- Email: [email protected]
The effect of dietary iron deficiency on the fatty acid composition of plasma and erythrocyte membrane phospholipids in the rat
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
© Pearson ProfessionalLtd 1997
The effect of dietary iron deficiency on the fatty acid composition of plasma and erythrocyte m e m b r a n e phospholipids in the rat H. Y. Tichelaar, 1 C. M. Smuts, 1 R. Gross, 2 P. L. Jooste, 1 M. Faber, 1 A. J. S. B e n a d 4 ~ ~National Research Programme for Nutritional Intervention of the Medical Research Council, Tygerberg, 7505, South Africa 2SEAMEO-TROPMED-GTZ, Community Nutrition Training Programme, University of Indonesia, Jakarta, Indonesia
Severe iron deficiency was introduced in rats by feeding outbred male Wistar rats a purified diet that was either adequate or deficient in iron. The rats were weighed regularly over 4 weeks to monitor body weight differences, after which blood was drawn from a subsample to determine the haemoglobin concentrations and fatty acid composition of plasma total phospholipids and to measure the erythrocyte membrane phosphatidylcholine and phosphatidylethanolamine levels. Comparisons between dietary iron adequate (control) and dietary iron deficient (experimental) rats showed that the experimental rats had lower body weight and plasma total phospholipid linoleic acid levels typical of the symptoms of essential fatty acid deficiency. Erythrocyte membrane phosphatidylethanolamine arachidonic acid levels were increased (P < 0.05) with concomitant decreases in oleic acid (P < 0.01). Correlations between fatty acids and growth suggest that the mechanism whereby iron deficiency affects growth is in some way related to abnormal fatty acid shifts that disturb the delicate balance of essential fatty acids in membranes. Additional 0)3 and 0)6 fatty acids may be necessary to counteract the effect of iron deficiency in rats. Summary
INTRODUCTION
The risks of iron deficiency (ID) are many and include preterm delivery and low birth weight of infants. 1 ID also impairs mental and motor development in infants 2 and may have poor long-lasting developmental outcomes 3 that can be corrected by early treatment with ferrous sulphate. 4 Fast growing healthy infants may even develop ID and need exogenous iron supplementation after 6 months of age? Therefore ID and growth present major nutritional problems among young children in developing countriesS -8 The role of iron in the brain is varied and includes the incorporation into enzymes and proteins, affecting behavioural systems through peroxide reduction, amino acid metabolism and fatty acid desaturation. 9 Clinical evidence suggests that ID influences essential fatty acid (EFA) metabolism 6,1° by reducing the Received 12 March 1996 Accepted 23 August 1996 Correspondence to: H. Y. Tichelaar, Tel. +27 21 938-0409; Fax. +27 21 9380321; E-mail. [email protected]..
activity of the A9- and A6-desaturase enzymes. The effects of an EFA metabolite deficiency have been overcome by supplementing oils containing a 4:1 ratio of ~-linolenic acid (ALA; C18:3o3) to linoleic acid (LA; C18:2~06) to rats, resulting in improved learning performance) 1 The rat presents a suitable and sensitive model to study ID, which rarely reaches levels of severity in humans that can be achieved in laboratory rats. l°,~a In the present study we have investigated the effect of ID on the fatty acid composition of plasma and erythrocyte membranes (EMB) in the rat to elucidate the involvement of iron in fatty acid metabolism and, indirectly, its effect on growth. MATERIALS AND METHODS Animals
Outbred male Wistar rats (n = 150) weighing =59 g were weaned at 3 weeks of age (21 + 1 day). The weaned rats received a powder rat diet, AIN-76 TM Purified Diet, 13,14ad libitum (Table 1) during an adaptation period of 3 days. 229
230
H. Y. Tichelaar et al
Table I Composition of the AIN-76TM purified dieta Ingredient
Diet (%)
Casein Methionine Cornstarch Sucrose Glucose Cellulose-type fibre Sunflower oil AIN mineral mixture b AIN vitamin mixture ° Cholien-bicartate
20.00 0.30 21.66 21.66 21.66 5.00 5.00 3.50 1.00 0.20
aRats received 5 g/kg/day. bComposition of the mineral mixture g/kg: CaHPO4, 500; NaCI, 74; K3C6HsOT.H20,220; K2SO4, 52; MgO, 24; MnC%, 3.5; Ferric citrate, 6; ZnC%, 1.6; CuC%, 0.3; KIO3, 0.01; Na2SEO3.5H20,0.01; CrK(SO,)2.12H20, 0.55; sucrose, 1000. °Vitamin mixture per kg: thiamin-HCI, 600 mg; riboflavin, 600 mg; pyridoxine-HCI, 700 mg; nicotinic acid, 3 g; D-calcium pantothenate, 1.6 g; folic acid, 200 mg; D-biotin, 20 mg; cyanobalamin, 1 mg; retinyl palmitate (vitamin A) 400 000 IU; d/-c~-tocopherolacetate (vitamin E), 5000 IU; cholecalciferol (vitamin D3), 2.5 mg; menaquinone (vitamin K), 50 mg; sucrose, 1000 g.
Their individual food intakes were measured daily and their body weights on day 1 and day 3. Rats with extreme food intakes and body weights during the adaptation period were withdrawn from the study until 132 rats remained. They were then randomly allocated to two groups each consisting of 66 housed rats. This sample size provided two groups of rats with no mean body weight difference at baseline. The experimental protocol was approved by the ethics committees of the South African Medical Research Council and the Regional SEAMEO-TROPMED Center on Community Nutrition, Indonesia. Experimental design
The experiment consisted of two groups of rats. A control group (n = 66) was fed the powdered AIN-76 TM diet ad libitum. The experimental group of rats (n = 66) received the same diet as the control group ad libitum, but with iron (ferric citrate) omitted from the mineral mixture. Mineral-free distilled water was freely available to the rats throughout the adaptation and experimental phases. Rats were weighed at the beginning of the experimental phase and weekly thereafter for four consecutive weeks on an electronic scale. The food intake of the rats was determined individually on alternate days for the duration of the experimental phase. Rats were ranked within the groups according to their final body weights. Every fourth rat was included in a subsample of 14 rats per group. Rats were anaesthetized with Sagatal (0.1 ml per 100 g body weight injected peritoneally) prior to blood
collection. About 4 ml blood was taken directly from the heart with an EDTA (ethylenediaminetetracetic acid) flushed syringe and the blood sample was transferred to an EDTA tube. Haemoglobin concentrations were determined spectrophotometrically on the subsample (n = 28) using the metcyanohaemoglobin method in preprepared cuvettes and the Minilab photometer from Ames. Erythrocyte membranes were prepared by haemolyzing erythrocytes with different phosphate buffersY '16 Lipids were extracted from plasma and EMB with chloroform/ methanol (2:1 vol/vol), separated by thin-layer chromatography and analysed for fatty acid composition of plasma total phospholipid (TPL), EMB phosphatidylcholine (-PC) and EMB phosphatidylethanolamine (-PE) by gas-liquid chromatography. 6,1z Statistical analysis
The difference between the fatty acid results obtained from the control and experimental animals was analysed statistically using the Wilcoxon two-sample test. Spearman correlation coefficients were calculated for fatty acids versus growth, irrespective of dietary treatment used. RESULTS
Table 2 shows the effect of dietary ID on growth of male Wistar rats after 4 weeks. The percentage weight gain of the control group was 33% (182.9% versus 150.4%) more than that of the experimental group. Control animals also had significantly higher (P < 0.0001) haemoglobin concentrations (132 + 3 g/L) than the experimental animals (86 + 8 g/L). For comparisons, the fatty acids of only nine control and 12 experimental rats could be used as the other five control and two experimental rats had either insufficient plasma or EMB for fatty acid analyses. The effect of dietary ID on plasma TPL, EMB-PC and EMB-PE fatty acid compositions of the rats is shown in Table 3. The fatty acids display significant differences between the two groups of rats after 4 weeks on their respective diets. In the plasma TPL fraction, stearic acid (C18:0) was higher (P < 0.05) in the experimental group. This was at the expense of LA, that was lower (P< 0.01), although 7-1inolenic acid (GLA; C18:3(06) and dihomoy-linolenic acid (DGLA; C20:3(06) were higher (P < 0.01). The (03 fatty acids, ALA and docosahexaenoic acid (DHA; C22:6(03) were respectively higher (P < 0.01 and P < 0.001). In the EMB-PC fraction, palmitic acid was lower (P < 0.05) in the experimental group. LA and docosatetraenoic acid (C22:4(06) were respectively higher (P< 0.05 and P < 0.01). Only the (03 fatty acid, ALA, was lower
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
© Pearson Professional Ltd 1997
Dietary iron deficiency and composition of rat membrane phospholipids
231
Table 2 Effect of dietary iron deficiency on body weight, weight gain and haemoglobin concentration compared with an adequate control diet
Baseline body weight (g) Body weight after 1 week (g) Body weight after 2 weeks (g) Body weight after 3 weeks (g) Body weight after 4 weeks (g) Weight gained (g) Haemoglobin (g/I)
Rats/group (n)
Control
Experimental
66 66 66 66 66 66 14
59.6 _+3.2 81.2 _+4.8 111.4_+ 8.0 136.8_+14.7 168.6_+17.1 109.0 + 16.6 132 + 3
59.3 _+2.8 79.9 _+4.2 107.5_+5.9 125.3_+13.6 148.5_+17.0" 89.2 _+ 16.3* 86 _+8**
Statistically significant difference (Wilcoxon two-sample test) between control and experimental rats: *P < 0.01 ; **P < 0.0001.
Table 3 Effect of dietary iron deficiency in plasma TPL, EMB-PC and EMB-PE fatty acid compositions of male Wistar rats Plasma TPL
EMB-PC
EMB-PE
Fatty acid
Control (n = 9)
Experimental (n = 12)
Control (n = 9)
Experimental (n = 12)
Control (n = 9)
Experimental (n = 12)
C16:0 C16:1 (07 C18:0 C18:1m9 C18:2m6 C18:3m6 C18:3m3 C20:3m9 C20:3m6 C20:4m6 C20:5m3 C22:4m6 C22:5m6 C22:5m3 C22:6m3
29.38 _+ 1.70 0.56 _ 0.22 22.71 _ 1.77 8.96_+1.28 18.07 _+ 1.93 0.16 _+0.10 0.02 -+ 0.03 0.53 _+0.22 0.31 _+0.13 13.88 _+ 1.43 ND 0.65 + 0.27 2.73 _+0.65 ND 2.03 _+0.35
28.02 _+ 1.55 0.48 _+0.24 25.19 _+2.10" 9.66_+0.83 15.66 -+ 1.33** 0.24 _+0.03** 0.10 -+ 0.07** 0.47 _+0.10 0.53+0.11"* 13.28 _+ 1.18 ND 0.83 _+0.11 2.36 _+0.45 ND 3.17 _+0.49***
34.85 + 1.73 1.01 + 1.80 16.90 _+ 1.17 11.35_+0.78 11.02 _+0.97 0.04 _+0.05 0.33 _+0.12 0.68 _+0.42 0.14_+0.13 21.17 _+3.23 0.02 _+0.05 0.49 _+0.09 0.91 _ 0.31 0.04 _+0.05 1.03 _+0.40
32.46 _+2.95* 0.45 _+0.32 17.80 _+ 1.02 11.39_+0.90 12.12 _+0.67** 0.05 _+0.06 0.11 _+0.07*** 0.75 _+0.41 0.21 _+0.20 21.84 _+2.91 0.04 _+0.06 0.60 _+0.04** 0.81 _+0.14 0.08 _+0.09 1.26 _+0.37
14.08 + 2.05 3.21 _+ 1.38 8.27 _+ 1.65 16.47_+1.18 6.24 _+0.67 ND 0.87 _+0.48 2.20 _+0.89 0.03_+0.09 38.04 _+3.01 0.03 _+0.09 5.57 _+ 1.08 0.88 _+0.94 1.27 _+0.85 2.82 _+0.78
12.10 + 1.95 2.38 _+2.22 7.24 _+ 1.01 14.92-+1.08"* 6.49 +_0.59 0.02 _+0.09 0.21 _+0.25*** 2.83 _+2.15 0.10_+0.30 41.36 _+3.45* 0.15 _+0.17 6.47 _+ 1.57 0.37 _+0.45 1.95 _+0.50 3.40 _+0.89
Statistically significant difference between control and experimental rats: *P < 0.05, **P < 0.01, ***P < 0.001. EMB, erythrocyte membrane; ND, Not Detected; PC, phosphatidylcholine; PE, phosphatidylethanolamine; TPL, total phospholipids.
(P< 0.001). The EMB-PE fraction, situated on the inside of the membrane, TM presented a different picture. Mthough ALA was lower (P< 0.001) as seen in the EMB-PC fraction, arachidonic acid (AA; C20:4co6) was higher (P < 0.05) in the experimental group, with a concomitant decrease (P< 0.05) of oleic acid (OA; C18:1m9). A trend to higher OA in plasma TPL and decreased OA of the EMB-PE fraction were respectively correlated ( r = -0.63, P = 0.0023 and r = 0.60, P = 0.0041) with the percentage growth expressed by weight gained after 4 weeks on a diet that was iron deficient (Table 4). DGLA in the plasma TPL correlated negatively (r = -0.69, P = 0.005) with growth. A positive correlation (r = 0.58 P = 0.0058) was seen with palmitic acid in the EMB-PC fraction. A shift to synthesis of m3 fatty acids in the EMBPC fraction, where ALA was lower and DHA higher, in response to a low iron intake, resulted in respective positive and negative correlations (r = 0.52, P = 0.0168 and r = -0.64, P = 0.0019). © Pearson Professional Ltd 1997
DISCUSSION
The resuks of the present study revealed a reduced growth rate (lower percentage body weight gained) and lower plasma TPL LA levels, known to be symptoms of classic EFA deficiency 19,2°in the experimental rats. Unlike the results reported by others m,12and despite an adequate supply of nutrients (except for iron) from a diet 13,14 designed for growth and maintenance of rats during the first year of life, dietary ID affected the incorporation of LA in plasma phospholipids in this study. As plasma TPL stearic acid and OA levels were higher, a reduced a6-desaturase activity was suspected initially. 22 However, significant increases in GLA and DGLA indicated that A6-desaturase activity was active. A trend towards lower AA levels was seen in the plasma TPL with dietary 11) that could possibly explain the reduced growth rate, as the status of AA has been shown to correlate with growth in preterm infants. = However, as
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
232
H. Y. Tichelaar et al
Table 4 Spearman correlation coefficients (r) for the relation between fatty acids and percentage growth of male Wistar rats (n = 21) r-value
P-value
Plasma TPL C18:10)9 C18:30)6 C20:30)6 C22:50)6
-0.63 -0.48 -0.69 0.56
0.0023 0.0277 0.0005 0.0083
EMB-PC C16:0 C18:30)3 C20:50)3 C22:5m6 C22:60)3
0.58 0.52 -0.47 -0.44 -0.62
0.0058 0.0168 0.0328 0.0448 0.0025
EMB-PE C18:1(o9 C18:30)3 C22:50)3 C22:60)3
0.60 0.47 -0.50 -0.64
0.0041 0.0332 0.0220 0.0019
EMB, erythrocyte membrane; PC, phosphatidylcholine; PE, phosphatidylethanolamine; TPL, total phospholipids.
the difference between the groups was not significant, an explanation for the effect of dietary ID on growth was required from factors other than metabolic LA deficiency. Although the percentage composition of AA that was decreased was insignificant, Foreman-Van Drongelen et a123 have found that the absolute fatty acid levels m a y be different from the relative fatty acid levels and this difference could explain the observed effect on growth. An increase of plasma TPL DHA levels ( P < 0.05) and a trend towards higher EMB-PE DHA levels are reassuring indications that neurological development is unlikely to be affected by dietary ID. 24 However, data obtained from rats should not be extrapolated to h u m a n s owing to differences in nutrient requirements, intake and metabolism; 25 we found that children with ID had lower EMB m3 fatty acids. 6 Lipid modifications of EMB and neural membranes are similar, 26 and as DHA is an essential component in the nervous system, 2z visual and cognitive development is unlikely to be affected by ID.2s The higher levels of EMB-PE AA suggest that the apparent metabolic LA deficiency in this study had no negative effect on the metabolism of tissue membranes. A higher incorporation of AA in neurological phospholipids m a y even be an indication of increased neurotransmitter activity 29 and active k5-desaturase activity, s° The higher LA (P < 0.05) in the EMB-PC of the experimental rats, contrary to the decreased LA incorporation into plasma TPL, could be a protection mechanism to prevent tissues from developing EFA deficiency. Higher levels of docosatetraenoic acid (P < 0.01) and lower levels of palmitic acid (P < 0.05) and ALA (P < 0.001) further demonstrate a delicate balance between fatty acids and
especially between m6 and m3 fatty acids. Reduced OA levels in EMB-PE (P < 0.01), lower ALA and concomitant higher AA levels with a tendency towards increased C22:4m6 and DHA levels are also clear indications of active A5- and A4-desaturase activity in the EMB. This study demonstrated that plasma and EMB fatty acids are very sensitive to changes in iron status, albeit with contrasting effects. This can be explained by the simukaneous involvement of both iron and fatty acids in the integrity of the EMB. This p h e n o m e n o n occurred irrespective of species investigated. 6 It seems that in the rat model, the EMB compensates by increasing the m3 and m6 fatty acid content of EMBs to stabilize m e m b r a n e integrity, despite a reduction in haemoglobin concentration. Significant correlations between fatty acids and growth are an indication that the mechanism, whereby dietary ID affects growth, is in some way related to abnormal fatty acid shifts that disturb the delicate balance of EFAs in membranes. To overcome the effects of dietary ID on fatty acid metabolism, the addition of GLA to the AIN7 6 T M Purified Diet should be investigated sl although a balanced addition of m3 and m6 fatty acids would probably be more appropriate. This opens exciting new prospects for future ID intervention programmes to study the effect of fatty acid supplementation on growth under conditions of known micronutrient deficiency.
ACKNOWLEDGEMENTS
The authors are very grateful to Mr P. de Lange, Mr J. V. Seier and Miss J. van Wyk for their technical assistance. This work was supported financially by Scotia Pharmaceuticals.
REFERENCES
1. Mien L. H. Iron-deficiency anemia increases risk of preterm delivery. Nutr Rev 1993; 51: 49-52. 2. Sheard N. F. Iron deficiency and infant development. Nutr Rev 1994; 52: 137-140. 3. LozoffB.,Jimenez E., Wolf A. W. Long-term developmental outcome of infants with iron deficiency. N Engl J Med 1991; 325: 687-694.
4. Idjradinata P., Pollitr E. Reversal of developmental delays in iron-deficient anaemic infants treated with iron. Lancet 1993; 341: 1-4.
5. Haschke F., Vanura H., Male C. et al. Iron nutrition and growth of breast- and formula-fed infants during the first 9 months of life. J Pediatr Gastroenterol Nutr 1993; 16: 151-156. 6. Smuts C. M., Tichelaar H. Y., Van Jaarsveld P.J. et al. The effect of iron fortification on the fatty acid composition of plasma and erythrocyte membranes in primary school children with and without iron-deficiency. Prostaglandins Leukot Essent FarO/Acids 1995; 52: 59-67. 7. Angeles I. T., Schultink W.J., Matulessi P., Gross R., Sastroamidjojo S. Decreased rate of stunting among anemic Indonesian preschool children through iron supplementation. Am J Clin Nulr 1993 58: 1-4. 8. Steyn N. P., Nel J. H., Dhansay M. A. et al. Differences in
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
© Pearson Professional Ltd 1997
Dietary iron deficiency and composition of rat membrane phospholipids
9. 10.
11.
12.
13. 14. 15.
16.
17. 18.
19.
20.
nutritional status between caretakers of underweight and normal weight Pedi preschool children. S Afr J Food Sci Nutr 1995; 7: 53-59. Beard J. L., Connor J. R., Jones B. C. Iron in the brain. Nutr Rev 1993, BI: 157-170. Culmane S. C., McAdoo K. R_ Iron intake influences essential fatty acid and lipid composition of rat plasma and erythrocytes. JNutr 1987; 117: 1514-1519. Yehuda S., Carasso R. L. Modulation of learning, pain thresholds, and thermoregulation in the rat by preparations of free purified a-linolenic and linoleic acids: determination of the optimal co3-to-¢06 ratio. Proc Natl Acad Sci USA 1993; 90: 10345-10349. Johnson S. B., Kramer T. R., Briske-Anderson M., Holman R. T. Fatty acid pattern of tissue phospholipids in copper and iron deficiencies. Lipids 1989; 24: 141-145. Ad Hoc Committee. Report on Standards for Nutritional Studies. J Nutr 1977; 107: 1340-1348. Ad Hoc Committee. Second report oil Standards for Nutritional Studies. JNutr 1980; 110: 1726. Steck T. L., Kant J. A. Preparation of impermeable ghosts and inside-out vesicles from h u m a n erythrocyte membranes. In: Methods in Enzymology. New York: Academic Press, 1974: 172-173. Burton G. W., Ingold K. U., Thompson K. E. An improved procedure for the isolation of ghost membranes from h u m a n red blood cells. Lipids 1981; 16: 946. Tichelaar H. Y., Steyn N. P., Nel J. H. et al. S AfrJ Food Sci Nutr 1995; 7: 27-32. Cartwright I.J., Pockley A. G., Galloway J. H., Greaves M., Preston F. E. The effect of dietary c0-3 polyunsaturated fatty acids and el3rthrocyte membrane phospholipids, erythrocyte deformability and blood viscosity in healthy volunteers. Atherosclerosis 1985; BS: 267-281. Burr G. O., Burr M. M. A new deficiency disease produced by the rigid exclusion of fat from the diet. JBiol Chem 1929; 99: 345-367. Holman I~ T. Essential Fatty Acids and Eicosanoids. American Oil Chemists' Society, Champaign, Illinois, 1992:3-17.
© Pearson Professional Ltd 1997
233
21. Rao G.A., Crane R. T., Larkin E. C. Reduction of hepatic stearoyl-CoA desaturase activity in rats fed iron-deficient diets. Lipids 1983; 18: 573-575. 22. 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 USA 1993; 90: 1073-1077. 23. Foreman-Van Drongelen M. M. H. P., Houwelingen A. C. V., Kester A. D. M. et al. Long-chain polyene status of preterm infants with regard to the fatty acid composition of their diet: comparison between absolute and relative fatty acid levels in plasma and erythrocyte phospholipids. BrJ Nutr 1995; 73: 405-422. 24. Carlson S. E. Lessons learned from randomized infants to marine off-supplemented formulas in nutritional trials. J Pediatr 1994; 125: $33-$38. 25. Innis S. M. Fatty acid requirements of the newborn. Can J PhysiolPharmaco11994; 72: 1483-1492. 26. Carlson S. E., Carver J. D., House S. G. High fat diet varying in ratios of polyunsaturated to saturated fatty acid and linoleic to linolenic acid: a comparison of rat neural and red cell membrane phospholipids. JNutr 1986; 116: 718-725. 27. Salem N., Niebylski C. D. The nervous system has an absolute molecular species requirement for proper function. Mol Mere Bio11995; 12: 131-134. 28. Carlson S. E., Werkman S. H., Peeples J. M., Wilson W. M. Growth and development of premature infants in relation to c03 to c06 fatty acid status. WorldRev NutrDiet 1994; 7B: 63-69. 29. Fonlupt P., Croset M., Lagarde M. Incorporation of arachidonic and docosahexaerloic acids into phospholipids of rat brain membranes. Neurosci Le# 1994; 171: 137-141. 30. Whelan J., Broughton K. S., Surette M. E., Kinsella J. E. Dietary arachidonic and linoleic acids: comparative effects on tissue lipids. Lipids 1992; 27: 85-88. 31. Horrobin D. F. Gamma-linolenic acid: An intermediate in essential fatty acid metabolism with potential as an ethical pharmaceutical and as a food. Rev Contemp Pharmacother 1990; 1: 1-45.
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
© Pearson ProfessionalLtd 1997
The effect of dietary iron deficiency on the fatty acid composition of plasma and erythrocyte m e m b r a n e phospholipids in the rat H. Y. Tichelaar, 1 C. M. Smuts, 1 R. Gross, 2 P. L. Jooste, 1 M. Faber, 1 A. J. S. B e n a d 4 ~ ~National Research Programme for Nutritional Intervention of the Medical Research Council, Tygerberg, 7505, South Africa 2SEAMEO-TROPMED-GTZ, Community Nutrition Training Programme, University of Indonesia, Jakarta, Indonesia
Severe iron deficiency was introduced in rats by feeding outbred male Wistar rats a purified diet that was either adequate or deficient in iron. The rats were weighed regularly over 4 weeks to monitor body weight differences, after which blood was drawn from a subsample to determine the haemoglobin concentrations and fatty acid composition of plasma total phospholipids and to measure the erythrocyte membrane phosphatidylcholine and phosphatidylethanolamine levels. Comparisons between dietary iron adequate (control) and dietary iron deficient (experimental) rats showed that the experimental rats had lower body weight and plasma total phospholipid linoleic acid levels typical of the symptoms of essential fatty acid deficiency. Erythrocyte membrane phosphatidylethanolamine arachidonic acid levels were increased (P < 0.05) with concomitant decreases in oleic acid (P < 0.01). Correlations between fatty acids and growth suggest that the mechanism whereby iron deficiency affects growth is in some way related to abnormal fatty acid shifts that disturb the delicate balance of essential fatty acids in membranes. Additional 0)3 and 0)6 fatty acids may be necessary to counteract the effect of iron deficiency in rats. Summary
INTRODUCTION
The risks of iron deficiency (ID) are many and include preterm delivery and low birth weight of infants. 1 ID also impairs mental and motor development in infants 2 and may have poor long-lasting developmental outcomes 3 that can be corrected by early treatment with ferrous sulphate. 4 Fast growing healthy infants may even develop ID and need exogenous iron supplementation after 6 months of age? Therefore ID and growth present major nutritional problems among young children in developing countriesS -8 The role of iron in the brain is varied and includes the incorporation into enzymes and proteins, affecting behavioural systems through peroxide reduction, amino acid metabolism and fatty acid desaturation. 9 Clinical evidence suggests that ID influences essential fatty acid (EFA) metabolism 6,1° by reducing the Received 12 March 1996 Accepted 23 August 1996 Correspondence to: H. Y. Tichelaar, Tel. +27 21 938-0409; Fax. +27 21 9380321; E-mail. [email protected]..
activity of the A9- and A6-desaturase enzymes. The effects of an EFA metabolite deficiency have been overcome by supplementing oils containing a 4:1 ratio of ~-linolenic acid (ALA; C18:3o3) to linoleic acid (LA; C18:2~06) to rats, resulting in improved learning performance) 1 The rat presents a suitable and sensitive model to study ID, which rarely reaches levels of severity in humans that can be achieved in laboratory rats. l°,~a In the present study we have investigated the effect of ID on the fatty acid composition of plasma and erythrocyte membranes (EMB) in the rat to elucidate the involvement of iron in fatty acid metabolism and, indirectly, its effect on growth. MATERIALS AND METHODS Animals
Outbred male Wistar rats (n = 150) weighing =59 g were weaned at 3 weeks of age (21 + 1 day). The weaned rats received a powder rat diet, AIN-76 TM Purified Diet, 13,14ad libitum (Table 1) during an adaptation period of 3 days. 229
230
H. Y. Tichelaar et al
Table I Composition of the AIN-76TM purified dieta Ingredient
Diet (%)
Casein Methionine Cornstarch Sucrose Glucose Cellulose-type fibre Sunflower oil AIN mineral mixture b AIN vitamin mixture ° Cholien-bicartate
20.00 0.30 21.66 21.66 21.66 5.00 5.00 3.50 1.00 0.20
aRats received 5 g/kg/day. bComposition of the mineral mixture g/kg: CaHPO4, 500; NaCI, 74; K3C6HsOT.H20,220; K2SO4, 52; MgO, 24; MnC%, 3.5; Ferric citrate, 6; ZnC%, 1.6; CuC%, 0.3; KIO3, 0.01; Na2SEO3.5H20,0.01; CrK(SO,)2.12H20, 0.55; sucrose, 1000. °Vitamin mixture per kg: thiamin-HCI, 600 mg; riboflavin, 600 mg; pyridoxine-HCI, 700 mg; nicotinic acid, 3 g; D-calcium pantothenate, 1.6 g; folic acid, 200 mg; D-biotin, 20 mg; cyanobalamin, 1 mg; retinyl palmitate (vitamin A) 400 000 IU; d/-c~-tocopherolacetate (vitamin E), 5000 IU; cholecalciferol (vitamin D3), 2.5 mg; menaquinone (vitamin K), 50 mg; sucrose, 1000 g.
Their individual food intakes were measured daily and their body weights on day 1 and day 3. Rats with extreme food intakes and body weights during the adaptation period were withdrawn from the study until 132 rats remained. They were then randomly allocated to two groups each consisting of 66 housed rats. This sample size provided two groups of rats with no mean body weight difference at baseline. The experimental protocol was approved by the ethics committees of the South African Medical Research Council and the Regional SEAMEO-TROPMED Center on Community Nutrition, Indonesia. Experimental design
The experiment consisted of two groups of rats. A control group (n = 66) was fed the powdered AIN-76 TM diet ad libitum. The experimental group of rats (n = 66) received the same diet as the control group ad libitum, but with iron (ferric citrate) omitted from the mineral mixture. Mineral-free distilled water was freely available to the rats throughout the adaptation and experimental phases. Rats were weighed at the beginning of the experimental phase and weekly thereafter for four consecutive weeks on an electronic scale. The food intake of the rats was determined individually on alternate days for the duration of the experimental phase. Rats were ranked within the groups according to their final body weights. Every fourth rat was included in a subsample of 14 rats per group. Rats were anaesthetized with Sagatal (0.1 ml per 100 g body weight injected peritoneally) prior to blood
collection. About 4 ml blood was taken directly from the heart with an EDTA (ethylenediaminetetracetic acid) flushed syringe and the blood sample was transferred to an EDTA tube. Haemoglobin concentrations were determined spectrophotometrically on the subsample (n = 28) using the metcyanohaemoglobin method in preprepared cuvettes and the Minilab photometer from Ames. Erythrocyte membranes were prepared by haemolyzing erythrocytes with different phosphate buffersY '16 Lipids were extracted from plasma and EMB with chloroform/ methanol (2:1 vol/vol), separated by thin-layer chromatography and analysed for fatty acid composition of plasma total phospholipid (TPL), EMB phosphatidylcholine (-PC) and EMB phosphatidylethanolamine (-PE) by gas-liquid chromatography. 6,1z Statistical analysis
The difference between the fatty acid results obtained from the control and experimental animals was analysed statistically using the Wilcoxon two-sample test. Spearman correlation coefficients were calculated for fatty acids versus growth, irrespective of dietary treatment used. RESULTS
Table 2 shows the effect of dietary ID on growth of male Wistar rats after 4 weeks. The percentage weight gain of the control group was 33% (182.9% versus 150.4%) more than that of the experimental group. Control animals also had significantly higher (P < 0.0001) haemoglobin concentrations (132 + 3 g/L) than the experimental animals (86 + 8 g/L). For comparisons, the fatty acids of only nine control and 12 experimental rats could be used as the other five control and two experimental rats had either insufficient plasma or EMB for fatty acid analyses. The effect of dietary ID on plasma TPL, EMB-PC and EMB-PE fatty acid compositions of the rats is shown in Table 3. The fatty acids display significant differences between the two groups of rats after 4 weeks on their respective diets. In the plasma TPL fraction, stearic acid (C18:0) was higher (P < 0.05) in the experimental group. This was at the expense of LA, that was lower (P< 0.01), although 7-1inolenic acid (GLA; C18:3(06) and dihomoy-linolenic acid (DGLA; C20:3(06) were higher (P < 0.01). The (03 fatty acids, ALA and docosahexaenoic acid (DHA; C22:6(03) were respectively higher (P < 0.01 and P < 0.001). In the EMB-PC fraction, palmitic acid was lower (P < 0.05) in the experimental group. LA and docosatetraenoic acid (C22:4(06) were respectively higher (P< 0.05 and P < 0.01). Only the (03 fatty acid, ALA, was lower
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
© Pearson Professional Ltd 1997
Dietary iron deficiency and composition of rat membrane phospholipids
231
Table 2 Effect of dietary iron deficiency on body weight, weight gain and haemoglobin concentration compared with an adequate control diet
Baseline body weight (g) Body weight after 1 week (g) Body weight after 2 weeks (g) Body weight after 3 weeks (g) Body weight after 4 weeks (g) Weight gained (g) Haemoglobin (g/I)
Rats/group (n)
Control
Experimental
66 66 66 66 66 66 14
59.6 _+3.2 81.2 _+4.8 111.4_+ 8.0 136.8_+14.7 168.6_+17.1 109.0 + 16.6 132 + 3
59.3 _+2.8 79.9 _+4.2 107.5_+5.9 125.3_+13.6 148.5_+17.0" 89.2 _+ 16.3* 86 _+8**
Statistically significant difference (Wilcoxon two-sample test) between control and experimental rats: *P < 0.01 ; **P < 0.0001.
Table 3 Effect of dietary iron deficiency in plasma TPL, EMB-PC and EMB-PE fatty acid compositions of male Wistar rats Plasma TPL
EMB-PC
EMB-PE
Fatty acid
Control (n = 9)
Experimental (n = 12)
Control (n = 9)
Experimental (n = 12)
Control (n = 9)
Experimental (n = 12)
C16:0 C16:1 (07 C18:0 C18:1m9 C18:2m6 C18:3m6 C18:3m3 C20:3m9 C20:3m6 C20:4m6 C20:5m3 C22:4m6 C22:5m6 C22:5m3 C22:6m3
29.38 _+ 1.70 0.56 _ 0.22 22.71 _ 1.77 8.96_+1.28 18.07 _+ 1.93 0.16 _+0.10 0.02 -+ 0.03 0.53 _+0.22 0.31 _+0.13 13.88 _+ 1.43 ND 0.65 + 0.27 2.73 _+0.65 ND 2.03 _+0.35
28.02 _+ 1.55 0.48 _+0.24 25.19 _+2.10" 9.66_+0.83 15.66 -+ 1.33** 0.24 _+0.03** 0.10 -+ 0.07** 0.47 _+0.10 0.53+0.11"* 13.28 _+ 1.18 ND 0.83 _+0.11 2.36 _+0.45 ND 3.17 _+0.49***
34.85 + 1.73 1.01 + 1.80 16.90 _+ 1.17 11.35_+0.78 11.02 _+0.97 0.04 _+0.05 0.33 _+0.12 0.68 _+0.42 0.14_+0.13 21.17 _+3.23 0.02 _+0.05 0.49 _+0.09 0.91 _ 0.31 0.04 _+0.05 1.03 _+0.40
32.46 _+2.95* 0.45 _+0.32 17.80 _+ 1.02 11.39_+0.90 12.12 _+0.67** 0.05 _+0.06 0.11 _+0.07*** 0.75 _+0.41 0.21 _+0.20 21.84 _+2.91 0.04 _+0.06 0.60 _+0.04** 0.81 _+0.14 0.08 _+0.09 1.26 _+0.37
14.08 + 2.05 3.21 _+ 1.38 8.27 _+ 1.65 16.47_+1.18 6.24 _+0.67 ND 0.87 _+0.48 2.20 _+0.89 0.03_+0.09 38.04 _+3.01 0.03 _+0.09 5.57 _+ 1.08 0.88 _+0.94 1.27 _+0.85 2.82 _+0.78
12.10 + 1.95 2.38 _+2.22 7.24 _+ 1.01 14.92-+1.08"* 6.49 +_0.59 0.02 _+0.09 0.21 _+0.25*** 2.83 _+2.15 0.10_+0.30 41.36 _+3.45* 0.15 _+0.17 6.47 _+ 1.57 0.37 _+0.45 1.95 _+0.50 3.40 _+0.89
Statistically significant difference between control and experimental rats: *P < 0.05, **P < 0.01, ***P < 0.001. EMB, erythrocyte membrane; ND, Not Detected; PC, phosphatidylcholine; PE, phosphatidylethanolamine; TPL, total phospholipids.
(P< 0.001). The EMB-PE fraction, situated on the inside of the membrane, TM presented a different picture. Mthough ALA was lower (P< 0.001) as seen in the EMB-PC fraction, arachidonic acid (AA; C20:4co6) was higher (P < 0.05) in the experimental group, with a concomitant decrease (P< 0.05) of oleic acid (OA; C18:1m9). A trend to higher OA in plasma TPL and decreased OA of the EMB-PE fraction were respectively correlated ( r = -0.63, P = 0.0023 and r = 0.60, P = 0.0041) with the percentage growth expressed by weight gained after 4 weeks on a diet that was iron deficient (Table 4). DGLA in the plasma TPL correlated negatively (r = -0.69, P = 0.005) with growth. A positive correlation (r = 0.58 P = 0.0058) was seen with palmitic acid in the EMB-PC fraction. A shift to synthesis of m3 fatty acids in the EMBPC fraction, where ALA was lower and DHA higher, in response to a low iron intake, resulted in respective positive and negative correlations (r = 0.52, P = 0.0168 and r = -0.64, P = 0.0019). © Pearson Professional Ltd 1997
DISCUSSION
The resuks of the present study revealed a reduced growth rate (lower percentage body weight gained) and lower plasma TPL LA levels, known to be symptoms of classic EFA deficiency 19,2°in the experimental rats. Unlike the results reported by others m,12and despite an adequate supply of nutrients (except for iron) from a diet 13,14 designed for growth and maintenance of rats during the first year of life, dietary ID affected the incorporation of LA in plasma phospholipids in this study. As plasma TPL stearic acid and OA levels were higher, a reduced a6-desaturase activity was suspected initially. 22 However, significant increases in GLA and DGLA indicated that A6-desaturase activity was active. A trend towards lower AA levels was seen in the plasma TPL with dietary 11) that could possibly explain the reduced growth rate, as the status of AA has been shown to correlate with growth in preterm infants. = However, as
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
232
H. Y. Tichelaar et al
Table 4 Spearman correlation coefficients (r) for the relation between fatty acids and percentage growth of male Wistar rats (n = 21) r-value
P-value
Plasma TPL C18:10)9 C18:30)6 C20:30)6 C22:50)6
-0.63 -0.48 -0.69 0.56
0.0023 0.0277 0.0005 0.0083
EMB-PC C16:0 C18:30)3 C20:50)3 C22:5m6 C22:60)3
0.58 0.52 -0.47 -0.44 -0.62
0.0058 0.0168 0.0328 0.0448 0.0025
EMB-PE C18:1(o9 C18:30)3 C22:50)3 C22:60)3
0.60 0.47 -0.50 -0.64
0.0041 0.0332 0.0220 0.0019
EMB, erythrocyte membrane; PC, phosphatidylcholine; PE, phosphatidylethanolamine; TPL, total phospholipids.
the difference between the groups was not significant, an explanation for the effect of dietary ID on growth was required from factors other than metabolic LA deficiency. Although the percentage composition of AA that was decreased was insignificant, Foreman-Van Drongelen et a123 have found that the absolute fatty acid levels m a y be different from the relative fatty acid levels and this difference could explain the observed effect on growth. An increase of plasma TPL DHA levels ( P < 0.05) and a trend towards higher EMB-PE DHA levels are reassuring indications that neurological development is unlikely to be affected by dietary ID. 24 However, data obtained from rats should not be extrapolated to h u m a n s owing to differences in nutrient requirements, intake and metabolism; 25 we found that children with ID had lower EMB m3 fatty acids. 6 Lipid modifications of EMB and neural membranes are similar, 26 and as DHA is an essential component in the nervous system, 2z visual and cognitive development is unlikely to be affected by ID.2s The higher levels of EMB-PE AA suggest that the apparent metabolic LA deficiency in this study had no negative effect on the metabolism of tissue membranes. A higher incorporation of AA in neurological phospholipids m a y even be an indication of increased neurotransmitter activity 29 and active k5-desaturase activity, s° The higher LA (P < 0.05) in the EMB-PC of the experimental rats, contrary to the decreased LA incorporation into plasma TPL, could be a protection mechanism to prevent tissues from developing EFA deficiency. Higher levels of docosatetraenoic acid (P < 0.01) and lower levels of palmitic acid (P < 0.05) and ALA (P < 0.001) further demonstrate a delicate balance between fatty acids and
especially between m6 and m3 fatty acids. Reduced OA levels in EMB-PE (P < 0.01), lower ALA and concomitant higher AA levels with a tendency towards increased C22:4m6 and DHA levels are also clear indications of active A5- and A4-desaturase activity in the EMB. This study demonstrated that plasma and EMB fatty acids are very sensitive to changes in iron status, albeit with contrasting effects. This can be explained by the simukaneous involvement of both iron and fatty acids in the integrity of the EMB. This p h e n o m e n o n occurred irrespective of species investigated. 6 It seems that in the rat model, the EMB compensates by increasing the m3 and m6 fatty acid content of EMBs to stabilize m e m b r a n e integrity, despite a reduction in haemoglobin concentration. Significant correlations between fatty acids and growth are an indication that the mechanism, whereby dietary ID affects growth, is in some way related to abnormal fatty acid shifts that disturb the delicate balance of EFAs in membranes. To overcome the effects of dietary ID on fatty acid metabolism, the addition of GLA to the AIN7 6 T M Purified Diet should be investigated sl although a balanced addition of m3 and m6 fatty acids would probably be more appropriate. This opens exciting new prospects for future ID intervention programmes to study the effect of fatty acid supplementation on growth under conditions of known micronutrient deficiency.
ACKNOWLEDGEMENTS
The authors are very grateful to Mr P. de Lange, Mr J. V. Seier and Miss J. van Wyk for their technical assistance. This work was supported financially by Scotia Pharmaceuticals.
REFERENCES
1. Mien L. H. Iron-deficiency anemia increases risk of preterm delivery. Nutr Rev 1993; 51: 49-52. 2. Sheard N. F. Iron deficiency and infant development. Nutr Rev 1994; 52: 137-140. 3. LozoffB.,Jimenez E., Wolf A. W. Long-term developmental outcome of infants with iron deficiency. N Engl J Med 1991; 325: 687-694.
4. Idjradinata P., Pollitr E. Reversal of developmental delays in iron-deficient anaemic infants treated with iron. Lancet 1993; 341: 1-4.
5. Haschke F., Vanura H., Male C. et al. Iron nutrition and growth of breast- and formula-fed infants during the first 9 months of life. J Pediatr Gastroenterol Nutr 1993; 16: 151-156. 6. Smuts C. M., Tichelaar H. Y., Van Jaarsveld P.J. et al. The effect of iron fortification on the fatty acid composition of plasma and erythrocyte membranes in primary school children with and without iron-deficiency. Prostaglandins Leukot Essent FarO/Acids 1995; 52: 59-67. 7. Angeles I. T., Schultink W.J., Matulessi P., Gross R., Sastroamidjojo S. Decreased rate of stunting among anemic Indonesian preschool children through iron supplementation. Am J Clin Nulr 1993 58: 1-4. 8. Steyn N. P., Nel J. H., Dhansay M. A. et al. Differences in
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233
© Pearson Professional Ltd 1997
Dietary iron deficiency and composition of rat membrane phospholipids
9. 10.
11.
12.
13. 14. 15.
16.
17. 18.
19.
20.
nutritional status between caretakers of underweight and normal weight Pedi preschool children. S Afr J Food Sci Nutr 1995; 7: 53-59. Beard J. L., Connor J. R., Jones B. C. Iron in the brain. Nutr Rev 1993, BI: 157-170. Culmane S. C., McAdoo K. R_ Iron intake influences essential fatty acid and lipid composition of rat plasma and erythrocytes. JNutr 1987; 117: 1514-1519. Yehuda S., Carasso R. L. Modulation of learning, pain thresholds, and thermoregulation in the rat by preparations of free purified a-linolenic and linoleic acids: determination of the optimal co3-to-¢06 ratio. Proc Natl Acad Sci USA 1993; 90: 10345-10349. Johnson S. B., Kramer T. R., Briske-Anderson M., Holman R. T. Fatty acid pattern of tissue phospholipids in copper and iron deficiencies. Lipids 1989; 24: 141-145. Ad Hoc Committee. Report on Standards for Nutritional Studies. J Nutr 1977; 107: 1340-1348. Ad Hoc Committee. Second report oil Standards for Nutritional Studies. JNutr 1980; 110: 1726. Steck T. L., Kant J. A. Preparation of impermeable ghosts and inside-out vesicles from h u m a n erythrocyte membranes. In: Methods in Enzymology. New York: Academic Press, 1974: 172-173. Burton G. W., Ingold K. U., Thompson K. E. An improved procedure for the isolation of ghost membranes from h u m a n red blood cells. Lipids 1981; 16: 946. Tichelaar H. Y., Steyn N. P., Nel J. H. et al. S AfrJ Food Sci Nutr 1995; 7: 27-32. Cartwright I.J., Pockley A. G., Galloway J. H., Greaves M., Preston F. E. The effect of dietary c0-3 polyunsaturated fatty acids and el3rthrocyte membrane phospholipids, erythrocyte deformability and blood viscosity in healthy volunteers. Atherosclerosis 1985; BS: 267-281. Burr G. O., Burr M. M. A new deficiency disease produced by the rigid exclusion of fat from the diet. JBiol Chem 1929; 99: 345-367. Holman I~ T. Essential Fatty Acids and Eicosanoids. American Oil Chemists' Society, Champaign, Illinois, 1992:3-17.
© Pearson Professional Ltd 1997
233
21. Rao G.A., Crane R. T., Larkin E. C. Reduction of hepatic stearoyl-CoA desaturase activity in rats fed iron-deficient diets. Lipids 1983; 18: 573-575. 22. 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 USA 1993; 90: 1073-1077. 23. Foreman-Van Drongelen M. M. H. P., Houwelingen A. C. V., Kester A. D. M. et al. Long-chain polyene status of preterm infants with regard to the fatty acid composition of their diet: comparison between absolute and relative fatty acid levels in plasma and erythrocyte phospholipids. BrJ Nutr 1995; 73: 405-422. 24. Carlson S. E. Lessons learned from randomized infants to marine off-supplemented formulas in nutritional trials. J Pediatr 1994; 125: $33-$38. 25. Innis S. M. Fatty acid requirements of the newborn. Can J PhysiolPharmaco11994; 72: 1483-1492. 26. Carlson S. E., Carver J. D., House S. G. High fat diet varying in ratios of polyunsaturated to saturated fatty acid and linoleic to linolenic acid: a comparison of rat neural and red cell membrane phospholipids. JNutr 1986; 116: 718-725. 27. Salem N., Niebylski C. D. The nervous system has an absolute molecular species requirement for proper function. Mol Mere Bio11995; 12: 131-134. 28. Carlson S. E., Werkman S. H., Peeples J. M., Wilson W. M. Growth and development of premature infants in relation to c03 to c06 fatty acid status. WorldRev NutrDiet 1994; 7B: 63-69. 29. Fonlupt P., Croset M., Lagarde M. Incorporation of arachidonic and docosahexaerloic acids into phospholipids of rat brain membranes. Neurosci Le# 1994; 171: 137-141. 30. Whelan J., Broughton K. S., Surette M. E., Kinsella J. E. Dietary arachidonic and linoleic acids: comparative effects on tissue lipids. Lipids 1992; 27: 85-88. 31. Horrobin D. F. Gamma-linolenic acid: An intermediate in essential fatty acid metabolism with potential as an ethical pharmaceutical and as a food. Rev Contemp Pharmacother 1990; 1: 1-45.
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(3), 229-233