The Journal of Nutrition Vol. 128 No. 8 August 1998, pp. 1257-1261

Docosahexaenoic Acid Increases Thyroid-Stimulating Hormone Concentration in Male and Adrenal Corticotrophic Hormone Concentration in Female Weanling Rats^1,2

M. Thomas Clandinin³, Donna L. Claerhout, and Eric L. Lien^*

Nutrition and Metabolism Research Group, Departments of Agricultural, Food and Nutritional Science and Medicine, University of Alberta, Edmonton, Alberta, Canada, T6G 2P5 and ^* Wyeth Nutrition International, Radnor, PA

	ABSTRACT

Abstract Introduction Methods Results & Discussion References

Circulating levels of thyroid-stimulating hormone (TSH), growth hormone (GH), adrenal corticotrophic hormone (ACTH) and prolactin (PRL) were assessed in suckling rats in the postweaning period after rats were fed diets that reflect the fat composition of a current infant formula with or without the addition of 1.2 g/100 g fatty acid arachidonic acid [20:4(n-6)] and 0.7 g/100 g fatty acid docosahexaenoic acid [22:6(n-3)] or both 20:4(n-6) and 22:6(n-3). At 2 wk of age, no effect of diet on circulating levels of TSH, ACTH, GH or PRL was apparent. By 6 wk of age (3 wk postweaning), male rats consuming the diet containing 22:6(n-3) had significantly elevated levels of TSH, and females had significantly higher ACTH concentrations than males. No effect of diet was observed on circulating GH or PRL levels. Male pups had higher levels of TSH than females (P < 0.0001), whereas female pups from the 22:6(n-3) diet treatment exhibited much higher levels of ACTH than all male pups from any of the other diet treatments. These results suggest that metabolic controls, functioning through endocrine mechanisms, can be altered by changing the 20:4(n-6) to 22:6(n-3) balance in the diet.

	INTRODUCTION

Abstract Introduction Methods Results & Discussion References

On the basis of the levels of essential fatty acids present in human milk and their accretion rate in the fetus during the last trimester of gestation, we concluded that infant formulas should contain the longer-chain homologues of 18:2(n-6) and 18:3(n-3) (Clandinin et al. 1982). It was our conviction that both 20:4(n-6) and 22:6(n-3) should be present in formulas in proportions that reflected the overall fatty acid balance found in the fatty acid mixture of human milk (Clandinin et al. 1989). We further suggested that supplementation of feedings with only very long-chain (n-3) fatty acids (>18C) and 18:2(n-6), although easily achieved, would not be as desirable because of predictable effects on the reduction in synthesis, metabolism and incorporation of 20:4(n-6) into membrane phospholipids. Feeding human milk results in increased content of 20:4(n-6) and 22:6(n-3) in erythrocyte phosphatidylethanolamine and phosphatidylcholine in term infants compared with similar membrane fractions for infants fed formula containing 18:2(n-6) and 18:3(n-3) but not their longer-chain homologues (Putnam et al. 1982). Moreover, when the formula has been modified to contain the longer-chain homologues of 18:2(n-6) and 18:3(n-3), the levels of 20:4(n-6) and 22:6(n-3) in plasma phospholipid and cholesterol ester are similar to those found in infants fed human milk (Clandinin et al. 1992, Koletzko et al. 1987). In studies in which the feedings have been supplemented with very long-chain (n-3) fatty acids alone, there is a clear increase in the incorporation of these fatty acids into membrane lipids in the developing infant (Carlson et al. 1986 and 1987) with a concomitant decrease in 20:4(n-6) levels.

Docosahexaenoic acid (DHA)⁴ is necessary for normal neural and visual development (Neuringer et al. 1988). When preterm infants were fed formula fortified with marine oil, their 22:6(n-3) status improved, and the rate of development of early visual acuity was accelerated (Carlson et al. 1989). However, 20:4(n-6) levels in phospholipids declined and the infants supplemented with only marine oil grew less well than those fed standard preterm formula (Carlson et al. 1992a, 1992b, 1992c, 1993 and 1996). When the formula contained both 20:4(n-6) and 22:6(n-3), no apparent differences in growth were observed (Clandinin et al. 1992 and 1997b, Koletzko et al. 1995). However, it should be noted that an insufficient number of infants were studied to test for small effects on growth.

Feeding lactating rats diets with or without preformed 20:4(n-6) and 22:6(n-3) markedly alters the phospholipid fatty acid composition in the developing brain of the rat pup (Jumpsen et al. 1995, 1996 and 1997). These changes in membrane composition vary between brain regions and the timing in development at which they occur, and are different when neuronal and glial cell types are compared (Jumpsen and Clandinin 1995).

The anterior hypophysis releases a variety of hormones, among them thyroid-stimulating hormone (TSH), adrenal corticotrophic hormone (ACTH), growth hormone (GH) and prolactin (PRL), that have a pervasive effect on the growth, differentiation and function of target organs throughout the body. Indeed, GH is synergistic with ACTH in increasing adrenal size, affecting reproductive organ development with androgens and increasing body protein content. Neuroregulation of the anterior pituitary arises from the portal hypophyseal vessels. In our previous studies of suckling rat pups (Jumpsen et al. 1996 and 1997), it was not feasible to dissect the pituitary and measure membrane changes that may have occurred in various cell types in response to feeding 20:4(n-6) or 22:6(n-3). However, it is logical to hypothesize that feeding these fatty acids may also affect the composition of neuronal membranes in the anterior and posterior regions of the pituitary. These types of diet-induced changes in membrane composition occurred in other brain regions. If analogous change in membrane composition occurs in the pituitary, it is reasonable to hypothesize that such changes in membrane structure will affect the function of this endocrine organ and perhaps the levels of hormones secreted from the pituitary that affect growth at different stages of neural development.

View larger version (36K): [in this window] [in a new window]   Fig 1. Plasma thyroid-stimulating hormone (TSH) concentrations in rats fed diets with or without arachidonic acid (AA), docosahexaenoic acid (DHA) or AA + DHA. At 6 wk of age, significant effects of diet treatment are indicated for male rats by different superscripts (a, b, c) (P < 0.05). Male vs. female, P < 0.01. Blood was drawn from rat pups at the time periods indicated. Values are means ± SD, n = 10.

View larger version (29K): [in this window] [in a new window]   Fig 2. Plasma adrenal corticotrophic hormone (ACTH) levels in rats fed diets with or without arachidonic acid (AA), docosahexaenoic acid (DHA) or AA + DHA. There were no significant effects of treatment at 2 wk of age. At 6 wk of age, significant effects of diet treatment are indicated for females by different superscripts (a, b). Blood was drawn from rat pups at the time periods indicated. Values are means ± SD, n = 10.

This study was designed to assess the circulating levels of hormones released from the anterior pituitary (TSH, GH, ACTH and PRL) in suckling rats and young rats in the postweaning period after they were fed diet fats that reflect the fat composition of a current infant formula with or without the addition of 20:4(n-6) and 22:6(n-3).

	MATERIALS AND METHODS

Abstract Introduction Methods Results & Discussion References

Animal breeding.  This research protocol was reviewed by the University Animal Care Committee and complies with the NIH guidelines (NRC 1985). Sprague-Dawley rats were bred in the laboratory for the study. All rats were housed in plastic cages fitted with stainless steel mesh covers in a temperature- and humidity-controlled room with a 12-h light:dark cycle. As soon as the pregnant dams delivered, they were given free access to one of the test diets. Litter size was equalized at 12 pups and assignment of dams and litters to treatment was random. A minimum of five litters were assigned to each diet treatment for each value illustrated at each age. Litters were used at 2 wk of age for blood collection and subsequent hormone analyses. Other litters were suckled until 3 wk of age, at which time the pups were weaned and given free access to the same diet as their mother. These litters were used at 6 wk of age for blood collection and hormone analyses.

Diets.  The semipurified diets fed contained 20 g/100 g total fat (Table 1) (Clandinin and Yamashiro 1982). The diet fat mixture was formulated to reflect the fatty acid composition of a current infant formula fat blend, the control blend (SMA, Wyeth-Ayerst Laboratories, Radnor, PA) or modified to provide levels of 20:4(n-6) or 22:6(n-3) that could be incorporated into a potentially new infant formula fat blend [arachidonic acid (AA), DHA or AA + DHA; Table 2], reflecting the content of these fatty acids in human milk. Both 20:4(n-6) and 22:6(n-3) were provided in the fat blend as a triglyceride.

Blood collection.  All blood samples were rapidly collected (i.e., within 30 s) by cardiac puncture in tubes after brief exposure to halothane at the same time of the morning. Blood was then centrifuged at 4°C (1855 × g for 10 min). The serum-plasma was removed, transferred to a microcentrifuge tube, immediately placed on dry ice and stored at 70°C under N₂ until analyzed. Samples at 2 and 6 wk of age were analyzed in a batch a few days after sample collection.

Hormone analysis.  TSH, GH, ACTH and PRL were analyzed by RIA procedures (TSH and PRL, Ciba Corning Canada, Markham, Canada; GH, Diagnostics Products, Los Angeles, CA; ACTH, Nichols Institute Diagnostics, San Juan Capistrano, CA). All hormones were analyzed for each rat. These assays were tested with rodent hormone as described in the following.

TSH was measured by using a Ciba-Corning TSH diagnostic kit (Ciba Corning Canada). The Ciba Corning TSH assay is a two-site chemiluminometric immunoassay, which uses constant amounts of two antibodies. The first antibody or Lite Reagent is a monoclonal mouse anti-TSH antibody labeled with an acridinium ester. The second antibody or Solid Phase is a polyclonal sheep anti-TSH antibody covalently coupled to a paramagnetic particle. A direct relationship exists between the TSH in a sample and the relative light units detected by the ACO:180 analyzer. The TSH assay measures TSH concentrations up to 150 mIU/L with a minimum detectable concentration of 0.03 mIU/L. The cross-reactivity of the TSH assay with luteinizing hormone (LH), follicle-stimulating hormone (FSH) and human chorionic gonadotropin (hCG) has been determined by adding these hormones to serum controls containing TSH. The level of TSH of the serum was not affected.

Rat serum PRL level was assayed using a Ciba Corning PRL kit (Ciba Corning Canada). PRL assay is a two-site chemiluminometric (sandwich) immunoassay, similar to the TSH assay. A direct relationship exists between the PRL in the serum sample and relative light units (RLµs) detected by the ACS:180 system. The prolactin assay measures prolactin concentrations up to 200 µg/L with a minimum detectable concentration of 0.3 µg/L. The cross-reactivity of the PRL assay with TSH, LH, hCG, FSH, human growth hormone (hGH) and human placental lactogen (hPL) was determined by adding these hormones to serum samples containing PRL. The presence of TSH, LG, hCG and FSH slightly increased the PRL level in the serum samples but hGH and hPL had no effect.

The GH level was measured by a double antibody hGH diagnostic kit, manufactured by Diagnostic Products. In the double antibody hGH procedure, ¹²⁵I-labeled GH competes with GH in serum or plasma for sites on the GH specific antibody. After incubation for a fixed time, separation of bound from free is achieved by the PEG-accelerated double-antibody method. The tube is then counted in a gamma counter; the counts are inversely related to the amount of GH present in the sample. By the more conservative definition, the GH assay has a detection limit of ~0.9 µg/L using the standard 2-h room-temperature procedure. The antiserum is highly specific for GH with the exception of PRL, which has a molecular structure similar to that of GH. The antiserum exhibits low cross-reactivity to other hormones.

View larger version (26K): [in this window] [in a new window]   Fig 3. Plasma growth hormone concentrations in rats fed diets with or without arachidonic acid (AA), docosahexaenoic acid (DHA) or AA + DHA. These values are just above detection limits. There were no significant effects of treatment. Blood was drawn from rat pups at the time periods indicated. Values are means ± SD, n = 10.

View larger version (32K): [in this window] [in a new window]   Fig 4. Plasma prolactin concentrations in rats fed diets with or without arachidonic acid (AA), docosahexaenoic acid (DHA) or AA + DHA. There were no significant effects of treatment. Blood was drawn from rat pups at the time periods indicated. Values are means ± SD, n = 10.

For the quantitative determination of ACTH in plasma, an ACTH 65T kit (Nichols Institute Diagnostics) was used. The ACTH immunoassay incorporates a monoclonal antibody and a polyclonal antibody, both with high affinity and specificity for defined amino acid regions of the ACTH molecule. The monoclonal antibody is radialabeled for detection, whereas the polyclonal antibody is coupled to biotin. The addition to the reaction mixture of an avidin-coated plastic bead allows for a specific and efficient means of binding the sandwich complex via the high affinity interactin between biotin and avidin. In the assay system, standards, controls and samples are incubated with a solution containing both the radialabeled antibody and biotin-coupled antibody, and an avidin-coated plastic bead. At the end of the assay incubation, the bead is washed to remove unbound components, and the radioactivity bound to the solid phase is measured in the gamma counter. A dose-response curve of radioactivity vs. concentration is generated by using results obtained from standards that are assayed concurrently with the unknowns. Concentrations of intact ACTH in the controls and samples are determined directly from the curve. The highest concentration of ACTH measurable without dilution is the value of the highest standard (20,000 ng/L) and the lowest measurable concentration is 1.0 ng/L.

Statistical analysis.  Each value illustrated is based on at least 10 replicates from a minimum of five different litters. One rat was considered a replicate. The effect of treatment was determined by three-way ANOVA procedures followed by a Duncan's multiple range test (Steel and Torrie 1980) when a significant effect of diet was observed. The sum of squares used was the error mean square.

	RESULTS AND DISCUSSION

Abstract Introduction Methods Results & Discussion References

Analyses of the stomach contents of the suckling rat pups at 2 wk of age indicated that a substantial amount of the 20:4(n-6) and 22:6(n-3) added to the dam's diet was transferred into the milk (data not shown). Addition of 1.2 g/100 g 20:4(n-6) or 0.7 g/100 g 22:6(n-3) increased the level of these fatty acids in the milk approximately two- to threefold and three- to sixfold, respectively (Jumpsen and Clandinin 1996). All groups exhibited similar weight gains (data not illustrated).

At 2 wk of age, no effect of diet on circulating levels of TSH (Fig. 1), ACTH (Fig. 2), GH (Fig. 3) or PRL (Fig. 4) was apparent. By 3 wk postweaning, 6-wk-old male pups consuming the diet containing 22:6(n-3) had significantly elevated levels of TSH and female pups had greater levels of ACTH. No effect of diet was observed on circulating GH or PRL levels. Male pups had higher levels of TSH than females (P < 0.0001; Fig. 1), whereas female pups from the 22:6 (n-3) diet treatment had significantly higher concentrations of ACTH than all male pups.

Growth is a function of many factors that can be limited by restricting the availability of essential nutrients and by metabolic interactions that may also serve to limit the availability of an essential precursor. Both 20:4(n-6) and 22:6 (n-3) are essential components of the membrane structural lipids (Clandinin et al. 1991 and 1994). It is therefore logical that a decrease in the status of one of these fatty acids, as measured by a variety of potential dynamic metabolic markers [for example, levels in plasma phospholipid pools, erythrocyte phosphatidylcholine, plasma triglycerides (Carlson et al. 1992c and 1993, Koletzko and Braun 1991)], may be reflected in a decrease in cell growth in some tissues. At present, however, it is hard to envisage how a change in these metabolic markers will be reflected in changes in gross body weight or the composition of growth. Moreover, it is not known which tissues or cell types are the first to be affected when essential fatty acid status declines to a critical or limiting level.

If similar changes occur in pituitary function in the human infant, the results of this study are suggestive of an entirely different basis for the differences in growth observed after varying intakes of (n-6) and (n-3) fatty acids in infant feedings. Is it reasonable to speculate that infants supplemented with fish oil containing 20:5(n-3) and 22:6(n-3) may have elevated TSH levels? It is also noteworthy that small preterm infants have a fragile capability to produce corticosteroids and are often given corticosteroids to stimulate phosphatidylcholine production for surfactant. Perhaps this need for corticosteroid is also a function of the infant's essential fatty acid status affecting ACTH release. The observations of this study, although not conclusive in terms of demonstrating a clear effect of diet on preweaning to postweaning growth, are suggestive of an effect of early diet on developing neuroendocrine regulatory mechanisms.

	FOOTNOTES

Manuscript received 3 June 1997. Initial reviews completed 11 July 1997. Revision accepted 19 March 1998.

	LITERATURE CITED

Abstract Introduction Methods Results & Discussion References

• Carlson S. E., Cooke R. J., Peeples J. M. Docosahexaenoate (DHA) and eicosapentaenoate (EPA) status of preterm infants: relationship to visual acuity in n-3 supplemented and unsupplemented infants. Pediatr. Res. 1989; 25:285A
• Carlson S. E., Cooke R. J., Rhodes P. G., Peeples J. M., Werkman S. H. Effect of vegetable and marine oils in preterm infants formulas on blood arachidonic and docosahexaenoic acids. J. Pediatr. 1992a; 120:S159-S167[Medline]
• Carlson S. E., Cooke R. J., Werkman S. H., Tolley E. A. First year growth of preterm infants fed standard compared to marine oil n-3 supplemented formula. Lipids 1992b; 27:901-907[Medline]
• Carlson S. E., Rhodes P. G., Ferguson M. G. Docosahexaenoic acid status of preterm infants at birth and following feeding with human milk or formula. Am. J. Clin. Nutr. 1986; 44:798-804[Abstract/Free Full Text]
• Carlson S. E., Rhodes P. G., Rao V. S., Goldgar D. E. Effect of fish oil supplementation on the n-3 fatty acid content of red blood cell membranes in preterm infants. Pediatr. Res. 1987; 21:507-510[Medline]
• Carlson S. E., Werkman S. H., Peeples J. M., Cooke R. J., Tolley E. A. Arachidonic acid status correlates with first year growth in preterm infants. Proc. Natl. Acad. Sci. U.S.A. 1993; 90:1073-1077[Abstract/Free Full Text]
• Carlson, S. E., Werkman, S. H., Peeples, J. M., Cooke, R. J., Tolley, E. A. & Wilson, W. M. (1992c) Growth and development of very low-birthweight infants in relation to n-3 and n-6 essential fatty acid status. In: Essential Fatty Acids and Eicosanoids (Sinclair, A. & Gibson, R., eds.), American Oil Chemists' Society, Champaign, IL.
• Carlson S. E., Werkmen S. H., Tolley E. A. Effect of long-chain n-3 fatty acid supplementation on visual acuity and growth of preterm infants with and without bronchopulmonary dysplasia. Am. J. Clin. Nutr. 1996; 65:687-697
• Clandinin M. T., Chappell J. E., Heim T. Do low birth weight infants require nutrition with chain elongation-desaturation products of essential fatty acids? Prog. Lipid Res. 1982; 20:901-904
• Clandinin M. T., Chappell J. E., Van Aerde J.E.E. Requirements of newborn infants for long-chain polyunsaturated fatty acids. Acta Paediatr. Scand. 1989; 351:63-71
• Clandinin M. T., Cheema S., Field C. J., Garg M. L., Venkatraman J., Clandinin T. R. Dietary fat: exogenous determination of membrane structure and cell function. FASEB J. 1991; 5:2761-2769[Abstract]
• Clandinin M. T., Foxwell A. F., Goh Y. K., Layne K., Jumpsen J. A. Omega-3 fatty acid intake results in a relationship between the fatty acid composition of LDL cholesterol ester and LDL cholesterol content in humans. Biochim. Biophys. Acta 1997a; 1346:247-252[Medline]
• Clandinin, M. T., Jumpsen, J. & Suh, M. (1994) Relationship between fatty acid accretion, membrane composition and biological functions. J. Pediatr. 125 (suppl.): S25-S32.
• Clandinin M. T., Parrott A., Van Aerde J. E., Hervada A., Lien E. Feeding preterm infants a formula containing C20 and C22 fatty acids simulates plasma phospholipid fatty acid composition of infants fed human milk. Early Hum. Dev. 1992; 31:41-51[Medline]
• Clandinin M. T., Van Aerde J. E., Parrott A., Field C. J., Euler A. R, Lien E. L. Assessment of the efficacious dose of arachidonic and docosahexaenoic acids in preterm infant formulas: fatty acid composition of erythrocyte membrane lipids. Pediatr. Res. 1997b; 42:819-825[Medline]
• Clandinin M. T., Yamashiro S. Dietary factors affecting the incidence of dietary fat-induced myocardial lesions. J. Nutr. 1982; 112:825-828
• Jumpsen, J. & Clandinin, M. T. (1995) Brain Development: Relationship to Dietary Lipid and Lipid Metabolism. AOCS Press, Champaign, IL.
• Jumpsen J. A., Lien E. L., Goh Y. K., Clandinin M. T. Small changes of dietary (n-6) and (n-3) fatty acid content ratio alter phosphatidylethanolamine and phosphatidylcholine fatty acid composition during development of neuronal and glial cells in rats. J. Nutr. 1997; 127:724-731[Abstract/Free Full Text]
• Koletzko B., Braun M. Arachidonic acid and early human growth: is there a relationship? Ann. Nutr. & Metab. 1991; 35:128-131[Medline]
• Koletzko B., Edenhofer S., Lipowsky G., Reinhardt D. Effects of a low birthweight infant formula containing human milk levels of docosahexaenoic and arachidonic acids. J. Pediatr. Gastroenterol. Nutr. 1995; 21:200-208[Medline]
• Koletzko B., Schmidt E., Bremer H. J., Haug M., Harzer G. Dietary long chain polyunsaturates for premature infants. J. Pediatr. Gastroenterol. Nutr. 1987; 6:997-998[Medline]
• National Research Council (1985) Guide for the Care and Use of Laboratory Animals. Publication No. 85-23 (rev.), National Institutes of Health, Bethesda, MD.
• Neuringer M., Anderson G. J., Connor W. E. The essentiality of n-3 fatty acids for the development and function of the retina and brain. Annu. Rev. Nutr. 1988; 8:517-541[Medline]
• Putnam J. C., Carlson S. E., DeVoe P. W., Barness L. A. The effect of variations in dietary fatty acids on the fatty acid composition of erythrocyte phosphatidylcholine and phosphatidylethanolamine in human infants. Am. J. Clin. Nutr. 1982; 36:106-114[Abstract/Free Full Text]
• Steel, R.G.D. & Torrie, J. H. (1960) Principles and Procedures of Statistics. McGraw Hill Book Co., Inc., Toronto, Canada.

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences