effect of nutritional status on the fatty acid composition of rat liver and cultured hepatocytes

454

Effect of Nutritional Status on the Fatty Acid Composition of Rat Liver and Cultured Hepatocytes Gary J. Helson, Darshan S. Kelley and James E. Hunt Biochemistry Research Unit, Western Human Nutrition Research Center, Agricultural Research Service, U.S. Department of Agriculture, Presidio of San Francisco, CA 94129

The lipid concentration and fatty acid composition of the whole liver and of cultured hepatocytes isolated from the livers of rats fed ad l ibitum (fed}, fasted for 24 hr (fasted}, or fasted for 48 hr and then refed a fat-free, high carbohydrate diet for 48 hr (refed) was studied. Hepatocytes were maintained as monolayer cultures in serum-free, lipid-free media and their fa t ty acid composition was analyzed at 3, 24, 48, 72 and 96 hr. The livers of fed animals, as well as their hepatocytes, contained less total lipid than those from animals on either of the other dietary regimes. Livers of fasted animals had three times the amount of lipid found in the livers of fed animals, and the livers of refed animals contained five t imes the amount of lipid as the livers of fed animals {all based on m g lipid/g wet weight of liver}. The fat ty acid composi- t ion of hepatocytes after 3 hr of culturing was very similar to that of fresh liver when compared in each of the dietary regimes. However, while the fat ty acid composi t ions of livers and hepatocytes from fed and fasted animals were similar, the pattern in liver of refed animals was quite distinct from that of the fed animals. In the fed and fasted animals palmitic acid {16:0}, stearic acid (18:0), oleic acid (18:1[n-9]}, linoleic acid (18:2[n-6]) and arachidonic acid {20:4[n-6]~ were the major fat ty acids of the liver; in refed animals 16:0, palmitoleic acid {16:1[n-7]), 18:0, 18:1(n-9) and c/s-vaccenic acid {the n-7 isomer of oleic acid} were the major fa t ty acids. During maintenance in culture the 18:1{n-9) content of the hepatocytes increased in cells from livers of animals on all three dietary regimes. The polyunsaturated fat ty acid content was similar in fresh livers and isolated hepatocytes in all samples when compared on the basis of pg fat ty acid/rag of hepatoeyte or liver protein. It was also found that the polyunsaturated fat ty acid content of hepatocytes was remarkedly stable with time of culture when the cells were incubated in serum-free, lipid-free medium. Thus, isolated hepatocytes maintained in serum-free medium appear to be a possible system for the evaluation of the effects of prior nutrit ional s ta tus on fat ty acid metabol i sm in the whole animal, not subject to hormonal and other somatic in- fluences which often complicate the interpretation of such nutrit ional studies. Lipids 21, 454-459 (1986).

The liver is the primary organ for the metabolism of dietary fatty acids (1-3}. It also is well established that dietary fat ty acids repress endogeneous fatty acid synthesis in the liver, while fat-free, high carbohydrate diets stimulate fat ty acid synthesis (4-7). Short term fasting in rats increases liver lipid content (8,9) while decreasing total liver mass and liver glycogen. The mechanisms of repression, induction and fat mobilization have not yet been completely elucidated (3,4,7}.

Hepatocytes can be isolated from rat liver and maintained in serum-free media for several days (10). After a recuperative period of variable length depending on

treatment and culture media, the cells recover many of their characteristic enzyme activities and cellular functions (11). In this work, the effects of either feeding ad libitum, fasting for 24 hr, or fasting for 48 hr and then refeeding a high carbohydrate, fat-free diet (referred to as fed, fasted and refed animals, respectively) on the fatty acid composition of fresh whole liver and isolated hepatocytes maintained in serum- and lipid-free medium for 96 hr was studied. The objective of this research was to determine the effect of prior nutritional status of the animals on the fatty acid composition of the hepatocytes in culture. It was also of interest to observe how the polyunsaturated fatty acids are affected in apparently healthy cells maintained in medium devoid of these essen- tial nutrients. It was suspected that they must be con- served under these conditions.

MATERIALS AND METHODS

Materials and animals. Male Sprague-Dawley rats (100-150 g) were obtained from Bantin and Kingman (Fre- mont, California). They Were fed laboratory chow obtained from Ralston-Purina (Richmond, Indiana). Waymouth's MB 752/1 and Swim's S-77 media were purchased from GIBCO (Santa Clara, California). Dexamethasone sodium phosphate and crystalline pig insulin were gifts of K. Bohra (Organon, West Orange, New Jersey) and W. W. Bromer (Eli Lilly and Co., Indianapolis, Indiana), respectively. Triiodothyronine and BSA Fr V (essentially fatty acid-free} were bought from Sigma Chemical Co. (St. Louis, Missouri). Collagenase type II was obtained from Worthington Biochemicals (Freehold, New Jersey). Purified fat ty acid methyl ester reference standard mixtures were purchased from Nu-Chek-Prep (Elysian, Min- nesota and Supelco Inc. (Bellefonte, Pennsylvania). Organic solvents were obtained from Burdick and Jackson (Muskegon, Michigan).

The animals were placed on one of three dietary regimes: 1, fed lab chow ad libitum; 2, fasted for 24 hr; 3, fasted for 48 hr, then refed a high carbohydrate diet for 48 hr. The high carbohydrate diet contained 5% glycerol, 30% egg albumin and 65% sucrose. The animals were maintained in individual cages with free access to water. The ambient temperature of the animal quarters was 25 C. The room was lighted between 6 a.m. and 6 p.m. Animals were anesthesized with nembutal between 8 and 9 a.m.; the livers were either removed immediately and extracted or perfused for isolation of hepatocytes as described below. Animals whose livers were perfused weighed 300-400 g at the time of perfusion. Animals whose livers were extracted directly weighed ca. 300 g.

Hepatocyte isolation and culture. Hepatocytes were isolated by the method of Berry and Friend (12) as modified by Bonney et al. (13) with the following additional changes. Livers in situ were perfused at 37 C with 300 ml of calcium-free Swim's S-77 medium containing BSA (1 mg/ml) at a flow rate of 25 ml/min. After

LIPIDS, Vol. 21, No. 7 (1986)

DIET AND HEPATIC FATTY ACIDS

455

perfusing for 10 min with the calcium-free medium, calcium {3 mM) and coUagenase (0.5 mg/ml) were included in the perfusion medium. The collagenase-containing medium was recirculated through the liver while the calcium-free medium was not. The liver cells were dis- persed after 7-8 min perfusion with the collagenase- containing medium. Liver cells were collected in calcium- free Swim's S-77 medium, treated with DNase (10~g/ml) and then cooled, after which trypsin inhibitor was added to the cell suspension as previously reported by Kelley and Potter (14). Hepatocytes were isolated by filtration and centrifugation as previously described (15). H e p a t o c y t e s were suspended (2 million/ml) in Waymouth's MB 752/1 medium supplemented with insulin (2 X 10 -7 M), dexamethasone (10 -6 M), triiodothyronine {10 -6 M), penicillin (100 U/ml), streptomycin (100 ~g/ml), gentamycin (20 ~g/ml) and defatted bovine serum albumin, fraction V (0.5 mg/ml). The cell suspension was inoculated (6 ml/dish) in 100-mm collagen-coated culture dishes, maintained at 37 C in an atmosphere of 95% air and 5% C02. Medium was initially changed at 3 hr after plating and then every 24 hr. The Waymouth's medium was aspirated, cells were washed four times with Hank's Hepes BSS and then cells from each culture dish were scraped into 5 ml of H~O containing hydroquinone (20 ~g/ml).

Lipid extraction. The lipid extractions were performed as described previously ( 16,17) with some modifications. Briefly, the samples, either washed hepatocyte suspension or homogenized fresh liver, were first lyophilized to remove water. The freeze-dried samples were extracted with CHCI3/MeOH (2:1, v/v). The extracts then were filtered to remove insoluble protein, inorganic salts and carbohydrates.

The extraction solvents were removed by evaporation with dry N2 at 40 C. The total lipid extracts were further purified by redissolving in dry chloroform, refiltering, evaporating and drying in a vacuum dessicator. Then they were weighed and either prepared for transmethylation or stored at low temperature ( - 20 C) until analyzed further.

Transmethylation. The total lipid extracts were transmethylated directly without prior separation of the various lipid classes. The samples were placed in 15 • 150 mm screw cap culture tubes fitted with reflux con- densers {18). Five ml of 7% methanolic-HC1 was added to each tube which was then heated to between 85 and 90 C for 2 hr. The tubes were cooled to room temperature and 10 ml of redistilled H20 was added to each tube. The fat ty acid methyl esters (FAME) then were extracted by three washings of the aqueous phase with 3 ml of hexane. The hexane washes were combined and dried over NaHCO3/Na~SO4 (4:1, w/w) for 1 hr and evaporated to dryness at 40 C. The FAME then were transferred to vials which had been previously tared. The weights of FAME were determined and the samples diluted to an appropriate concentration in hexane and prepared for analysis by gas liquid chromatography (GLC).

GLC. The fat ty acid methyl ester samples were analyzed on fused silica, wall-coated capillary column, 0.025 mm I.D. • 30 m long. The columns, coated with SP-2340, were obtained from Supelco Inc. Samples were run on either a Hewlett-Packard 5880A gas chromatograph with an attached Hewlett-Packard automated data station or

a Perkin-Elmer Sigma 2000 gas chromatograph coupled to a Perkin-Elmer 7500 computer with a Chrom 3 data analysis program.

The column conditions were as follows: The carrier gas was helium at 14.5 psi inlet pressure. The samples were injected in 1 gl of hexane at a split ratio of 50:1. The FAME were separated on the column using a temperature program from 150 to 190 C. The first ramp was 2 C/min for 10 min to 170 C, with a hold at that temperature for 5 min. The second ramp was 5 C/min for 4 min to 190 C, with a hold at that temperature for 15 min. The FAME were tentatively identified by comparison of their reten- tion times (RT) to RT of reference standards obtained from either Nu-Chek-Prep or Supelco Inc. The GLC detector was a hydrogen flame ionization detector.

The quantitative accuracy of the GLC procedures was evaluated by using either purified single FAME or reference mixtures, and the reference FAME or mixtures were selected to cover the range of fat ty acid methyl esters present in the samples. Quantitative reference standards (either single FAME standards or mixtures) were prepared in the concentration ranges expected for the FAME mixtures in the natural samples. Accuracy of the quantitation was estimated to be within 5% for major components in the samples (those present in amounts greater than 10% of total FAME in the sample) and within 10% for components present in less than 10% of total FAME in the sample.

I t should be noted that the cholesterol extracted into the hexane washes of the transmethylation mixture was not separated from the FAME prior to injection of the samples in the chromatograph. However, free cholesterol does not elute as a discrete peak from a SP-2340 column operated under the conditions described here. Hence, there is no interference in FAME analysis either quan- titatively or qualitatively.

The percentage compositions of FAME in the rat hepatocyte samples were calculated as wt % transformed from the area percentages in the chromatogram. The transformation was accomplished by using a response factor for each individual FAME calculated from purified reference standards and the calibration procedures pro- vided by the instrument manufacturers. Generally, the flame ionization detector is quite linear in response over the range of FAME found in the samples with respect to both the mass of compound and its RT, particularly for the sample sizes used with capillary columns (19,20).

Only the major fat ty acids are listed individually in Tables 2 and 3 as the capillary GLC analytical system used to collect this data yielded between 40 and 80 discrete peaks depending on the load of sample placed on the column. Usually individual minor peaks contributed less than 0.1% to the total area of the chromatogram, and many were unidentified. In a typical analysis, 20 identified components comprised 97-98% of the total fat ty acid present, while 20 to 40 unidentified components accounted for the remaining 2-3%. Thus, to simply the presentation of the data, the 10 fat ty acids identified in Tables 2 and 3 are the only compounds for which individual data are presented. The residue is collectively presented as "sum of trace components" and varies from ca. 5-9% of the total fat ty acids present. Both identified and unidentified components are grouped together in this category.

LIPIDS, Vol. 21, No. 7 (1986)

456

G.J. NELSON ET AL.

The d a t a in F igures 1, 2 and 3 are p r e s e n t e d in t e rms of ~g f a t t y ac ids /mg cel lular p ro te in because th is allows the a m o u n t of each f a t t y acid to be compared in abso lu te t e rms r a the r t h a n the normal ized va lues g iven by a wt % calcula t ion. P ro te ins were ana lyzed by the me thod of Lowry et al. (21).

RESULTS

Table 1 gives da t a for the con ten t of l ipids and to ta l f a t ty acids in the l ivers of r a t s in the th ree n u t r i t i o n a l s t a tes s t ud i ed in th is work. Af te r f a s t i ng for 24 hr the l iver had lost 1/3 (5-6 g) of i t s wet weight . Livers f rom ra t s fas ted for 48 hr and t h e n refed a fat-free diet for 48 hr rega ined 2 to 3 g of wet weight . However , b o t h the fas ted and refed an ima l s had a larger pe rcen tage of the i r l iver weight as l ipid t h a n the fed an imals when resu l t s were compared as percentage of wet weight or ~g l ipid/mg of liver protein.

U n d e r our e x p e r i m e n t a l condi t ions , m g pro te in /g wet weight of l iver did no t change s ign i f i can t ly a m o n g t h e three d ie t a ry g roups of ra ts . These va lues were 149.8 + 14.4, 164.3 _+ 8.0 a n d 150.8 +_ 20.5 for the fed, fas ted and refed ra ts , respect ively .

F a t t y acids made up ca. 32% of the to t a l l iver l ipid in fed an ima l s a nd 24% in the fas ted animals , while in the refed an imals f a t t y acids were over 45% of the to t a l l ipid fract ion. The phosphol ipids comprised c& 14, 46 and 61% of the to t a l l ipid in the l iver of refed, fed a nd s t a rved animals , respect ively .

Table 2 gives the fa t ty acid composi t ion of fresh whole liver lipids from ra ts on the three dietary regimes. In the fed animals the five fa t ty acids, palmitic acid (16:0), stearic acid (18:0), oleic acid (18:1[n-9]), linoleic acid (18:2[n-6]) and arachidonic acid (20:4[n-6]), made up ca. 80% of tota l fa t ty acids, while docosahexaenoic acid (22:6[n-3]) accounted for a lmost 8% of the total fa t ty acids.

TABLE 1

Effect of Dietary Status on Lipid Content of Rat Liver

Dietary status Weight of Wt % % Phospholipid ~g Fatty acid Wt % fatty acid of animals fresh liver (g) lipid a of total lipid b mg Protein c of total lipid d

Ad lib ted 15.97 • 1,95 3.67 • 0.70 45,8 • 8.0 65.85 • 21.28 31.7 • 2.7 24-hr fasted 9.69 • 0.71 4.69 • 1.15 60.5 • 9.3 163.37 • 69.57 24.2 • 6.6 48-hr fasted,

48-hr refed fat-free diet 12.13 • 2.53 5.59 • 0.51 13.5 • 4.5 245.77 • 47.58 45.4 • 3.6

Data shown are averages • S.D. (n = 3 for each treatment). aBoth starved and refed group significantly different than fed, P < 0.01. bCalculated using ~g phosphorus time 25. Both fed and starved group significantly different than refed, P < 0.05. cCalculated from GLC mass areas and Lowry Protein determination. Refed group significantly different than fed, P < 0.05. dCalculated from GLC mass areas and initial gravimetric wt of total lipid extract. All three groups significantly different from each other, P < 0.01.

TABLE 2

Fatty Acid Composition of Whole Rat Liver as Affected by Dietary Status

Ad libitum-fed

Fatty acid Wt % ~g FA/mg protein

24-Hr fasted 48-Hr fasted, 48-hr refed fat-free diet

Wt % ~g FA/mg protein Wt % ~g FA/mg protein

14:0 0.34 -- 0.17 0.22 _ 0.11 0.26 • 0.03 0.42 --+ 0.05 1.25 • 0.39* 3.07 • 0.96* 16:0 21.06 -- 0.75* 13.87 -- 0.49* 16.30 +_ 0.79* 26.63 _ 1.29 32.53 -- 3.07* 79.95 -- 7.55* 16:1(n-9) 0.33 • 0.09* 0.22 • 0.06* 0.18 • 0.04* 0.29 • 0.07* 0.75 • 0.09* 0.61 • 0.22* 16:1(n-7) 1.08 • 0.17" 0.71 • 0.11" 0.60 • 0.05* 0.98 • 0.08* 15.39 • 1.85" 37.82 • 4.55* 18:0 14.89 • 1.99" 9.81 • 1.31" 21.03 • 0.31" 34.36 • 0.51" 3.51 -- 0.52* 8,63 • 1.28" 18:1(n-9) 11.76 • 1.88" 7.74 • 1.29" 6.00 +_ 0.25* 9.80 • 0.41" 28.30 • 1.03" 69.55 • 2.53* 18:1(n-7) 2.77 • 0.13" 1.82 • 0.09* 1.76 _+ 0.10" 2.88 • 0.16" 6.70 • 0.68* 16.47 • 1.67" 18:2(n-6) 16.80 • 0.26* 11.86 • 0.17" 15.02 • 0.64* 24.54 • 1.05" 2.39 • 1,56" 5.87 • 3.83* 20:4(n-6) 14.22 • 0.36* 9.36 • 0.24* 21.44 • 0.50* 35.03 • 0.82* 2.89 • 1.17" 7.10 • 2.88* 22"6(n-3) 7.68 • 1.06 5.86 • 0.70 8.42 • 1.28 13.76 _ 2.09 1.31 • 0.52* 3.22 • 1.28" Sum of trace

components 9.00 • 0.36 5.73 - 0.24 8.68 • 0.61 14.18 • 1.00 4.57 • 0.84* 11.23 • 2.06

Fatty acid designated by chain length, number of double bonds and the position of the first double bond from the methylene end of the molecule. Data shown are the average + S.D. (n -- 3 for each treatment). *Concentration of fatty acids in the fed and starved groups indicated by asterisks are significantly different at P < 0.05, and both are significantly different from the corresponding values in the refed group by at least P < 0.01.

I_IPIDS, VOl. 21, No. 7 (1986)


457

The fa t ty acid pa t t e rn was markedly different in livers of the refed animal compared to tha t in livers of fed or fasted rats. The percentage of palmitic, palmitoleic acid (16:1[n-7]) and oleic acid was elevated in liver of refed ra ts and accounted for over 75% of the to ta l f a t ty acids present, but stearic acid and the po lyunsa tura ted f a t t y acids were noticeably decreased and accounted for ca. 10% of to ta l f a t ty acids collectively.

Table 3 gives the distr ibution of f a t t y acid in hepatocytes which were isolated f rom the livers of r a t s on the three different d ie tary regimes and mainta ined in the described culture medium for 3 hr. The fa t ty acid pat terns o f the isolated hepatocytes were quite similar to tha t found in the corresponding fresh intact livers; the cells from fed and fasted animals had a similar f a t t y acid composition which was dist inctly different from the composi- t ion found in the refed animal.

While the experiments reported here were not designed to measure secretion of f a t t y acids into the medium, in a few instances all medium was collected mid pooled from each change of medium and analyzed for to ta l f a t t y acid content. Significant amounts of f a t ty acids could n o t b e detected in medium used to culture cells f rom fas ted or refed animals (detection limits about 5% of the cellular f a t ty acid content). Medium collected for 96 hr f rom cultured cells f rom fed animals had a total f a t t y acid con- ten t a t the level of 10% of the cellular f a t t y acid content. The major f a t t y acids present in the medium were palmitic, palmitoleic, stearic and oleic. Thus, cellular fa t ty acid content or composit ion would not be markedly altered by secretion of fa t ty acids into the culture medium in these series of experiments .

Figure 1 shows the concentrat ion of six major f a t t y acids for hepa tocytes isolated f rom the liver of a fed r a t and analyzed at five different times in culture. I t was only when the da ta were analyzed in te rms of ~g f a t t y acid/mg cellular protein tha t the cons tancy of the concentrat ion of po lyunsa tura ted f a t ty acids in the cell with t ime in the medium became readily apparent . Conversely, the satu- ra ted and monounsa tu ra t ed f a t t y acid concentrat ion in

the cells was al tered significantly with t ime in culture. As with the whole liver, we could detect no significant changes in the average protein content per hepa tocy te during the t ime course of this experiment; thus, the ~g f a t t y acid/mg cellular protein is an accurate reflection of amount of f a t ty acid per cell in these experiments .

Figure 2 gives t ime-dependent composit ional informa- tion on the identical f a t t y ac ids f rom the hepatocytes isolated f rom livers of fas ted animals and mainta ined in culture for 96 hr. The changes parallel those observed in the cells f rom fed animals (Fig. 1} with only minor varia- tions. Figure 3 shows the da ta for these f a t t y acids in hepatocytes isolated f rom the liver of refed animals and maintained in culture for 96 hr. In these cells the pa t t e rn of the changes in the cellular fa t ty acid concentration with t ime in culture was str ikingly different f rom those of ei ther the fed or fas ted animals. The concentrat ion of palmitic acid decreased significantly while the concentra- t ion of stearic and oleic acids increased and the concen- t rat ion of polyunsaturated fa t ty acids appeared relatively unaltered.

DISCUSSION

The liver weights and lipid content of the livers repor ted here are consistent with literature values (8,22,23) for normal animals of the age and size used in this work. A 24-hr fas t decreased the to ta l liver wet weight by one-third, p resumably by the loss of liver glycogen (24,25), while refeeding a fat-free diet high in ca rbohydra te for 48 hr res tored some of the wet weight of the liver by replacing some of the liver glycogen and increasing the triglyceride content of the liver (6-8). I t is interesting, however, t ha t in the liver of fas ted animals, the to ta l lipid as well as the phospholipid content was higher than the corresponding values in the liver of fed animals, even when corrected for the decrease in the weight of the liver in the s ta rved animals.

Even though de novo f a t ty acid synthesis is repressed in fasted animals (4,5}, phospholipid synthesis or turnover

TABLE 3

Fatty Acid Composition of 3-Hr Cultured Hepatocytes as Affected by Prior Nutritional Status of Intact Animal

Ad libitum-fed

Fatty acid Wt % ~g FA/mg protein

24-Hr fasted 48-Hr fasted, 48-hr refed fat-free diet

Wt % ~g FA/mg protein Wt % pg FA/mg protein

14:0 0.69 • 0.10" 0.67 + 0.09* 0.42 __ 0.13" 0.64 • 0.20* 1.57 • 0.05 5.62 • 0.18" 16:0 21.30 __ 0.37* 20.87 • 0.36* 17.58 • 0.78* 26.90 + 1.19" 28.28 • 0.41 101.27 • 1.47" 16:1(n-9) 0.41 • 0.07* 0.40 +_ 0.07* 0.23 • 0.07* 0.35 • 0.11" 0.65 • 0.05 2.33 • 0.18" 16:1(n-7) 1.69 • 0.09* 1.66 • 0.09* 0.65 +_ 0.13" 0.99 • 0.20* 13.02 • 0.21 46.62 • 0.75* 18:0 12.99 • 0.22* 12.73 • 0.22* 20.60 • 1.54" 31.52 • 2.36* 4.29 • 0.33 15.36 • 1.18" 18:1(n-9) 11.76 • 0.23* 11.52 • 0.22* 6.11 • 0.46* 9.35 • 0.70* 30.18 • 1.19 108.07 • 4.26* 18:1(n-7) 4.45 • 0.05* 4.36 +_ 0.05* 2.26 • 0.17" 3.46 • 0.26* 7.37 • 0.13 26.39 • 0.47* 18:2(n-6} 18.12 • 0.21" 17.76 • 0.21" 14.76 • 0.50* 22.58 • 0.77* 4.34 • 0.17 15.54 • 0.61" 20:4(n-6} 14.24 • 0.27* 13.96 • 0.26* 22.04 • 1.22" 33.72 • 1.87" 4.77 • 0.44 17.08 • 1.58" 22:6(n-3) 4.92 • 0.28* 4.82 • 0.27* 7.00 • 0.34* 10.71 • 0.52* 1.62 • 0.16 5.80 • 0.57* Sum of trace

components 9.42 +_ 0.55 8.25 • 0.54 7.72 • 1.23" 11.81 • 1.88 3.92 • 1.11 14.04 • 3.97

Data shown are the average • S.D. (n -- 4 for each treatment). *Concentration of fatty acids in the fed and starved groups indicated by asterisks are significantly different at P < 0.05, and both are significantly different from the corresponding values in the refed group by at least P < 0.01.

LIPIDS, Vol. 21, No. 7 (1986)

458

G.J. NELSON ET AL.

presumably continues using fatty acids mobilized from peripheral tissues and adipose stores (8,26,27). The livers of animals refed a fat-free diet exhibit very active synthesis of fat ty acids (6,7) which are rapidly incorporated into triglycerides (28-30). This concept is consistent with data in Table 1 which indicates that the wt % of fatty acids of the total lipid in the liver was highest in refed animals and lowest in the fasted animal. Apparently fasted animals rapidly hydrolyze any liver triglyceride for energy or convert it to phospholipid (27,28).

The data reported here on the fatty acid pattern of whole rat liver is consistent with that reported previously (31-35). There was little difference between the overall

40.

.E 30 '

< 2 0

10-

- 4 0

- 3 0

- 2 0

. 10

I I I

2 4 4 8 7 2

Hours in Culture Medium

96

FIG. 1. Individual fatty acid content in I~g fatty acid per mg cellular protein for six major fat ty acids of rat hepatocytes isolated from livers of ad libitum-fed rats. The cells were maintained in a lipid- free, serum-free, high carbohydrate medium for 96 hr. Each point represents an average value from four experiments. Standard devia- tions are not shown to avoid clutter on figure, but they were generally less than 10% of the mean value. Fatty acids are designated by chain length, number of double bonds, and the position of first double bond from methylene end of the chain. In comparing the 3-hr and 96-hr points, all values for the fat ty acids are significantly different (P < 0.0D except for 18:0, which is not.

7 5 -

.E e o 5 0 -

r 50

4 24 48 72 96

14ours in Culture Medium

FIG. 2. Individual fatty acid content in ~g fatty acid per mg cellular protein for s ix major fat ty acids of rat hepatocytes isolated from livers of rats fasted for 24 hr. In comparing the 3-hr and 9Ghr points, values for 16:0, 16:1, 18:0 and 18:1 are significantly different (P 0.05); 18:2 and 20:4 values are not. Refer to the legend for Fig. I for additional details and explanation of shorthand designations of fatty acids.

pattern found in the livers of fed and fasted animals. Fat ty acid synthesis is inhibited in livers of the fasted animals and existing triglyceride fatty acids are con- verted to carbohydrate or phospholipids (28). However, chain elongation and some desaturation may still be oc- curring in the liver in both fed and fasted animals (36,37).

The liver from refed rats showed a very different fatty acid pattern relative to the wt % distribution of the major fatty acids. Palmitate, palmitoleate, oleate and cis- vaccenic acids (the n-7 isomer of oleic acid} were the predominant species present. The increase in palmitoleate was particularly striking as it rose from about 1% in the livers of fed animals to over 15% in the livers of refed animals. Almost equally dramatic was the decrease in the polyunsaturated fatty acids, linoleate, arachidonate and docosahexaenoate, which all decreased by a factor of 5 or more. This marked change in fatty acid pattern in the whole liver between fed and refed animals has been observed by others (38-41}.

These differences between the refed and the fed or fasted animals were probably caused by the rapid synthesis of saturated and monounsaturated fat ty acids in the liver of the refed animals (3,28,40,41), which were receiving a fat-free, high carbohydrate diet. As no polyunsaturated fat ty acids were present in the diet of these animals, saturated and monounsaturated fat ty acids increased in their liver due to de novo synthesis, while the amount of the polyunsaturated fat ty acids was constant in terms of ~g fatty acid/mg protein and decreased in percentage of total fat ty acids.

Several metabolic processes in hepatocytes are initially disturbed by their isolation and maintenance in culture; however, most of these functions recover by 24 hr (11,42). The fatty acid composition of the cells at 3 hr after plating (Table 3) resembled the fatty acid composition of the corresponding whole liver (Table 2). Various changes in the cellular fatty acid composition occurred during 24-96 hr in culture. These changes indicate that there was vigorous metabolic activity related to fat ty acid lipogenesis in cultured hepatocytes. This supposition is supported by

2 0 0 - - 2 0 0

150- .E

e

E ~ 100-

~ 50-

A

A /

16:1 (n -7 ) II -------.-.---__._._. i I

18:2 (n-6) e-

Hours in Cu l tu re M e d i u m

96

�9 150

�9 100

50

FIG. 3. Individual fat ty acid content in ~g fatty acid per mg cellular protein for six major fatty acids of rat hepatocytes isolated from livers of rats starved for 48 hr, then refed for 48 hr a high carbohydrate, fat-free diet. In comparing the 3-hr and 96-hr points, values for the fat ty acid content of the cells are significantly different for 16:0, 16:1 and 18:1 (P < 0.001) but not for 18:0, 18:2 and 20:4. Refer to the legend for Fig. I for additional details and explanation of shorthand designations of fa t ty acids.

LIPIDS, Vol. 21, No. 7 (1986)


459

our data for the activity of several lipogenic enzymes in these cells (43).

To our knowledge there are no previous reports on the fa t ty acid composition of primary cultures of hepatocytes maintained for 96 hr in serum-free medium, nor are there any reports on the fa t ty acid composition of hepatocytes cultured from rats fed high carbohydrate, fat-free diets prior to isolation of the cells. Previous reports (32,40,42, 44,45) on the fa t ty acid composition of cultured hepatocytes were mainly for hepatoma cell lines rather than primary hepatocyte cultures; however, our data obtained with primary cultures are consistent with the earlier reported values for these cultures.

When hepatocytes were maintained in culture, de novo synthesis of fa t ty acids occurred primarily in the cells from fed and fasted animals (Kelley et al., in preparation). Where there was de novo synthesis of fat ty acids, the content of palmitic acid in the cells remained constant (fasted animals) or increased (fed animals). Conversely, in the cells from the refed animals, the palmitic acid content decreased continuously with time in culture, presumably because it was being elongated to stearic acid which in turn was desaturated to oleic acid (36,37). A similar hypothesis can be constructed to explain the changes in the other saturated and monounsaturated fa t ty acids for which data are presented in Figures 1, 2 and 3. I t is particularly interesting to note the relation between palmitic, palmitoleic and cis-vaccenic acids. These fa t ty acids appear to have a metabolic pathway somewhat in- dependent of oleic acid. Indeed, in many reports, the cis- vaccenic acid content of mammalian tissue is ignored, assumed to be trivial or lumped with oleic acid; these are all unwarranted assumptions, in our opinion, as it appears cis-vaccenic is a major fa t ty acid at least in rat liver.

Our results appear to indicate tha t active synthesis of saturated and monounsaturated fat ty acids occurred during maintenance of pr imary cultures of rat hepatocytes for 96 hr while polyunsaturated fa t ty acids were con- served during this time. This observation does not rule out subtle changes in the polyunsaturated fa t ty acid metabol i sm, such as convers ion of l inoleate to arachidonate at a level below that detectable with the techniques used in this work. Thus, maintenance of hepatocytes in serum-free, lipid-free medium offers an op- por tuni ty to s tudy fa t ty acid metabolism in a system unperturbed by both hormonal agents and extraneous lipids associated with heterologous serum.

ACKNOWLEDGMENT Claire Serrato and Perla Schmidt gave technical assistance.

REFERENCES 1. Geelen, M.J.H., Harris, R.A., Beynan, A.C., and McClure, S.A.

(1980) Diabetes 29, 1006-1022. 2. Wakil, S.J., Stoops, J.K., and Joshi, V.C. (1983) Ann. Rev.

Biochem. 52, 537-579. 3. Herzberg, G.R. (1983) Ann. Rev. Nutr. 5, 221-253. 4. Hodge, H.C., MacLachlan, P.L., Bloor, W.R., Stoneburg, C.A.,

Oleson, M.C., and Whitehead, R. (1941) J. Biol. Chem. 139, 897-916.

5. Masoro, E.J., Chaikoff, I.L., Chernick, S.S., and Felts, J.M., (1950) J. Biol. Chem. 185, 845-854.

6. Gibson, D.M., Hicks, S.E., and Allmann, D.W. (1966)Adv. En- zyme Regul. 4, 239-246.

7. Kim, M., and Elson, C.E. (1981) J. Nutr. i l l , 1985-1995. 8. Suzuki, H., Goshi, H., and Sugisawa, H. (1975) J. Nutr. 105,

90-95. 9. Barley, J.C., and Abraham, S. (1972) Biochim. Biophys. Acta

280, 258-266. 10. Kelley, D.S., and Kletzien, R.F. (1984) Biochem. J. 217, 543-549. 11. Ichihara, A., Nakamura, T., Tanaka, K., Tomita, Y., Aoyama,

K., Kato, S., and Shinno, H. (1980) in Differentiation and Car- cinogenesis in Liver Cell Cultures (Borek, C., and Williams, G.M., eds.) pp. 77-84, New York Academy of Sciences, New York.

12. Berry, M.N., and Friend, D.S. (1969) J. Cell. Biol. 43, 506-520. 13. Bonney, R.J., Becker, J.E., Walker, J.E., and Potter, V.R. (1974)

In Vitro 9, 399-413. 14. Kelley, D.S., and Potter, V.R. (1979) J. Biol. Chem. 254,

6691-6697. 15. Kelley, D.S., and Potter, V.R. (1978) J. Biol. Chem. 253,

9009-9017. 16. Nelson, G.J. (1972) in Blood Lipids andLipoproteins: Quantita-

tion, Composition, and Metabolism (Nelson, G.J., ed.) pp. 3-24, Wiley-Interscience, New York.

17. Nelson, G.J. (1975) in Analysis of Lipids and Lipoproteins (Perkins, E.G., ed.) pp. 1-22, American Oil Chemists' Society, Champaign, IL.

18. Booth, R.A. (1971) in Laboratory Procedures for Analysis of Blood Lipids, Lawrence Livermore Laboratory Technical Report No. UCRL-51108, p. 17, University of California, Livermore.

19. Ackman, R.G. (1972) Progr. Chem. Fats Other Lipids 12, 165-284.

20. Slover, H.R., and Lanza, E. (1979) J. Am. Oil Chem. Soc. 56, 933-943.

21. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275.

22. Wadhwa, P.S., Elson, C.E. and Pringle, D.J. (1973)J. Nutr. 103, 899-903.

23. Suzuki, H., Tanada, M., and Imamura, M. (1979)J. Nutr. 109, 1405-1412.

24. Harper, A.E., Monson, W.J., Arata, D.A., Benton, D.A., and E1vehjem, C.A. (1953)J. Nutr. 51, 523-537.

25. Aoyama, Y., and Ashida, K. (1972) J. Nutr. 102, 631-638. 26. Swaggerson, E.D., and Bates, E.J. (1981} in Short-term Regula-

tion of Liver Metabolism (Hue, L., and Van de Werve, G., eds.) pp. 247-262, Elsevier/North Holland, Amsterdam.

27. Brindley, D.N. (1984) Prog. Lipid Res. 23, 115-133. 28. Zammit, V.A. (1984} Prog. Lipid Res. 23, 39-67. 29. Ide, T., and Ontko, J.A. (1981)J. Biol. Chem. 256, 10247-10255. 30. McGarry, J.D., and Foster, D.W. (1974) J. Biol. Chem. 249,

7984-7990. 31. Wood, R., Chumbler, F., and Wiegand, R.D. (1978) Lipids 13,

232-238. 32. Ruggieri, S., and Fallani, A. (1979) Lipids 14, 323-333. 33. Ciccoli, L., and Comporti, M. (1983) Lipids 18, 363-370. 34. Blomstrand, R., Diczfalusy, U., Sisfontes, G., and Svensson, L.

(1985) Lipids 20, 283-295. 35. Alam, B.S., and Alam, S.Q. (1983) J. Nutr. 113, 64-69. 36. Sprecher, H. (1974) Biochim. Biophys. Acta 360, 113-123. 37. Brenner, R. (1974) Mol. Cell. Biochem. 3, 41-52. 38. Allman, D.W., and Gibson, D.M. (1965) J. LipidRes. 6, 51-62. 39. Allman, D.W., Hubbard, D.D., and Gibson, D.M. (1965) J. Lipid

Res. 6, 63-74. 40. Wood, R., Falch, J., and Weigand, R.D. (1975) Lipids 10,

202-207. 41. Wood, R. (1975) Lipids 10, 736-745. 42. Pariza, M.W., Yager, J.D., Goldfarb, S., Gurr, J.A., Yanagi, S.,

Grossman, S.H., Becker, J.E., Barber, T.A., and Potter, V.R. (1975) in Gene expression and carcinogenesis in cultured liver (Gerschen, L.E., and Thompson, E.B., eds.), pp. 137-167, Academic Press, New York.

43. Kelley, D.S., Nelson, G.J., and Hunt, J.E. (1986) Biochem. J. 235, 87-90.

44. Wood, R., and Wiegand, R.D. (1975) Lipids 10, 746-749. 45. Zoeller, R.A., and Wood, R. (1984) Lipids 19, 529-538.

[Received October 7, 1985]

LIPIDS, Vol. 21, No. 7 (1986)

effect of nutritional status on the fatty acid composition of rat liver and cultured hepatocytes

Documents