![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Research Institute, Bassett Healthcare, Cooperstown, NY 13326
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
FOOTNOTES
LITERATURE CITED
ABSTRACT 
In this study, we tested the hypothesis that dietary linoleic acid intake controls the arterial blood plasma linoleic acid concentration and the rates of tumor growth and linoleic acid metabolism in vivo. Seven groups of young male Buffalo rats (11-21 rats/group) were given free access to semipurified diets containing different amounts of corn and/or olive oils. Four other groups (7-11 rats/group) were 30% energy-restricted. Each experiment included periods for rat growth and plasma lipid stabilization (6 wk), measurement of mean daily arterial blood plasma fatty acid concentrations (3 wk), surgical implantation of a subcutaneous tissue-isolated hepatoma 7288CTC, tumor growth and harvest (2-4 wk). Linoleic + arachidonic acid (P = 0.007) and oleic acid (P = 0.002) concentrations in arterial blood plasma were increased as dietary intake of linoleic and oleic acids was increased, respectively. In rats given free access to food, tumor growth was directly dependent on the plasma concentrations of linoleic (P < 0.001) and arachidonic acids (P = 0.04). Tumor growth in energy-restricted rats was dependent only on the linoleic acid concentration (P = 0.008). Energy restriction itself caused a growth inhibition independent of plasma linoleic acid. The linoleic acid and total fatty acid concentrations of tumor triacylglycerols were directly dependent on the plasma linoleic acid concentration in rats given free access to food (P = 0.009). Hepatoma 7288CTC (both in vivo and during perfusion in situ) supported a dose-dependent conversion (P < 0.001) of plasma linoleic acid to the mitogen, 13-hydroxy-9,11-octadecadienoic acid. We conclude that increased arterial blood plasma linoleic acid concentrations, caused by increased dietary intakes, specifically stimulate growth, lipid storage and linoleic acid metabolism in hepatoma 7288CTC in vivo.
KEY WORDS: linoleic acid · hepatoma 7288CTC · tumor growth · tumor lipid content · 13-hydroxyoctadecadienoic acid · ratsINTRODUCTION
Dietary fat has an important role in the growth of spontaneous, chemically induced and transplantable tumors in rodents (Klurfeld 1995
). High fat diets, in particular those containing linoleic acid as a major fatty acid, were shown to increase the growth rates of established rodent tumors (Abraham and Hillyard 1983, Hillyard and Abraham 1979) and human tumor xenografts in nude mice (Rose et al. 1993, Wang et al. 1995). Diets supplemented with pure arachidonic acid were ineffective (Hillyard and Abraham 1979). Increased dietary linoleic acid also appeared to increase the weights of chemically induced mammary tumors in 25% energy-restricted rats (Klurfeld et al. 1989). The stimulative effects of dietary linoleic acid on tumor growth rate (Rao and Abraham 1976) and chemically induced carcinogenesis (Ip et al. 1985, Roebuck et al. 1985) were saturable processes. These studies defined a unique tumor growth requirement for linoleic acid in vivo and suggested that the fatty acid itself or a metabolite initiated or enhanced specific tumor growth processes. The mechanism of tumor growth stimulation by dietary linoleic acid is not yet known.
Previous studies performed in this laboratory showed that increased plasma concentrations of linoleic and arachidonic acids were responsible for the increase in tumor growth rate observed in vivo during an acute fast (Sauer et al. 1986). Solid tumors in vivo removed linoleic and arachidonic acids from the arterial blood in concentration- and supply-dependent processes (Sauer and Dauchy 1992a). Also, incorporation of 3H-thymidine into DNA of hepatoma 7288CTC perfused in situ was directly dependent on the ambient arterial blood concentrations and uptakes of linoleic and arachidonic acids (Sauer and Dauchy 1988 and 1992b), suggesting that the fatty acids were the active agents that interacted with the tumor. Rates of 3H-thymidine incorporation were about three times faster with linoleic acid than with arachidonic acid; oleic and palmitic acids were not active (Sauer and Dauchy 1992b). We proposed that increased concentrations of linoleic acid in arterial blood plasma, resulting from increased linoleic acid ingestion, have a direct stimulative effect on tumor growth and linoleic acid metabolism in vivo (Sauer and Dauchy 1988). The experiments performed in this study were designed to test this hypothesis.
0.678, P = 0.009, r = 0.991, respectively. Correlations between plasma linoleic and arachidonic acid concentrations and plasma EFA in rat groups given free access to food, inset A, were as follows: plasma linoleic acid concentration = 0.645 (plasma EFA) 0.402, P < 0.001, r = 0.983 and plasma arachidonic acid concentration = 0.356 (plasma EFA) + 0.399, P < 0.001, r = 0.95. The slopes were different, P = 0.003, df = 10. Correlations between plasma linoleic and arachidonic acid concentrations and plasma EFA in energy-restricted rat groups, inset B, were as follows: plasma linoleic acid concentration = 0.931 (plasma EFA) 0.990, P = 0.001, r = 0.999 and plasma arachidonic acid concentration = 0.059 (plasma EFA) + 1.011, P = 0.251, r = 0.749. The slopes were different, P = 0.00007, df = 4. The x-axes for insets A and B are linoleic + arachidonic acids concentrations, mmol/L arterial blood plasma.
MATERIALS AND METHODS
). A brief description of the implantation procedure is as follows: the skin over the lower abdomen and groin of a rat anesthetized with sodium pentobarbital (25 mg/kg, Abbot Laboratories, North Chicago, IL) was shaved and cleaned with Betadine solution (Purdue Frederick, Norwalk, CT). An incision was made to expose the vessels in the left inguinal region. The deep branches of the femoral vessels and the femoral vein distal to the origin of the superficial epigastric vein were ligated. The superficial epigastric artery and vein were cleaned of adherent fat and ligated about 2 cm distal to their origin from the femoral vessels to form a vascular stalk; a 3-mm cube of tumor tissue was sutured to the end of this vascular stalk. The tumor implant and adjacent stalk were enclosed in a small Parafilm envelope (American National Can, Greenwich, CT), placed in the inguinal fossa and the skin incision was closed. Vascular connections to the epigastric vessels form within the Parafilm envelope during tumor-initiated angiogenesis. These tumors typically lack a central necrotic region seen in subcutaneous implants that obtain their blood supply from the periphery (Sauer et al. 1982). The latent period between implantation and first evidence of a subcutaneous tumor was recorded, and the weight of the established tumor was estimated every other day over a 10- to 18-d period (Sauer et al. 1986). Growth rates of individual tumors (g/d) represent the slopes of linear regression lines developed from these data. The actual tumor weight (measured on the day of harvest) was the final point in the regression. Tumor growth rates for a group are means ± SEM of the rates determined for individual rats.
and 1992b). [14C(U)]-Linoleic acid (38.7 GBq/mmol) was purchased from NEN Products, Boston, MA and [1-14C]-arachidonic acid (2.04 GBq/mmol) was purchased from American Radiolabeled Chemicals, St. Louis, MO. Tumors for perfusion weighed 4-6 g and were collected from Buffalo rats fed either the 5% corn oil diet or nonpurified diet. Donor blood was collected at 0800-1000 h from adult male Sprague-Dawley rats (250-300 g) fed nonpurified diet. In some experiments, the lipid concentration of the donor blood was increased by depriving the donor rats of food for 24 h. All procedures were approved by the Institutional Animal Care and Use Committee.
20°C. Rats with free access to food were housed two rats per cage, whereas the energy-restricted rats were housed singly. The experimental diets were fed in wide-mouthed glass jars with stainless steel screw tops with a central opening that restrained the animals from pulling food from the jars. Energy-restricted rats received an amount of diet equal to 70% of the mean amount of food eaten the day before by cohort rats given free access to food. All rats were fed at 1500 h and uneaten rations were discarded. Animal weights were recorded weekly.
|
Table 1. Composition of the experimental diets |
|
Table 2. Fatty acid composition of the experimental diets |
20°C. Tumors were implanted 3 d after the final blood collection.
Lipid extraction and analysis.
Plasma free and total plasma, tumor and dietary fatty acids were extracted as previously described (Sauer and Dauchy 1988 and 1992a). All tumors were carefully cleaned of external fat on removal from the rat; microscopic examination showed that adipose tissue was not present in the tumor interior (Sauer et al. 1982). The portions of sample extracted were as follows: 0.1-0.2 mL plasma, 0.1 mL of a 20% homogenate of tumor tissue and 25 mg diet. Heptadecanoic acid (100 µg) was added as an internal standard before extraction. The major lipid classes in plasma and tumor were separated by TLC, eluted from the resin, saponified and the fatty acids methylated. Internal standards for each lipid class were added to the plasma and tumor homogenates before extraction and TLC. Lipid extracts were saponified in methanolic-NaOH (0.5 mmol/L) for 5 min at 100°C and the fatty acids methylated using boron triflouride-methanol reagent for 2 min at 100°C. Fatty acid methyl esters were measured using a Hewlett-Packard (Palo Alto, CA) model 5890A gas chromatograph equipped with a flame-ionization detector, an integrator (model 3396A) and autoinjector (model 7673S). Separations were performed on a 0.25 mm × 30 m capillary column (model 2330, Supelco Inc., Bellefonte, PA) at 190°C with helium as the carrier gas (linear gas rate, 19 cm/s). Fatty acid methyl esters were identified by their retention times compared with known standards. Plasma and tumor fatty acid concentrations were expressed as mmol/L plasma or mmol/g tumor wet weight, respectively. Plasma free fatty acid concentrations were 20-25% of the total concentration.
). Tracer amounts of 3H-oleic acid, [9,10-3H(N)]-triolein (931.6 GBq/mmol) and [2-palmitoyl-9,10-3H(N)]-L-
-dipalmitoylphosphatidylcholine (1.2 TBq/mmol), purchased from NEN, were added before extraction for recovery estimation. Radiolabeled compounds were measured by liquid scintillation; counting efficiency was determined by internal standardization.
). Relationships between tumor growth rates and plasma fatty acid levels for the dietary groups within and between the dietary regimens were compared by linear regression. P-values < 0.05 were considered significant. Statistical tests were performed using SigmaStat (Jandel Scientific, San Rafael, CA) and True Epistat (True Epistat Services, Richardson, TX) software packages.
RESULTS
|
Table 3. Number of rats per group, period of dietary treatment, rat carcass weights, daily diet consumption per rat, day of tumor implantation, tumor growth rates and final tumor weights for the dietary groups |
Fig. 3.
The effect of mean daily arterial blood plasma concentrations of (A) palmitic + stearic, (B) linoleic, (C) arachidonic and (D) oleic acids on mean tumor growth rates in rats given free access to food or energy-restricted. Symbols are the same as in Figure 2. Regression lines and 95% confidence intervals indicate significant correlations. In (B), mean plasma linoleic acid concentrations were different (P < 0.05) for all dietary group pairs given free access to food, except groups fed the 7.5% corn oil (CO) and 5% CO + 5% olive oil (OV) diets and the 10% OV and 2% CO diets. Mean plasma linoleic acid concentrations in energy-restricted groups were different (P < 0.05), except for groups fed 10% CO and 7.5% CO diets. In (B), for rats given free access to food, tumor growth rate = 0.425 (plasma linoleic acid concentration) + 0.697, P < 0.001, r = 0.956 and for energy-restricted rats, tumor growth rate = 0.812 (plasma linoleic acid concentration) 0.172, P = 0.008, r = 0.992. Slopes were not different, P = 0.317; intercepts were different, P = 0.000013. In (C), for rat groups given free access to food, tumor growth rate = 0.605 (plasma arachidonic acid concentration) + 0.482, P = 0.04, r = 0.776.
[View Larger Version of this Image (27K GIF file)]
0.172, respectively, were significantly different (P = 0.001); relative to tumor growth in rats given free access to food, energy restriction caused a decrement in tumor growth equivalent to about 0.9 g wet weight/d.
|
Table 4. Effect of dietary fat level and feeding regimen of rats on the linoleic acid and total fatty acid concentrations of tumor total lipids and triacylglycerols |
Table 5.
Incorporation of plasma free 14C-linoleic and 14C-arachidonic acids into 13-HODE and 13-KODE in tumor venous blood and tumor lipids during perfusion of hepatoma 7288CTC in situ
). 13-HODE and 13-KODE were released into the tumor venous blood during perfusion with 14C-arachidonic acid but were not radiolabeled. No 12- or 15-HETE, either radiolabeled or nonradiolabeled, was detected in tumor venous blood from either experiment. A portion of the radiolabeled linoleic and arachidonic acids removed from the arterial blood was incorporated into tumor lipids. 14C-Linoleic acid was incorporated into tumor triacylglycerols and phospholipids; however, 14C-arachidonic acid was incorporated mainly into phospholipids and cholesterol esters.
Fig. 4.
Correlation between the rates of release of 13-hydroxy-9,11-octadecadienoic acid (13-HODE) into the tumor venous blood and plasma linoleic acid uptake by hepatoma 7288CTC in vivo and during perfusion in situ. Each point represents a single measurement: rate of tumor 13-HODE release = 0.035 (rate of tumor linoleic acid uptake) + 0.038, P < 0.001, r = 0.956, n = 54.
[View Larger Version of this Image (24K GIF file)]
DISCUSSION
, Hirose et al. 1990, Sundram et al. 1989), but in those experiments single blood samples were drawn, the times of blood sampling during the diurnal feeding cycle were not designated and the fatty acid compositions were not quantified. Quantitative data described here show changes in plasma fatty acid concentrations and tumor growth rates associated with ingestion of seven semipurified diets containing different corn and olive oil concentrations. The central findings were that tumor growth, lipid storage and rate of 13-HODE formation were dependent on dietary linoleic acid intake and the plasma linoleic acid concentration. Growth of hepatoma 7288CTC became limited at plasma linoleic acid concentrations 
3 mmol/L.
), it is not known which lipid class is most important in supplying the linoleic acid-dependent growth requirement. Some evidence suggests that free linoleic acid performs this function. We have previously shown that 0.6-0.7 mmol/L plasma free linoleic acid supported peak rates of 3H-thymidine incorporation in hepatoma 7288CTC perfused in situ (Sauer and Dauchy 1992b). A similar plasma free linoleic acid concentration (0.78 ± 0.15 mmol/L) was present in arterial blood of the rats given free access to the 10% CO diet. Also, free linoleic acid has a positive growth effect when added to tumor cells cultured in serum-free media. If 13-HODE production is necessary for the linoleic acid-dependent growth requirement, the discrepancy between the specific activities of 14C-13-HODE in tumor venous blood and plasma free 14C-linoleic acid in arterial blood suggests that unlabeled linoleic acid sources, e.g., triacylglycerols, phospholipids and/or cholesterol ester, contributed to 13-HODE production. However, the 14C-13-HODE specific activity may have been underestimated. Because of the low concentrations of 13-HODE present in tumor venous blood, the determinations were made using pooled plasma samples. This may have obscured increases in specific activity that occurred during the perfusion. Also, because no adequate tracer was available, we were unable to correct for 13-HODE losses during the extraction and chromatographic procedures. The results in Table 5 clearly show that plasma free linoleic acid was converted to 13-HODE, but other plasma linoleic acid sources cannot yet be ruled out.
). Plasma free arachidonic acid concentrations (generally about 25% of the total concentration) equaled or exceeded these values in all of the rat groups examined here except rats fed the fat-free diet (Fig. 3, 4). Hillyard and Abraham (1979) demonstrated a linoleic acid growth requirement in mammary adenocarcinoma growing in BALB/c mice given free access to a fat-free diet but found no arachidonic acid growth requirement. Preferential conversion of linoleic acid to arachidonic acid in rats fed low linoleic acid concentrations (Fig. 2) may provide sufficient plasma arachidonic acid to satisfy the arachidonic acid-dependent growth requirement. Evidence for this conversion was also observed in normal suckling (Iritani et al. 1993) and adult rats (Nelson et al. 1987) fed low linoleic acid concentrations. Unless animals are EFA-deficient, the specific tumor arachidonic acid-dependent growth processes are probably saturated.
, Keler and Sorof 1993, Wicha et al. 1979) and human breast carcinoma cells (Rose and Connolly 1990, Shultz et al. 1992) grown in fatty acid-deficient media showed a positive growth response to added linoleic acid. Saturated fatty acids were either without effect (Holley et al. 1974) or were inhibitory (Wicha et al. 1979). However, equivalent positive growth responses were also observed after addition of either arachidonic (Holley et al. 1974, Keler and Sorof 1993) or oleic acid (Kasayama et al. 1994, Rose and Connolly 1990, Wicha et al. 1979) to the culture medium. The in vitro requirement for linoleic, arachidonic or oleic acid is most convincingly demonstrated after preculture in serum-free or low serum media, or in delipidized serum (Holley et al. 1974, Kasayama et al. 1994, Keler and Soroff 1993, Rose and Connolly 1990, Wicha et al. 1979). This acute fatty acid deprivation may reveal requirements that do not exist in vivo. For example, oleic acid, one of the most abundant fatty acids in rodent plasma, is unlikely to ever be limiting in vivo. Perhaps, as suggested by Kasayama et al. (1994), the in vitro growth responses to linoleic, oleic and arachidonic acids in fatty acid-deprived cell lines represent a nonspecific requirement for unsaturated fatty acids.
, Klurfeld 1995, Kritchevsky 1990, Welsch 1994). The exact cause of the tumor growth inhibition is not yet known. In Buffalo rats, energy restriction reduced hepatoma 7288CTC growth and plasma EFA and linoleic acid concentrations to narrow ranges (Fig. 1-3). Interestingly, the slope of the regression lines for tumor growth on plasma linoleic acid concentrations in energy-restricted rats was identical to that for rats given free access to food. An increment in plasma linoleic acid concentration caused an equivalent increment in tumor growth in rats on both feeding regimens. The intercepts of the regression lines on the y-axis were different (Fig. 3B), however, indicating that energy restriction removed a tumor growth stimulative effect that was expressed in rats given free access to food. To our knowledge, these data are the first evidence showing that decreased plasma linoleic acid concentrations and decreased energy intakes may contribute separately to reduce tumor growth rates during energy restriction.
). Inhibitors of cyclooxygenase and lipoxygenase are known to inhibit carcinogen-induced tumorigenesis and tumor growth promoted by high fat diets. We investigated 13-HODE production in hepatoma 7288CTC in vivo because of the strong linoleic acid growth requirement and the weak or absent arachidonic acid requirement. 13-HODE was released into the tumor venous blood at substantial rates in vivo and during perfusion in situ (Fig. 4), and rates of release were directly dependent on the plasma linoleic acid concentrations. Radiolabeled 13-HODE and 13-KODE were formed from plasma free 14C-linoleic acid. Factors that blocked 13-HODE production inhibited tumor growth. For example, the inhibition of linoleic acid uptake in hepatoma 7288CTC by physiological amounts of melatonin inhibited 13-HODE release and tumor growth (Sauer et al. 1996). [Also, recent unpublished results (Sauer, L.A., Blask, D.E. and Dauchy, R.T.) show that the lipoxygenase inhibitor, nordihydroguaiaretic acid, inhibited tumor growth and 13-HODE release without affecting fatty acid uptake.] Thus, these data support and extend the hypothesis that 13-HODE is the critical substance, generated by the tumor from plasma linoleic acid, that links increased linoleic acid ingestion to increased tumor growth.
ACKNOWLEDGEMENTS
FOOTNOTES
Manuscript received 1 July 1996. Initial reviews completed 9 August 1996. Revision accepted 17 March 1997.
LITERATURE CITED
This article has been cited by other articles:
|
|
 |
|
  L. A. Sauer, R. T. Dauchy, D. E. Blask, J. A. Krause, L. K. Davidson, E. M. Dauchy, K. J. Welham, and K. Coupland Conjugated Linoleic Acid Isomers and Trans Fatty Acids Inhibit Fatty Acid Transport in Hepatoma 7288CTC and Inguinal Fat Pads in Buffalo Rats J. Nutr., August 1, 2004; 134(8): 1989 - 1997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Blask, R. T. Dauchy, L. A. Sauer, and J. A. Krause Melatonin uptake and growth prevention in rat hepatoma 7288CTC in response to dietary melatonin: melatonin receptor-mediated inhibition of tumor linoleic acid metabolism to the growth signaling molecule 13-hydroxyoctadecadienoic acid and the potential role of phytomelatonin Carcinogenesis, June 1, 2004; 25(6): 951 - 960. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Sauer, R. T. Dauchy, and D. E. Blask Mechanism for the Antitumor and Anticachectic Effects of n-3 Fatty Acids Cancer Res., September 1, 2000; 60(18): 5289 - 5295. [Abstract] [Full Text]
|
|
|
|
|
|
L. A. Sauer, R. T. Dauchy, D. E. Blask, B. J. Armstrong, and S. Scalici 13-Hydroxyoctadecadienoic Acid Is the Mitogenic Signal for Linoleic Acid-dependent Growth in Rat Hepatoma 7288CTC in Vivo Cancer Res., September 1, 1999; 59(18): 4688 - 4692. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Blask, L. A. Sauer, R. T. Dauchy, E. W. Holowachuk, M. S. Ruhoff, and H. S. Kopff Melatonin Inhibition of Cancer Growth in Vivo Involves Suppression of Tumor Fatty Acid Metabolism via Melatonin Receptor-mediated Signal Transduction Events Cancer Res., September 1, 1999; 59(18): 4693 - 4701. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
