoA-I mRNA message was normalized to values of GPDH as an internal standard.

Table 1

Composition of the different experimental dietsa

Cholesterol Cholesterol

Control

+coconut +corn oil +40% olive +10% olive +5% olive

Energy supply 10.7 10.7 21.4 21.4 21.4 13.4 12

(kJ/g)

39 37

Carbohydrate 41 41 25 25 25

Protein 19 19 11 11 11 17 18

42 13 8

Fat 3.2 3.2 42 42

0.1 0.1 0.1

0.1

Cholesterol 0.02 0.1 0.1

Main fatty acids

BC12:0 13.3

0.1 0.1 46

Lauric (12:0) Trace Trace Trace Trace

0.4 0.2

0.1 Trace

Myristic (14:0) 1.5 1.5 17

21.6 21.6 9.6

Palmitic (16:0) 11.2 12 13.3 14.7

2.8 3.3 3.9

Stearic (18:0) 7.0 7.0 2.6 2.7

33 33 8.2

Oleic (18:1) 33.6 74.6 70 65.3

Linoleic 30 30 2.8 52.1 9.5 12.2 14.9

(18:2n-6)

Trace 0.3

0.3 0.8

Linolenic 0.9 1 0.9

(18:3n-3)

18.7 16.9

Total saturated 33.8 33.8 88.5 13.9 14.9

35.2

Monounsaturated 35.2 8.2 33.6 74.6 70 65.3

0.9 0.9 3.8 0.7 0.8

P/S ratio 0.03 0.83

a_{Components are expressed as g% (w}_/_{w). Other components of chow diet are crude fiber, 5%, and minerals, 7%. A total dry matter of 75%.}

(4)

Table 2

Nutritional evaluation of experimental groupsa

Weight gain (g/day)

Experimental group Food intake (g/day per rat) Energy supply/weight gain (kJ/g per rat)

Control (n=5) 3595 2.590.4 148921

2.790.4 125919b

Cholesterol (n=4) 3295 2193bc

Cholesterol+40% coconut 2.090.4 216928bc

(n=5)

Cholesterol+40% corn (n=5) ₂₁₉₃bc 1.790.5bc ₂₅₆₉₃₅bcd

1.790.2bc

2093bc ₂₄₃₉₄₈bcd

Cholesterol+40% olive (n=5)

2.690.2e

Cholesterol+10% olive (n=4) ₂₆₉₄bce 137918fe

2.690.4e

3194bef ₁₄₁₉₁₉e

Cholesterol+5% olive (n=5)

PB0.0001

ANOVA PB0.0035 PB0.0001

a_{Data are shown as means and their standard deviations. Statistical analysis was done using one-way ANOVA and Tukey–Kramer Multiple}

Comparisons as post hoc test. Different superscripts (b_{versus control;}c_{versus cholesterol;}d_{versus cholesterol}₊_{coconut oil;}e_{versus cholesterol}₊

40% olive; andf_{versus cholesterol}₊_{10% olive) are significantly different from each other at}_P_B_0.001.

2.8. Statistical analysis

Results are shown as means and their standard devi-ations. Analysis of data was done using Instat 2.0 for Macintosh software (GraphPad, San Diego, CA, USA). Some of the analyzed parameters in this study did not show normal distribution according to Shapiro – Wilk test, or failed in homology of variance. Therefore, analysis of statistically significant differences was car-ried out using Mann – Whitney U-test for unpaired data, and one-way ANOVA was performed according to Kruskal – Wallis test. Alternatively, one-way ANOVA, according to Tukey, was carried out and Tukey – Kramer multiple comparison test as post hoc analysis was used. Differences were considered non-sig-nificant when P\0.05. Association between variables was assessed by Spearman’s rank-order correlation co-efficient (rs) [31].

3. Results

3.1. Dietary characteristics

Table 1 summarizes the composition of different diets. As shown, distinctive general features were lower carbohydrate and protein, higher percentage of fat and an increase in cholesterol in the different diets, chow being the control. Diets containing different amounts of fat (5, 10 and 40%) were not isocaloric as were diets providing different types of fat at 40% (w/w) content. Chow and cholesterol diets presented a monounsatu-rated fatty acid content of 33% and almost equal amounts of polyunsaturated and saturated fatty acids (P/S ratio 0.9). Coconut oil diet provided a high con-tent of saturated short chain fatty acids (88.5) and a low content of monounsaturated fatty acids. Corn oil diet was particularly enriched in linoleic acid and levels of monounsaturation were similar to the control diet. Olive oil diets were high in monounsaturated fatty

acids, differing in their percentages in inverse propor-tion to the fat content.

3.2. Nutritional e6aluation

Table 2 shows that increasing food fat content to percentages higher than 5% produced significant de-creases (PB0.001) in the food intake, and these de-creases were the highest with 40% (w/w) rates without any difference among type of oil. However, weight gain was decreased only in those groups of experimental animals consuming 40% fat containing diets with again no differences among oils. As results of these data, the ratio of energy supplied related to weight gain was significantly (PB0.001) higher in animals consuming diets enriched in 40% fat, independent of its source. 3.3. Plasma analysis

Table 3 shows the total plasma cholesterol for each of the seven groups at the end of the experimental period. Dietary cholesterol was a hypercholesterolemic agent in these animals since plasma cholesterol signifi-cantly (PB0.05 vs. control) increased when rats were fed diets enriched in this compound. When compared to the cholesterol group, simultaneous administration of saturated fat (coconut diet) and cholesterol did not increase plasma cholesterol. Corn oil administration not only prevented hypercholesterolemia caused by cholesterol administration (PB0.001 vs. cholesterol), but reduced plasma cholesterol to levels lower than control group (PB0.05 vs. control). Olive oil adminis-tration prevented the hypercholesterolemic effect of cholesterol (PB0.05 vs. cholesterol) maintaining simi-lar values to those found in animals fed on a high carbohydrate diet (control group), irrespective of the dose employed.

The hypercholesterolemic effect of dietary cholesterol (Table 3) is mainly due to an increase in LDL-choles-terol (PB0.001 vs. control), and the

(5)

hypocholes-Table 3

Plasma lipid parametersa

HDL-cholesterol LDL-cholesterol

Experimental condition Total cholesterol Triglycerides

4493

Control (n=5) 6793 1994 41910

Cholesterol (n=4) 8597b ₄₅₉₄ ₃₃₉₃b ₆₂₉₁₀b

5094b

Cholesterol+40% coconut (n=5) ₇₅₉₅ 2393c ₆₉₉ ₁₄b

2994bcd ₁₅₉₃cd

4897bcd ₄₄₉₉cd

Cholesterol+40% corn (n=5)

67912ce

Cholesterol+40% olive (n=5) 3895d ₂₈₉₆bde ₄₇₉₁₉

4493 2493c

Cholesterol+10% olive (n=4) 7394 5097

4097 2494c

69912 4399c

Cholesterol+5% olive (n=5)

PB0.003

ANOVA PB0.002 PB0.0015 PB0.02

a_{Data, expressed as mg}_/_{dl, are means and their standard deviations. Statistical analysis was done using one way ANOVA according to Tukey}

and Tukey–Kramer Multiple Comparison Test as post hoc analysis. Different superscripts (b _{versus control;} c _{versus cholesterol;} d _versus

cholesterol+coconut oil; ande_{versus cholesterol}₊_{corn oil), are significantly different from each other at}_P_B_0.05.

terolemic effect of corn oil is due to significant (PB 0.001 vs. control) reductions in HDL-cholesterol as well as LDL-cholesterol. LDL-cholesterol levels were lower in the coconut oil group compared to the cholesterol group (PB0.05 vs. cholesterol) which explains the de-crease in total cholesterol observed for this group. The hypocholesterolemic effect of olive oil is mainly due to a decrease (PB0.05 vs. cholesterol) in LDL-cholesterol and is more pronounced in 5 and 10% olive oil contain-ing diets. LDL-cholesterol in saturated and monounsat-urated diets was similar and lower than in the cholesterol group. The 40% coconut oil diet was better at reducing LDL-cholesterol compared to 40% olive oil (PB0.05), and as effective as other doses of olive oil. HDL-cholesterol levels of monounsaturated fatty acid diets were found among those of saturated and polyun-saturated diets, reaching the maximum value in animals consuming 10% olive oil diets. The HDL:LDL ratio was actually more favorable regarding the chow diet (2.3) than the olive oil diets (mean of 1.6). Also, the ratio of 2.2 for the cholesterol+coconut oil diet was relatively favorable when compared with the olive oil diets.

Plasma triglycerides (Table 3) increased (PB0.05 vs. control) when rats were fed on cholesterol. This effect was not enhanced by simultaneous administration of saturated fat. Rats receiving polyunsaturated (corn oil) and 5% monounsaturated diets showed a decrease in plasma triglycerides when compared to the cholesterol group (PB0.05 vs. cholesterol).

Fig. 1A shows that apoA-I used to generate the polyclonal antibody was homogeneously purified and panel B shows proof that purified rabbit Ig G anti-apoA-I recognized only this apolipoprotein. Plasma apoA-I concentration (Table 4) appeared significantly (PB0.05 vs. control and vs. cholesterol) increased in saturated and monounsaturated fatty acid provided at 40%. A 10% monounsaturated diet induced levels of apoA-I that were significantly (PB0.05) different from the cholesterol and 40% the olive oil group.

Fig. 1. Quality characteristics of apolipoprotein A-I, antibody, hybri-dation conditions and probes used in the present study. (A) Denatur-ing polyacrylamide electrophoresis showDenatur-ing homogeneously purified apolipoprotein A-I stained with Coomassie blue. (B). Western blot-ting showing the specificity of inmunopurified rabbit Ig G anti-rat apoA-I. Molecular markers used in A and B were: bovine albumin, 66 kDa; egg albumin, 45 kDa; glyceraldehyde dehydrogenase, 36 kDa; carbonic anhydrase, 29 kDa; and a-lactoalbumin, 14 kDa. Northern blot analysis reflecting the specificity of hybridation condi-tions for apoA-I (C) and GPDH (D) probes. A total of 5mg of total RNA was subjected to electrophoresis in a formaldehyde 1% agarose gel. The RNA was transferred to a nylon membrane and hybridized to radioactive cDNAs. mRNA sizes are indicated on the autoradio-gram in relation to the ribosomal RNAs.

(6)

Table 4

Plasma concentration and hepatic expression of rat apolipoprotein A-I in the different experimental dietsa

Hepatic expression (pg of Plasma

Experimental

condition concentration apoA-I mRNA/mg total RNA)

(mg/dl)

66934

Control (n=5) 2194

Cholesterol 49935 1695 (n=4)

Cholesterol+40% 101934bc ₁₆₉₃

coconut (n=5)

71911 1693 Cholesterol+40%

corn (n=5)

102912bce

Cholesterol+40% 28911cde

olive (n=5)

8996cf

Cholesterol+10% 2796c

olive (n=4)

Cholesterol+5% 89916 19913 olive (n=5)

ANOVA PB0.04 PB0.05

a_{Data are means and their standard deviations. Hepatic}

expres-sions are normalized to the levels of GPDH expression. Statistical analysis was done using non-parametric one-way ANOVA according to Kruskal–Wallis test and unpaired Mann–Whitney U-test to test pair-wise differences. Different superscripts (b_{versus control;}c_versus

cholesterol;d _{versus cholesterol}₊_{coconut oil;}e_{versus cholesterol}₊

corn oil; andf_{versus cholesterol}₊_{40% olive) are significantly}

differ-ent from each other atPB0.05.

analyzed against coconut or corn oils because the con-ditions are not comparable due to two different parameters involved: source of fat and fat intake.

The degree of relationship among parameters in the different dietary groups was established using an associ-ation analysis, and the values of Spearman correlassoci-ation coefficient (rs) are shown in Table 5. Although a general positive correlation was found between plasma apoA-I and HDL-cholesterol, the degree of association ap-peared highly variable depending on the diet. While highly significant correlations were noted for corn and 10% olive oil groups, weaker associations were ob-served in the remaining groups.

Table 5 also shows the relationship between plasma apolipoprotein A-I concentration and hepatic apoA-I mRNA. According to the observed correlations, the increase in plasma apoA-I in olive oil containing diets is strongly related (rs=0.97,PB0.01) to changes in hep-atic apoA-I mRNA. However this dramhep-atic effect is only observed when supplied dietary fat is 40% (w/w). A significant negative correlation (rs= −0.97, PB 0.01) was found in the corn oil group and no significant association was observed in the remaining groups.

4. Discussion

In this experiment we used male Wistar rats, an animal model that we had previously found to undergo changes in plasma lipids when saturation of fat was changed in the presence of cholesterol [32]. In that study, an extremely high fat percentage 40% (w/w) and 0.1% (w/w) cholesterol were used, and a 60% increase of hepatic apoA-I mRNA was found when olive oil was the source of fat. The present work is an extension of the previous study in order to determine whether the described effect of olive oil could exist at lower fat contents with the additional presence of 0.1% choles-terol and if the observed effect at the hepatic mRNA level is reflected at all by plasma apoA-I concentration. The energy supply in animal groups consuming 40% fat was higher than in groups on chow, cholesterol, 5% olive oil or 10% olive oil as shown in Table 1. Animals 3.4. RNA analysis

As depicted in Fig. 1, the used cDNA probes (1C, apoA-I and 1D, GPDH) hybridized to a single elec-trophoretic mRNA species of the appropriate molecu-lar size in each case. Table 4 shows the values of hepatic apolipoprotein apoA-I mRNA obtained in the different groups consuming the experimental diets. When compared to control group fed chow diet, there was no significant change in this message in animals consuming coconut, corn or olive oil. However, when cholesterol fed animals were taken as the reference group, a significant increase was observed in olive oil groups when fat percentage was higher than 10% (w/w) without any significant difference between the two pro-vided doses (10 or 40%). The 10% olive oil was not Table 5

Relationship between parameters in the different experimental dietsa

Plasma apoA-I concentration

Control Cholesterol Coconut Corn 40% Olive 10% Olive 5% Olive (n=5)

(n=5) (n=5) (n=4)

(n=4) (n=5)

(n=5)

0.4 0.6

0.4

HDL-cholesterol 0.9** 0.4 0.82* 0.7

ApoA-I mRNA 0.5 0.8 −0.15 −0.97** 0.97** 0.4 0.2

a_{Spearman correlation coefficients}_r

sare indicated.n, number of animals in each experimental group.

*PB0.05. **PB0.01.

(7)

consuming 40% fat content had less food intake and less weight gain (Table 2). The latter effect has also been reported by other authors [33] and fuels the controversy that high lipid diets, at least in this animal, are not responsible for obesity [34]. When the amount of dietary energy required to gain 1 g is considered, it is clearly observed that a lower metabolic efficiency was noted in those groups consuming 40% fat since the values of that ratio are higher. This observation pre-sents an interesting aspect that needs further research to establish whether a lesser absorption is involved, or whether an induction of metabolism takes place.

Dietary cholesterol induced hypercholesterolemia in rats confirming our previous results [32]. The increase of plasma cholesterol is mainly due to an increase in LDL-cholesterol (Table 3). Our dietary cholesterol per-centage (0.1%, w/w) was not high enough to elicit changes in LDLr mRNA levels [32]. Woollett et al. [35] and Dietschy et al. [36] with similar dietary cholesterol amount have described reduction of LDLr activity. In HepG2 cells, LDL receptor activity was reported to have been reduced without any changes in LDL recep-tor mRNA [37]. In rats and mice, changes in activity were not associated with protein or mRNA changes [38]. Therefore LDL receptor activity may be more important than protein and mRNA levels to explain plasma LDL levels. The lack of hypercholesterolemic effect of lauric acid (main fatty acid of coconut oil) in this study corroborates our previous work [32] and is in agreement with results obtained by Jackson et al. [39] in hamsters and by Denke and Grundy in humans [40]. Coconut oil induced an increase in HDL-cholesterol in agreement with previous work done in other experi-mental animals and in humans [10,11,17]. In this study, we found that corn oil is very effective in preventing hypercholesterolemia caused by dietary cholesterol. This effect could be due to the high linoleic acid content of corn oil as reported by Grundy and Denke [11]. This oil decreases both LDL- and HDL-choles-terol, the latter effect being of higher intensity. De-crease in HDL-cholesterol has also been described in monkeys and it seems to be due to lesser hepatic secretion of these particles. Animals fed olive oil in the different percentages showed similar values of choles-terolemia as animals fed chow diet and a hypocholes-terolemic effect when compared to cholesterol fed animals. The latter is mainly due to a decrease in LDL-cholesterol and not to changes in HDL-choles-terol, in agreement with other studies using monounsat-urated fatty acid diets [10,14]. When provided at 40%, the coconut oil diet surprisingly decreased LDL-choles-terol at a higher rate than the same percentage of olive oil. An explanation of this fact could be that the food provided to the coconut group had a lower amount of palmitic acid than the 40% olive oil (Table 1). It is well documented that palmitic acid raises LDL-cholesterol

[40]. Furthermore, the amount of linoleic in the coconut blend could be high enough to neutralize the negative impact of myristic acid, as suggested by Khosla and Sundram [41]. Our data support the hypothesis raised by these authors in the sense that the effect of some oils could be modified by the accompanying fats and results could be different from those obtained when the pure fat is added to a semisynthetic diet.

Dietary cholesterol induced hypertriglyceridemia in rats confirming our previous results [32]. Othani et al. [42] reported a similar effect without any change in the synthesis of apoB and we did not find any change in apoB message caused by dietary cholesterol [32]. In rats, most of VLDL remnants are eliminated from circulation by LDL receptor [11]. The hypertriglyce-ridemia could be due to the lower activity of LDL receptor found when cholesterol is added to food. No additional effect on the elevation of plasma triglyceride concentration induced by dietary cholesterol was ob-served when the diet was supplemented with saturated coconut oil (Fig. 1), a similar finding to that reported by Van Heek et al. [43] in rabbits. Plasma triglyceride levels of groups consuming corn or olive oil were similar to those found in control animals and lower than rats fed cholesterol, which may be regarded as an abolition of the hypertriglyceridemic effect of dietary cholesterol. These effects are similar to those found in humans and in rabbits [11,43] and corroborate our previous data [32] regarding neutralizing the cholesterol effect. Percentage of provided olive oil was irrelevant regarding this matter. Our results suggest that, in the rat model, unsaturation of diet is a potent modulator of plasma triglycerides.

Saturated diet did increase apoA-I concentration as described in other animal species and in humans [11,44]. Feeding monounsaturated fatty acids also in-creased plasma apoA-I levels with high percentages of fat. A similar effect has been described in humans [14] with diets containing 20% fat. A striking finding of our research is the variability in the association between plasma apoA-I and HDL-cholesterol levels in the dif-ferent diets. Several explanations could be proposed to explain this observation. Firstly, cholesterol and apoA-I are two independent components of HDL particles with different metabolic origins and fates, both potentially modulated by dietary components [6]. While cholesterol could be eliminated through SR-B1, this does not seem to be the case for apoA-I. Secondly, the rat is an HDL animal with a high content of HDL containing apoE, levels which in different diets need to be ascertained. Finally, apoA-I in rodents is not only present in HDL type particles but also as free protein as well in LDL floating particles [45]. The contribution of different diets to the latter aspect warrants further research (preliminary data have confirmed that distribution of apoA-I changes depending on diets). In corn oil, a high

(8)

correlation is found between HDL-cholesterol and apoA-I levels. Considering that an enhanced catabolism of apoA-I for these types of diets has been found in other studies [46], it is highly possible that cholesterol content of HDL together with apoA-I are equally removed from the HDL particles in these diets. This high apoA-I catabolism would also explain the low plasma levels despite normal apoA-I mRNA and would be consistent with the strong and negative correlation between apoA-I mRNA and plasma apoA-I (Table 5). Although a lower translation of apoA-I mRNA is possible, this mechanism would have to be proved.

The amount of apoA-I mRNA found in the livers from control animals was 2194 pg/mg total RNA, a

figure similar to that published by Srivastava et al. [47]. Hepatic expression was similar among cholesterol and coconut, corn oil groups and chow diet group. Mo-nounsaturated fatty acid diets induced an increased expression of this message depending on the amount of fat provided. When either 10 or 40% fat was provided, a 43% increase was found, a result that corroborates our previous study, although the magnitude of response was of less intensity [32]. These data indicate that at least 10% olive oil is required to induce hepatic expres-sion. Based on the correlation data shown in Table 5, an even higher olive oil percentage is needed to show a high correlation between plasma apoA-I and its hepatic expression. Thus, dependence of apoA-I on its hepatic mRNA as a main factor determining plasma levels would only happen at these high intakes of olive oil. At this percentage of fat, dissociation of behavior of apoA-I and other HDL components may be important as is suggested by the low correlation coefficient between apoA-I and HDL-cholesterol. Using saturated fat, Dreon et al. [48] have shown that fat intake modulated hepatic lipase activity in humans. An inverse correla-tion between hepatic lipase and HDL concentracorrela-tions is currently found [49]. These facts suggest that a similar situation could be involved in olive oil diet which could explain the HDL-cholesterol levels observed at 40% olive oil diet. The extent of dependence of all these mechanisms on olive oil intake in other non-HDL animals seems plausible. In cynomolgus monkeys, when fed a 14% monounsaturated fat diet, there is a trend to increase plasma apoA-I although not statistically sig-nificant [44], and in humans, a 20% supply of monoun-saturated fatty acids induces an increase in plasma apoA-I [13,14]. As posed by the latter authors, there seems to be a threshold for the effect of monounsatu-rated fatty acids. Whether this dependence on the amount of monounsaturated fat is a general phe-nomenon in all species or is something specific for rats, due to the fact that it is an HDL animal, requires more experimental work due to the paucity of studies using monounsaturated fatty acid diets.

In saturated fat-fed animals, hepatic mRNA is not responsible for plasma levels, suggesting that a slow catabolism of apoA-I may be the determinant cause, as described for cynomolgus monkeys [46]. In hamsters consuming saturated fat, an increase in HDL particle size was found also suggesting impairment in the catabolic processing of these particles in the presence of this specific kind of fat [50]. Besides these aspects related to HDL dynamics, other molecular mechanisms might be implicated. Arzolan et al. [51], in transgenic mice, and Srivastava et al. [21], in normal mice, showed that saturated fat induced a more efficient translation of apoA-I mRNA. This mechanism acting for saturated fatty acid diets seems to be specific for rodents since in primates fed these types of diets there is a close correla-tion between hepatic mRNA and plasma apoA-I levels. In conclusion, it is clear that the plasma apoA-I increase in saturated and monounsaturated fatty acid diets has different mechanisms. Likewise, for the satu-rated diet there are posttranscriptional or posttransla-tional mechanisms with different degree of involvement depending on species. In the case of monounsaturated diets, our study indicates that hepatic mRNA may be responsible for the increase observed in plasma. This dependence on apoA-I mRNA levels is clearly observed at a high fat content. Our results, regarding regulation of plasma apoA-I, provide further support for the notion of different mechanisms of action for the differ-ent fatty acids, influence of dietary fat contdiffer-ent and species-specific mechanisms. Further research is neces-sary to gain more insight into those mechanisms and to achieve a complete picture of them.

Acknowledgements

We would like to thank Dr M. Uceda for food analysis, Dr M. Iturralde and M.V. Martı´nez for their critical reading of this manuscript and Erik Lundin and Patricia Sabatini for their help in preparing this manuscript. This work was supported by grants awarded by DGICYT (SM90-002), FISS (95/704), CI-CYT (OLI96/2204) and Fundacio´n Mapfre-Medicina. C. Carrizosa was the recipient of a doctoral fellowship from the Universidad Complutense de Madrid.

References

[1] Castelli WP, Garrison RJ, Wilson PWF, Abbott RD, Kalous-dian S, Kannel WB. Incidence of coronary heart disease and lipoprotein cholesterol levels. The Framingham Study. J Am Med Assoc 1986;256:2835 – 8.

[2] Gordon DJ, Probstfield JL, Garrison RJ. High density lipo-protein cholesterol and cardiovascular disease: four prospective American studies. Circulation 1989;79:8 – 15.

(9)

[3] Dahlen GH, Guyton JR, Attar M, Farmer JA, Kautz JA, Gotto AM. Association of levels of Lp(a), plasma lipids, and other lipoproteins with coronary artery disease documented by angiog-raphy. Circulation 1986;74:758 – 65.

[4] Williams RR, Hopkins PN, Hunt SC, Wu LL, Hasstedt SJ, Lalouel JM, et al. Population-based frequency of dyslipidemia syndromes in coronary-prone families in Utah. Arch Intern Med 1990;150:582 – 8.

[5] Brown BG, Fuster V. Impact of management in stabilization of coronary disease. In: Fuster V, Ross R, Topol EJ, editors. Atherosclerosis and coronary artery disease. Philadelphia, PA: Lippincott-Raven, 1996:191 – 205.

[6] Kashyap ML. Mechanistic studies of high-density lipoproteins. Am J Cardiol 1998;82:42U – 8U.

[7] Schaefer EJ, Levy RI, Ernst ND, Van Sant FD, Brewer HB Jr. The effects of low cholesterol, high polyunsaturated fat and low fat diets on plasma lipid and lipoprotein cholesterol level in normal and hypercholesterolemic subjects. Am J Clin Nutr 1981;34:1758.

[8] Chong KS, Nicolosi RJ, Rodger RF, Arrigo DA, Yuan RW, Mackey JJ, et al. Effect of dietary fat saturation on plasma lipoproteins and high density lipoprotein metabolism of the Rhesus monkey. J Clin Invest 1987;79:675.

[9] Keys A, Anderson JT, Grande F. Serum cholesterol response to changes in the diet. Particular saturated fatty acids in the diet. Metabolism 1965;14:776 – 87.

[10] Rudel LL, Haines JL, Sawyer JK. Effects on plasma lipoproteins of monounsaturated, saturated, and polyunsaturated fatty acids in the diet of African green monkeys. J Lipid Res 1990;31:1873 – 82.

[11] Grundy SM, Denke MA. Dietary influences on serum lipids and lipoproteins. J Lipid Res 1990;31:1149 – 72.

[12] Mattson FH, Grundy SM. Comparison of effects of dietary saturated, monounsaturated, and polyunsaturated fatty acids on plasma lipids and lipoproteins in men. J Lipid Res 1985;26:194 – 202.

[13] Mata P, Garrido JA, Ordova´s JM, Bla´zquez E, Alvarez-Sala L, Rubio MJ, et al. Effect of dietary monounsaturated fatty acid on plasma lipoproteins and apolipoproteins in women. Am J Clin Nutr 1992;56:77 – 83.

[14] Mata P, Alvarez-Sala L, Nun˜o J, De Oya M. Effects of long-term monounsaturated — versus polyunsaturated — enriched diets on lipoproteins in healthy men and women. Am J Clin Nutr 1992;55:846 – 50.

[15] Breslow JL. Genetic basis of lipoprotein disorders. J Clin Invest 1989;84:373 – 80.

[16] Breslow JL. Lipoprotein transport gene abnormalities underlying coronary heart disease susceptibility. Annu Rev Med 1991;42:357 – 71.

[17] Stucchi AF, Hennessy LK, Vespa DB, Weiner EJ, Osada J, Ordovas JM, et al. Effect of corn- and coconut-oil containing diets with and without cholesterol on HDL ApoA-I metabolism and hepatic ApoA-I mRNA levels in Cebus monkeys. Arte-rioscler Thromb 1991;11:1719 – 29.

[18] Hennessy LK, Osada J, Ordovas JM, Nicolosi RN, Brousseau ME, Schaefer EJ. Effects of dietary fats and cholesterol on liver lipid content and hepatic apolipoprotein A-I, B and E and LDL receptor mRNA levels in Cebus monkeys. J Lipid Res 1992;33:351 – 60.

[19] Sorci-Thomas M, Prack MM, Dashti N, Johnson F, Rudel LL, Williams DL. Apolipoprotein A-I production and mRNA abun-dance explain plasma apoA-I and high density lipoprotein differ-ences between two non-human primate species with high and low susceptibilities to diet-induced hypercholesterolemia. J Biol Chem 1988;263:5183 – 9.

[20] Sorci-Thomas M, Prack MM, Dashti N, Johnson F, Rudel LL, Williams DL. Differential effects of dietary fat on the

tissue-spe-cific expression of the apolipoprotein A-I gene: relationship to plasma concentration of high density lipoproteins. J Lipid Res 1989;30:1397 – 403.

[21] Srivastava RAK. Saturated fatty acid, but not cholesterol, regu-lates apolipoprotein AI gene expression by posttranscriptional mechanism. Biochem Mol Biol Int 1994;34:393 – 402.

[22] Fox JC, McGill HC Jr, Carey KD, Getz GS. In vivo regulation of hepatic LDL receptor mRNA in the baboon: differential effects of saturated and unsaturated fat. J Biol Chem 1987;262:7014 – 20.

[23] Osada J, Ferna´ndez-Sa´nchez A, Diaz-Morillo JL, Aylagas H, Miro´-Obradors MJ, Ordova´s JM, et al. Hepatic expression of apolipoprotein A-I gene in rats is upregulated by monounsatu-rated fatty acid diet. Biochem Biophys Res Commun 1991;180:162 – 8.

[24] European Union Commission. Regulation ECC/2568/91 on the characteristics of olive and olive pomace oils and on their analytical methods. Off J Eur Commun 1991;L248 Anex II. [25] Siedel J, Hagele EO, Ziegenhorn J, Wahlefeld AW. Reagent for

the enzymatic determination of serum total cholesterol with improved lipolytic efficiency. Clin Chem 1983;29:1075 – 9. [26] Bucolo G, David H. Quantitative determination of serum

triglyceride by use of enzymes. Clin Chem 1973;19:476 – 82. [27] Kingston RE. In: Asubel FM, Brent R, Kingston RE, Moore

DO, Seldman JG, Smith JA, et al., editors. Current protocols in molecular biology. New York: Greene/Wiley-Interscience, 1987. [28] Haddad IA, Ordovas JM, Fitzpatrick T, Karathanasis SK. Linkage, evolution, and expression of the rat apolipoprotein A-I, C-III, and A-IV genes. J Biol Chem 1986;261:13268 – 77. [29] Feinberg AP, Vogelstein B. A technique for radiolabeling DNA

restriction endonuclease fragments to high specific activity. Anal Biochem 1983;132:6 – 13.

[30] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989:2.6 – 8.

[31] Sohal RR, Rohlf FJ. In: Cotter S, editor. The principles and practice of statistics in biological research, 2nd ed. New York: W.H. Freeman, 1981.

[32] Osada J, Fernańdez-Sańchez A, Diaz-Morillo JL, Miro´-Obra-dors MJ, Cebriań JA, Ordova´s JM, et al. Differential effect of dietary fat saturation and cholesterol on hepatic apolipoprotein gene expression. Atherosclerosis 1994;108:83 – 90.

[33] Morson LA, Michael TM, Clandinin PD. Dietary linoleic acid modulates liver plasma membrane unsaturated fatty acid compo-sition, phosphatidylcholine and cholesterol content, as well as glucagon stimulated adenylate cyclase activity. Nutr Res 1985;5:1113 – 20.

[34] Willett WC. Dietary fat and obesity: an unconvincing relation. Am J Clin Nutr 1998;68:1149 – 50.

[35] Woollett LA, Spady DK, Dietschy JM. Regulatory effects of the saturated fatty acids 6:0 through 18:0 on hepatic low density lipoprotein receptor activity in the hamster. J Clin Invest 1992;89:1133 – 41.

[36] Dietschy JM, Turley SD, Spady DK. Role of liver in the maintenance of cholesterol and low density lipoprotein homeostasis in different animal species, including humans. J Lipid Res 1993;34:1637 – 59.

[37] Srivastava RAK, Ito H, Hess M, Srivastava N, Schonfeld G. Regulation of low density lipoprotein receptor gene expression in HepG2 and Caco2 cells by palmitate, oleate and 25-hydroxyc-holesterol. J Lipid Res 1995;36:1434 – 46.

[38] Srivastava RAK. Regulation of the apolipoprotein E by dietary lipids occurs by transcriptional and post-transcriptional mecha-nisms. Mol Cell Biochem 1996;155:153 – 62.

[39] Jackson B, Gee AN, Martı´nez-Cayuela M, Sackling KE. The effects of feeding a saturated fat-rich diet on enzymes of choles-terol metabolism in the liver, intestine and aorta of the hamster. Biochim Biophys Acta 1990;1045:21 – 8.

(10)

[40] Denke MA, Grundy SM. Comparison of effects of lauric acid and palmitic acid on plasma lipids and lipoproteins. Am J Clin Nutr 1992;56:895 – 8.

[41] Khosla P, Sundram K. Effects of dietary fatty acid composition on plasma cholesterol. Prog Lipid Res 1996;35:93 – 132. [42] Othani H, Hayashi K, Hirata Y, Dojo S, Nakashima K, Nishio

E, et al. Effects of dietary cholesterol and fatty acids on plasma cholesterol level and hepatic lipoprotein metabolism. J Lipid Res 1990;31:1413 – 22.

[43] Van Heek M, Zilversmit DB. Postprandial lipemia and lipo-protein lipase in the rabbit are modified by olive and coconut oil. Atherosclerosis 1989;10:421 – 9.

[44] Brousseau ME, Stucchi AF, Vespa DV, Schaefer EJ, Nicolosi RJ. A diet enriched in monounsaturated fats decreases low density lipoprotein concentrations in cynomolgus monkeys by different mechanism than does a diet enriched in polyunsatu-rated fats. J Nutr 1993;123:2049 – 58.

[45] De Silva HS, Mas-Oliva J, Taylor JM, Mahley RW. Identifica-tion of apolipoprotein B-100 low density lipoproteins, apoB-48 remnants, and apolipoprotein E-rich high density lipoproteins in the mouse. J Lipid Res 1994;35:1297 – 310.

[46] Brousseau ME, Schaefer EJ, Stucchi AF, Osada J, Vespa DV, Ordovas JM, et al. Diets enriched in unsaturated fatty acids

enhance apolipoprotein A-I catabolism but do not affect either its production or hepatic mRNA abundance in cynomolgus monkeys. Atherosclerosis 1995;115:107 – 19.

[47] Srivastava RAK, Srivastava N, Schonfeld G. Expression of low density lipoprotein receptor, apolipoprotein AI, AII, and AIV in various rat organs utilizing an efficient and rapid method for RNA isolation. Biochem Int 1992;27:85 – 95.

[48] Dreon DM, Fernstrom HA, Campos H, Blanche P, Williams PT, Krauss RM. Change in dietary saturated fat intake is correlated with change in mass of large low-density-lipoprotein particles in men. Am J Clin Nutr 1998;67:828 – 36.

[49] Kuusi T, Ehnholm C, Viikari J, Harkonen R, Vartiainen E, Puska P, et al. Postheparin plasma lipoprotein and hepatic lipase are determinants of hypo- and hyperlaphalipoproteinemia. J Lipid Res 1989;30:1117 – 26.

[50] Ahn YS, Smith D, Osada J, Li Z, Schaeffer EJ, Ordovas JM. Dietary fat saturation affects apolipoprotein gene expression and high density lipoprotein size distribution in golden Syrian ham-ster. J Nutr 1994;124:2147 – 55.

[51] Arzolan N, Odaka H, Breslow JL, Fisher EA. Dietary fat elevates hepatic ApoAI production by increasing the fraction of apolipoprotein A-I mRNA in the translating pool. J Biol Chem 1995;270:19833 – 8.

(1)

Table 3

Plasma lipid parametersa

HDL-cholesterol LDL-cholesterol

Experimental condition Total cholesterol Triglycerides

4493

Control (n=5) 6793 1994 41910

Cholesterol (n=4) 8597b ₄₅₉₄ ₃₃₉₃b ₆₂₉₁₀b

5094b

Cholesterol+40% coconut (n=5) ₇₅₉₅ 2393c ₆₉₉ ₁₄b

2994bcd ₁₅₉₃cd

4897bcd ₄₄₉₉cd

Cholesterol+40% corn (n=5)

67912ce

Cholesterol+40% olive (n=5) 3895d ₂₈₉₆bde ₄₇₉₁₉

4493 2493c

Cholesterol+10% olive (n=4) 7394 5097

4097 2494c

69912 4399c

Cholesterol+5% olive (n=5)

PB0.003

ANOVA PB0.002 PB0.0015 PB0.02

a_{Data, expressed as mg/dl, are means and their standard deviations. Statistical analysis was done using one way ANOVA according to Tukey}

and Tukey–Kramer Multiple Comparison Test as post hoc analysis. Different superscripts (b _{versus control;} c _{versus cholesterol;} d _versus

cholesterol+coconut oil; ande_{versus cholesterol+corn oil), are significantly different from each other at}_P_B0.05.

(2)

Table 4

Plasma concentration and hepatic expression of rat apolipoprotein A-I in the different experimental dietsa

Hepatic expression (pg of Plasma

Experimental

condition concentration apoA-I mRNA/mg total RNA)

(mg/dl) 66934

Control (n=5) 2194

Cholesterol 49935 1695

(n=4)

Cholesterol+40% 101934bc ₁₆₉₃

coconut (n=5)

71911 1693

Cholesterol+40% corn (n=5)

102912bce

Cholesterol+40% 28911cde

olive (n=5)

8996cf

Cholesterol+10% 2796c

olive (n=4)

Cholesterol+5% 89916 19913

olive (n=5)

ANOVA PB0.04 PB0.05

a_{Data are means and their standard deviations. Hepatic}

expres-sions are normalized to the levels of GPDH expression. Statistical analysis was done using non-parametric one-way ANOVA according to Kruskal–Wallis test and unpaired Mann–Whitney U-test to test pair-wise differences. Different superscripts (b_{versus control;}c_versus

cholesterol;d _{versus cholesterol+coconut oil;}e_{versus cholesterol+}

corn oil; andf_{versus cholesterol+40% olive) are significantly}

differ-ent from each other atPB0.05.

analyzed against coconut or corn oils because the con-ditions are not comparable due to two different parameters involved: source of fat and fat intake.

positive correlation was found between plasma apoA-I and HDL-cholesterol, the degree of association ap-peared highly variable depending on the diet. While highly significant correlations were noted for corn and 10% olive oil groups, weaker associations were ob-served in the remaining groups.

hep-atic apoA-I mRNA. However this dramhep-atic effect is only observed when supplied dietary fat is 40% (w/w). A significant negative correlation (rs= −0.97, PB

0.01) was found in the corn oil group and no significant association was observed in the remaining groups.

4. Discussion

Table 5

Relationship between parameters in the different experimental dietsa

Plasma apoA-I concentration

Control Cholesterol Coconut Corn 40% Olive 10% Olive 5% Olive

(n=5)

(n=5) (n=5) (n=4)

(n=4) (n=5)

(n=5)

0.4 0.6

0.4

HDL-cholesterol 0.9** 0.4 0.82* 0.7

ApoA-I mRNA 0.5 0.8 −0.15 −0.97** 0.97** 0.4 0.2

a_{Spearman correlation coefficients}_r

sare indicated.n, number of animals in each experimental group.

*PB0.05. **PB0.01.

(3)

(4)

The amount of apoA-I mRNA found in the livers from control animals was 2194 pg/mg total RNA, a figure similar to that published by Srivastava et al. [47]. Hepatic expression was similar among cholesterol and coconut, corn oil groups and chow diet group. Mo-nounsaturated fatty acid diets induced an increased expression of this message depending on the amount of fat provided. When either 10 or 40% fat was provided, a 43% increase was found, a result that corroborates our previous study, although the magnitude of response was of less intensity [32]. These data indicate that at least 10% olive oil is required to induce hepatic expres-sion. Based on the correlation data shown in Table 5, an even higher olive oil percentage is needed to show a high correlation between plasma apoA-I and its hepatic expression. Thus, dependence of apoA-I on its hepatic mRNA as a main factor determining plasma levels would only happen at these high intakes of olive oil. At this percentage of fat, dissociation of behavior of apoA-I and other HDL components may be important as is suggested by the low correlation coefficient between apoA-I and HDL-cholesterol. Using saturated fat, Dreon et al. [48] have shown that fat intake modulated hepatic lipase activity in humans. An inverse correla-tion between hepatic lipase and HDL concentracorrela-tions is currently found [49]. These facts suggest that a similar situation could be involved in olive oil diet which could explain the HDL-cholesterol levels observed at 40% olive oil diet. The extent of dependence of all these mechanisms on olive oil intake in other non-HDL animals seems plausible. In cynomolgus monkeys, when fed a 14% monounsaturated fat diet, there is a trend to increase plasma apoA-I although not statistically sig-nificant [44], and in humans, a 20% supply of monoun-saturated fatty acids induces an increase in plasma apoA-I [13,14]. As posed by the latter authors, there seems to be a threshold for the effect of monounsatu-rated fatty acids. Whether this dependence on the amount of monounsaturated fat is a general phe-nomenon in all species or is something specific for rats, due to the fact that it is an HDL animal, requires more experimental work due to the paucity of studies using monounsaturated fatty acid diets.