- Email: [email protected]
Intrauterine fatty acid accretion rates in human brain: implications for fatty acid requirements
Early Human Development, 1980, 412, 121-129 0 ElsevieriNorth-Holland Biomedical Press
121
Intrauterine fatty acid accretion rates in human brain: implications for fatty acid requirements M.T. CLANDININ, J.E. CHAPPELL, S. LEONG, T. HEIM, P.R. SWYER and G.W. CHANCE Department of Nutrition and Food Science and Department of Pediatrics, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
Accepted for publication 26 February 1980
SUMMARY
Fatty acid components of infant brain were determined to assess fatty acid requirements for synthesis of structural lipids in brain tissue during the last trimester of development in the fetus. Quantitative fatty acid analysis of cerebellum, frontal and occipital brain lobes indicated rapid accretion of chain elongation and desaturation products during the last trimester of brain growth. Frontal and occipital brain lobes were similar in fatty acid content. Fatty acid accretion rates were determined by regression analysis of tissue fat components at varying gestational ages. Tissue accretion of saturated and w-9 fatty acids, as well as total fatty acid content, paralleled increases in whole brain weight. Levels of linoleic (Clsz2, w -6) and linolenic (C1siB, w -3) acids were consistently low in brain during the last trimester of development, while marked substantial accretion of long chain desaturation products, arachidonic (C,,,,, w-6) and docosahexaenoic (C,,,,, w-3) acids occurred. Accretion of individual fatty acids of cerebellum also reflected changes in tissue total fatty acid content, with exception of the levels of C1aE3, w-3 and its chain elongation products present in cerebellum during the last trimester. These developmental changes and estimates ot fatty acid incorporation into whole brain and cerebellum are quantitatively relevant to estimation of fatty acid requirements of the low birth weight neonate. fetal; brain; fatty acid; accretion Direct correspondence to: Dr. M.T. Clandinin, Department of Nutrition and Food Science, Faculty of Medicine, University of Toronto, 150 College Street, Toronto, Ontario M5S lA8, Canada.
122 INTRODUCTION
Metabolism of polyunsaturated fatty acids by common pathways of chain elongation and desaturation results in formation of long chain polyenoic fatty acids for complex structural lipid synthesis (reviewed in ref. 16). These structural precursors, derived from dietary or maternal sources of ClsL2, w-6 and C 18,3, w -3 are necessary for appropriate physiological maturation of developing homeothermic animals [10,11,18]. The essential nature of C18:2, WS or its homologue Czoz4, w-6 is accepted in man (reviewed in ref. 9), while the requirement for C,,:,, w -3 is not yet established. Furthermore, little quantitative evidence is available to suggest whether in the human fetus Czzz4, w-6, CZZz5, w-6, C,,:,, o-3, C,,:,, w-3 fatty acids present in developing brain are elaborated from shorter chain precursors (i.e. C&, w-6 and C18:3r w-3) or, if the fetus is dependent upon mechanisms of placental synthesis and transfer in order to obtain longer chain polyenoic fatty acids required for synthesis of brain structural lipids. During the third trimester of human development, rapid synthesis of brain tissue occurs [4] in association with increasing neuromotor activity. Cell size, type and number increase, necessitating de novo synthesis of structural lipids by the developing fetus. Quantitative fatty acid analysis of brain throughout this period of organogenesis has not been reported, and thus reliable determination of accretion rates for each fatty acid constituent in normally nourished human brain has not been made. Clearly, further quantitative understanding of this question is particularly relevant to feeding of low birth weight (< 1500 g) premature infants fed humanized formulas or by intravenous routes particularly, as these formulations contain no C&, w-6, C,,:,, w-3, C 22:6, w-3 and varied levels of other w -3 fatty acids normally present in breast milk [ 31. Therefore, the current study was initiated to quantitate the fatty acid constituents present in human brain from 22 weeks of gestation to term. Regression analysis was applied to determine timing and accretion rates for all fatty acids quantitated.
MATERIALS AND METHODS
Sample population. All aspects of this study protocol have been ethically approved by the Human Experimentation Review Committee of the Hospital for Sick Children and the Department of Pediatrics, University of Toronto. Infant tissues were obtained during autopsy from a refrigerated cadaver within 16 hours post mortum. Ten male and 5 female infants were sampled from 22 weeks of gestational age (by dates, cf. ref. 6) to term. These infants died within 3 days of birth and were appropriate for gestational age (AGA) for birth weight, length and head circumference, with the exception of one infant, who was small for gestational age (SGA). Causes of death, as determined by clinical diagnosis and autopsy, were intrapartum asphyxia (N = 9), prematurity combined with intracerebral hemorrhage or asphyxia (N = 4),
123
congenital heart disease (N = 1) and diaphragmatic hernia (N = 1). Infants were therefore apparently normally nourished in utero, with no evidence of placental infarcts or maternal toxemia. Fresh cross-sectional samples (1 g) of cerebellum and the tip of the frontal and occipital brain lobes were placed in screw cap vials. Tissue samples were frozen and placed in a low temperature freezer (-60 “C) until extraction. Extraction and fatty acid analyses. Tissue lipids were quantitatively extracted under nitrogen with 50 vols of chloroform:methanol (2:1, v/v) utilizing a Folch extraction apparatus [ ‘7,19 ] and n-heptadecanoic acid as an internal standard. Fatty acid methyl esters were prepared with boron trifluoride methanol reagent [ 121. Methyl esters were quantitated by gas liquid chromatography in a dual column gas chromatography, equipped with flame ionization detectors and a chromatography data system for semi-automated operation (Varian model 3700 GLC with a CDS 111). Three-meter glass columns (3 mm I.D.), packed with silar-5CP (10% w/w) coated on acidwashed chromosorb W (80-100 mesh), were utilized. The GLC was operated utilizing temperature programming from 210 “C at 0.5 “C per min for 45 minutes with subsequent isothermal operation of the oven. Identification of methyl esters was accomplished by comparison of retention data with those of known standards, and by the method of ‘equivalent chain length’ [13]. Tissue fatty acid content was calculated on a pg of fatty acid per gram wet weight of tissue basis. Statistical analysis. Regression analysis of fatty acid composition was performed by computer and plotted for infants from 27 weeks of gestational age to term (N = 14). Accretion of fatty acids was determined by a general linear models procedure fitting a third-degree polynomial equation in the regression analysis. Degrees of freedom, sample correlation coefficients, significance levels and polynomial functions are indicated (Figs. l-4). Significant values of r were determined by reference to Steel and Torrie [ 151. Weekly accretion rates for each fatty acid were calculated by solving the regression equation. RESULTS
Brain weight increased some 4-5-fold during the last trimester of fetal development (Fig. 1). Individual cerebellum weights were not measured;
. .
25
30
35
40
4.5
Gestational Age (weeks)
Fig. 1. Brain weight as a function of gestational age. Brain weight (8) = 1050 - 111x + 3.glx2 -0.039x’, where x = gestational age in weeks. *** = P < 0.001, df = 14.
124
Past Conceptional Age (weeks)
Fig. 2. Cerebellum weight (% of whole brain weight) as a function of postconceptional age. Cerebellum weight data were abstracted from ref. 14. Cerebellum weight (W of whole brain weight) = 0.9815 + 0.1274 x, where x = postconceptional age in weeks. ** = P i
0.01. therefore, previously reported data for cerebellum weight, expressed as a percentage of whole brain weight [l] , was utilized to estimate total cerebellum weight for each brain sampled. The relationship derived (Fig. 2) was utilized to calculate total cerebellum fatty acid content (Fig. 4). Comparison of quantitative fatty acid analyses of frontal and occipital brain lobes by analysis of variance indicated no significant difference between these brain regions for fatty acid content per gram of tissue. Therefore, in subsequent regression analysis, values obtained for these two brain regions were averaged as representative of the fatty acid composition for each individual whole brain. Fatty acid content per gram of tissue also reflected total fatty acid content on a whole brain basis. During the last trimester, whole brain levels of C18:2, w -6 and C&, w -3 were low relative to other fatty acids with no significant accretion noted (Fig. 3A and B, respectively). However, marked accretion of products of chain elongation-desaturation, such as C&0:4, w-6 and C,,:,, w-3 occurred (Fig. 3C and D, respectively) resulting in accretion of w-6 and w-3 fatty acids (Fig. 3E and F, respectively) some 50-80-fold greater than for the essential precursors C!,,:,, w-6 and C,,:,, w -3. Accretion of w -9, saturated and total fatty acids also occurred during the last trimester (Fig. 3G, H and I, respectively) reflecting changes in brain weight (Fig. 1) and presumably myelination. Growing cerebellum contained remarkably lower levels of C1sL2, w-6 and ClsZ3, w-3 relative to levels of longer chain homologues (Fig. 4A-E). During the last trimester of cerebellum development accretion of all fatty acids reflected the accretion of total fatty acids in cerebellum (Fig. 4A, C, D, F, G and H) with the notable exception of the w -3 fatty acids (Fig. 4B and E). Accretion rates for C18:3, w-3 remained low in cerebellum during the normal gestational period (Fig. 4B and Table I). By solving the regression equations for fatty acid accretion at each week of gestation during the third trimester, mean accretion rates for each fatty acid can be computed (Table I). The sum of these means plus two standard errors provide estimates of the typical requirement for each fatty acid normally utilized per week for intrauterine synthesis of brain and cerebellum tissues.
125
Gestational Age (weeks) Fig. 3. Fatty acid content of infant brain expressed as a function of gestational age. Values of r and significance levels are indicated (df = 13). Totals for each fatty acid (mg) are computed where x = gestational age (weeks). A: total C,,:,, w-6 = -606.2 + 48.81x 1.269x’ + 0.0108x”. * = P < 0.05. B: total C,,:,, w-3 = 109.4 - 8.008~ + 0.1916xz 0.0014~~. C: total C,,:,, w-6 = 7049 - 599.4x + 16.59x1 - 0.145x’. *** = P < 0.001. D: total C,,;,, w-6 = 4621 + 410.9x - 12.05x’ + 0.1191x3. ** = P < 0.01. E: total w-6 fatty acids = 6269 - 521.6x + 14.09x’ - 0.1156x3. *** = P < 0.001. F: total w-3 fatty acids = - 9842 + 855.9x - 24.43~~ + 0.2321~‘. ** = P < 0.01. G: total w-9 fatty acids = 2983 - 255.9x + 7.134~~ - 0.0571~). ** = P < 0.01. H: total saturated fatty acids = 20,270 - 1739x + 48.73x z - 0.4188x”. ** = P < 0.01. I: total fatty acids = 34,060 -2998x + 86.34x’ - 0.7632~“. ** = P < 0.01.
1
126 TABLE I Accretion rates for fatty acids in infant brain and cerebellum Fatty acids
C Is:,> C,,:,, C,,:,, Total Total Total Total Total
w-6 w-3 w -6 w -6 w -3 w -9 saturates fatty acids**
Accretion rate (mg fatty acid/week)* Whole brain
Cerebellum
0.457 0.132 18.8 31.3 14.5 28.8 101 181
0.037 * 0.012 0.026 + 0.036 1.59 t 0.558 1.53 L 0.252 0.127 2 0.238 2.41 f 0.994 6.1 + 2.22 11.5 * 3.4 ~~___~ _ _ _ _ _ _ _ _
t 0.622 _+ 0.194 * 7.24 f 9.64 t 7.34 t 6.16 + 24.18 I 39.2
*From 26 weeks of gestation to term si + 2 SEZ. **Total fatty acids were calculated from individual tissue analyses.
Gestat~onal
Age
(weeks)
DISCUSSION
Development of infant brain occurs in association with a rapid rise in cellularity and declining water content 151. In the current study, this increase in cellular membranes is quantitatively reflected by increases in long-chain polyenoic fatty acid content in both brain and cerebellum. Absolute accretion rates of w -3 fatty acids, particularly C&, w-3, were greater in the antenatal period (Table I) relative to postnatal rates (to be published), suggesting that accretion of long-chain w-3 fatty acids during the third trimester may be particularly important in the normal rise in brain cellularity during this period of development. In the animal model specific fetal long-chain polyenoic fatty acids originate primarily from synthesis in maternal liver [14]. In the human, placental synthesis of CZoZ4, w -6 occurs [ 231 with concomitant fetal accretion of both o -6 and w-3 chain elongation-desaturation products (Figs. 3 and 4). However, it is not known whether these long-chain polyenoic components accrue as a result of placental transfer or fetal synthesis. In this regard, if microsomal chain elongation+esaturation of C!,,:,, w-6 and C&, w-3 occurs in infant brain or liver by enzyme systems analogous to those of mouse brain where at least 3 separate acyl-CoA elongating enzymes are involved [ 81, then developmental examination of chain elongating enzymes could have implications for the nutritional support of the low birth weight infant with fatty acids. Quantitation of the essential fatty acid composition of cerebrum and cerebellum clearly indicates how vulnerable the very low birth weight infant is in respect to essential fatty acid deficiency at birth. A premature AGA infant of 1300 g born at 30 weeks of gestation has a brain weight of = 170 g and cerebellum weight of == 8.5 g (Figs. 1 and 2). According to our present results the total w -6 and w -3 fatty acid content of these two organs is = 300 mg (Figs. 3 and 4). Analysis of the chemical composition of the developing fetus indicates that the total fat content of such an infant is approximately 28 g [20,21], and the estimated essential fatty acid content of adipose tissue is 1.8% [2,22], i.e. some 0.5 g. Thus, it is apparent that the reserve of essential fatty acids in this prematurely born infant is marginally greater than the essential fatty acid content and accretion rate in brain. Consequently, these vital structural components of brain can be provided from endogenous sources for only a very limited period. Fig. 4. Fatty acid content of infant cerebellum as a function of gestational age. Values of r and significance levels are indicated (df = 13). Totals for each fatty acid (mg) are como u t e d w h e r e Y = eestational see Iweeks). A: total C...,. w-6 = 14.67 - 1.296x + :-51 + 4.394x - 0.1238;’ + b.0372 x1 - 0:003x;. * = P < rj.O$. B: t&al C,,:,, w-3’ 0.0011~‘. * = P < 0.05. C: total C,,.., w-6 = -466.9+41.54x - 1.228~’ + 0.0123x”. ** = P < 0.01. D: total w-6 fatty acids = 208.7 - 20.11~ + 0.6224~’ -0.0058~‘. * = I’ < 0.05. E: total w-3 fatty acids = 157.4 - 15.21~ + 0.4827~~ -0.0049x’ . F: total w-9 fatty acids = -865.9 + 77.54~ - 2.301~’ + 0.0229x3. * = P < 0.06. G: total saturated fatty acids = -1842 + 164.7~ -4.879x’ + 0.0488~‘. ** = P < 0.01. H: total fatty acids = -2699 + 239.4x -7.039x1 + 0.0707~~. ** = P < 0.01.
128
The physician faces a quite different situation in the nutritional management of a more mature neonate. For example, if an infant is born after 35 weeks of gestation with a body weight of 2200 g (AGA) the essential fatty acid content can be similarly estimated at = 2.5 g of w -6 and w -3 fatty acids. At this developmental age the essential fatty acid content of the brain is = 0.45 g. Consequently, this infant is less at risk of essential fatty acid deficiency, and when under nutritional stress would be expected to maintain an essential fatty acid supply from endogenous sources for continued development of the central nervous system for a longer period of time. In emergency situations when only fat free intravenous alimentation is indicated, this knowledge of minimal essential fatty acid requirements necessary to preserve normal development of the central nervous system may prove invaluable. Furthermore, since growth and composition of tissues after birth are influenced by manipulations of dietary fatty acids, the present results suggest guide-lines for ‘natural’ or close to ‘optimal’ nutritional support with essential fatty acids for the premature human neonate. Current studies of energy metabolism in low birth weight infants fed mothers’ own milk indicate [17] that, for net energy intakes of 120 kcal/kg/ day, some 20 k&/kg/day of fat fed will be deposited in tissues (2.2 g/kg/ day). Based on quantitative fatty acid analysis of human milk [3] and third-trimester accretion rates (Fig. 3), it can be estimated that a significant quantity of w6 and o-3 fatty acids provided for tissue deposition at this level of net energy intake would be utilized in brain synthesis if postnatal accretion of these components is to occur in low birth weight infants (< 1500 g) at rates analogous to intrauterine accretion rates (Table I). Such calculations regarding utilization of fatty acids in tissue synthesis do not wholly consider tissue synthesis costs; however, they do indicate that a significant proportion of long-chain polyenoic fatty acids present in human milk may be of particular importance to nutriture of low birth weight neonates if brain synthesis and accretion of fatty acids in brain is to occur at rates analogous to those in the normally growing fetus.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the technical assistance of Mrs. S. Salciccioli. This study was supported by The Hospital for Sick Children Foundation and H.J. Heinz Foundation.
REFERENCES 1 Btinkov, S.M. and Gleeer, I. (1968): In: The Human Brain in Figures and Tables; A Quantitative Handbook, pp. 334-335 and 338-339. Basic Books, Plenum Press, New York. 2 Boersma, E.R. (1979): Changes in fatty acid composition of body fat before and after birth in Tanzania: an international comparative study. Br. Med. J., 1.850-853.
129
3 Chappell, J.E., Clandinin, M.T., Heim, T., Swyer, P.R. and Chance, G.W. (1980): Levels of long chain polyenoic fatty acids in early human milk, to be submitted. 4 Dobbing, J. (1972): Vulnerable periods of brain development. In: Lipids, Mainutrition and the Developing Brain, pp. 9-29. Ciba Foundation Symposia. Editors: K. Elliot and J. Knight. Elsevier, Amsterdam. 5 Dobbing, J. and Sands, J. (1973): The quantitative growth and development of human brain. Arch. Dis. Child., 48, 757-767. 6 Dubowitz, L.M.S., Dubowitz, V. and Goldberg, G. (1970): Clinical assessment of gestational age in the newborn infant. J. Pediatr., 77, l-10. 7 Folch, J., Lees, M. and Sloane-Stanley, G.H. (1957): A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem., 226, 497507. 8 Goldberg, I., Schechter, I. and Bloch, K. (1973): Fatty acyl-xxnzyme A elongation in brain of normal and quaking mice. Science, 182,497499. 9 Holman, R.T. (1973): Essential fatty acid deficiency in humans. In: Dietary Lipids and Postnatal Development, pp. 127-143. Editors: C. GaIli, G. Jacini and A. Pecile. Raven Press, New York. 10 Lamptey, M.S. and Walker, B.L. (1978): Physical and neurological development of the progeny of female rats fed an essential fatty acid-deficient diet during pregnancy and/orlactation. J. Nutr., 108,351-357. 11 Lamptey, MS. and Walker, B.L. (1978): Learning behaviour and brain lipid composition in rats subjected to essential fatty acid deficiency during gestation, lactation and growth. J. Nutr., 108,358-367. 12 Metcaife, L.D. and Schmitz, A.A. (1961): The rapid preparation of fatty acid esters for gas chromatographic analysis. Analyt. Chem., 33,363-364. 13 Miwa, T.K., Mikolajaczak, K.L., Earle, F.R. and Wolff, J.A. (1960): Gas chromatographic characterization of fatty acids. Identification of constants for mono and dicarboxylic methyl esters. Analyt. Chem., 32, 1739-1742. 14 Pascaud, M., Phan, H. and Renard, J.L. (1979): Transfert materno-foetai et captation des acides gras essentiels chez le rat. Ann. Biol. Anim. B&hem. Biophys., 19, 251256. 15 Steel. R.G.D. and Torrie. J.H. (1960): Principles and Procedures of Statistics. MC Graw-Hill, N e w Y o r k . 16 Tinoco, J., Babcock, R., Hincenbergs, I., Medwadowski, B., Miljanich,P. and Williams, M.A. (1979): Linolenic acid deficiency. Lipids, 14,166-173. 17 Verellen, G., Heim, T., Smith, J.M., Swyer, P., Atkinson, S.A. and Anderson, G.H. (1979): Partition of metabolisahle energy (ME) in AGA premature infants < 1300 g birth weight, age 14 weeks. Pediatr. Res., 13,409 (Ahstr.). 18 Walker, B.L. (1967): Maternal diet and brain fatty acids in young rats. Lipids, 2, 497600. 19 Walker, B.L. (1974): Convenient apparatus for extraction of tissue lipids. Lipids, 9, 938-940. 20 Widdowson, E.M. (1968): Growth and composition of the human foetus and newborn. In: Biology of Gestation, Vol. 2, pp. 148. Editor: N.S. Assah. Academic Press, New York. 21 Widdowson, E.M. and Dickerson, J.W.T. (1960): The effect of growth and function on the chemical composition of soft tissue. B&hem. J., 77, 30-33. 22 Widdowson, E.M., Dauncey, M.J., Gairdner, D.M.T., Jonxis, J.H.P. and PelikanFilipkora, M. (1975): Body fat of British and Dutch Infants. Br. Med. J., 1, 653-655. 23 Zimmermann, T., Winkler, L., Moller, U., Schubert, H. and Goetse, E. (1979): Synthesis of arachidonic acid in the human placenta in vitro. Biol. Neon., 36, 209-212.
121
Intrauterine fatty acid accretion rates in human brain: implications for fatty acid requirements M.T. CLANDININ, J.E. CHAPPELL, S. LEONG, T. HEIM, P.R. SWYER and G.W. CHANCE Department of Nutrition and Food Science and Department of Pediatrics, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
Accepted for publication 26 February 1980
SUMMARY
Fatty acid components of infant brain were determined to assess fatty acid requirements for synthesis of structural lipids in brain tissue during the last trimester of development in the fetus. Quantitative fatty acid analysis of cerebellum, frontal and occipital brain lobes indicated rapid accretion of chain elongation and desaturation products during the last trimester of brain growth. Frontal and occipital brain lobes were similar in fatty acid content. Fatty acid accretion rates were determined by regression analysis of tissue fat components at varying gestational ages. Tissue accretion of saturated and w-9 fatty acids, as well as total fatty acid content, paralleled increases in whole brain weight. Levels of linoleic (Clsz2, w -6) and linolenic (C1siB, w -3) acids were consistently low in brain during the last trimester of development, while marked substantial accretion of long chain desaturation products, arachidonic (C,,,,, w-6) and docosahexaenoic (C,,,,, w-3) acids occurred. Accretion of individual fatty acids of cerebellum also reflected changes in tissue total fatty acid content, with exception of the levels of C1aE3, w-3 and its chain elongation products present in cerebellum during the last trimester. These developmental changes and estimates ot fatty acid incorporation into whole brain and cerebellum are quantitatively relevant to estimation of fatty acid requirements of the low birth weight neonate. fetal; brain; fatty acid; accretion Direct correspondence to: Dr. M.T. Clandinin, Department of Nutrition and Food Science, Faculty of Medicine, University of Toronto, 150 College Street, Toronto, Ontario M5S lA8, Canada.
122 INTRODUCTION
Metabolism of polyunsaturated fatty acids by common pathways of chain elongation and desaturation results in formation of long chain polyenoic fatty acids for complex structural lipid synthesis (reviewed in ref. 16). These structural precursors, derived from dietary or maternal sources of ClsL2, w-6 and C 18,3, w -3 are necessary for appropriate physiological maturation of developing homeothermic animals [10,11,18]. The essential nature of C18:2, WS or its homologue Czoz4, w-6 is accepted in man (reviewed in ref. 9), while the requirement for C,,:,, w -3 is not yet established. Furthermore, little quantitative evidence is available to suggest whether in the human fetus Czzz4, w-6, CZZz5, w-6, C,,:,, o-3, C,,:,, w-3 fatty acids present in developing brain are elaborated from shorter chain precursors (i.e. C&, w-6 and C18:3r w-3) or, if the fetus is dependent upon mechanisms of placental synthesis and transfer in order to obtain longer chain polyenoic fatty acids required for synthesis of brain structural lipids. During the third trimester of human development, rapid synthesis of brain tissue occurs [4] in association with increasing neuromotor activity. Cell size, type and number increase, necessitating de novo synthesis of structural lipids by the developing fetus. Quantitative fatty acid analysis of brain throughout this period of organogenesis has not been reported, and thus reliable determination of accretion rates for each fatty acid constituent in normally nourished human brain has not been made. Clearly, further quantitative understanding of this question is particularly relevant to feeding of low birth weight (< 1500 g) premature infants fed humanized formulas or by intravenous routes particularly, as these formulations contain no C&, w-6, C,,:,, w-3, C 22:6, w-3 and varied levels of other w -3 fatty acids normally present in breast milk [ 31. Therefore, the current study was initiated to quantitate the fatty acid constituents present in human brain from 22 weeks of gestation to term. Regression analysis was applied to determine timing and accretion rates for all fatty acids quantitated.
MATERIALS AND METHODS
Sample population. All aspects of this study protocol have been ethically approved by the Human Experimentation Review Committee of the Hospital for Sick Children and the Department of Pediatrics, University of Toronto. Infant tissues were obtained during autopsy from a refrigerated cadaver within 16 hours post mortum. Ten male and 5 female infants were sampled from 22 weeks of gestational age (by dates, cf. ref. 6) to term. These infants died within 3 days of birth and were appropriate for gestational age (AGA) for birth weight, length and head circumference, with the exception of one infant, who was small for gestational age (SGA). Causes of death, as determined by clinical diagnosis and autopsy, were intrapartum asphyxia (N = 9), prematurity combined with intracerebral hemorrhage or asphyxia (N = 4),
123
congenital heart disease (N = 1) and diaphragmatic hernia (N = 1). Infants were therefore apparently normally nourished in utero, with no evidence of placental infarcts or maternal toxemia. Fresh cross-sectional samples (1 g) of cerebellum and the tip of the frontal and occipital brain lobes were placed in screw cap vials. Tissue samples were frozen and placed in a low temperature freezer (-60 “C) until extraction. Extraction and fatty acid analyses. Tissue lipids were quantitatively extracted under nitrogen with 50 vols of chloroform:methanol (2:1, v/v) utilizing a Folch extraction apparatus [ ‘7,19 ] and n-heptadecanoic acid as an internal standard. Fatty acid methyl esters were prepared with boron trifluoride methanol reagent [ 121. Methyl esters were quantitated by gas liquid chromatography in a dual column gas chromatography, equipped with flame ionization detectors and a chromatography data system for semi-automated operation (Varian model 3700 GLC with a CDS 111). Three-meter glass columns (3 mm I.D.), packed with silar-5CP (10% w/w) coated on acidwashed chromosorb W (80-100 mesh), were utilized. The GLC was operated utilizing temperature programming from 210 “C at 0.5 “C per min for 45 minutes with subsequent isothermal operation of the oven. Identification of methyl esters was accomplished by comparison of retention data with those of known standards, and by the method of ‘equivalent chain length’ [13]. Tissue fatty acid content was calculated on a pg of fatty acid per gram wet weight of tissue basis. Statistical analysis. Regression analysis of fatty acid composition was performed by computer and plotted for infants from 27 weeks of gestational age to term (N = 14). Accretion of fatty acids was determined by a general linear models procedure fitting a third-degree polynomial equation in the regression analysis. Degrees of freedom, sample correlation coefficients, significance levels and polynomial functions are indicated (Figs. l-4). Significant values of r were determined by reference to Steel and Torrie [ 151. Weekly accretion rates for each fatty acid were calculated by solving the regression equation. RESULTS
Brain weight increased some 4-5-fold during the last trimester of fetal development (Fig. 1). Individual cerebellum weights were not measured;
. .
25
30
35
40
4.5
Gestational Age (weeks)
Fig. 1. Brain weight as a function of gestational age. Brain weight (8) = 1050 - 111x + 3.glx2 -0.039x’, where x = gestational age in weeks. *** = P < 0.001, df = 14.
124
Past Conceptional Age (weeks)
Fig. 2. Cerebellum weight (% of whole brain weight) as a function of postconceptional age. Cerebellum weight data were abstracted from ref. 14. Cerebellum weight (W of whole brain weight) = 0.9815 + 0.1274 x, where x = postconceptional age in weeks. ** = P i
0.01. therefore, previously reported data for cerebellum weight, expressed as a percentage of whole brain weight [l] , was utilized to estimate total cerebellum weight for each brain sampled. The relationship derived (Fig. 2) was utilized to calculate total cerebellum fatty acid content (Fig. 4). Comparison of quantitative fatty acid analyses of frontal and occipital brain lobes by analysis of variance indicated no significant difference between these brain regions for fatty acid content per gram of tissue. Therefore, in subsequent regression analysis, values obtained for these two brain regions were averaged as representative of the fatty acid composition for each individual whole brain. Fatty acid content per gram of tissue also reflected total fatty acid content on a whole brain basis. During the last trimester, whole brain levels of C18:2, w -6 and C&, w -3 were low relative to other fatty acids with no significant accretion noted (Fig. 3A and B, respectively). However, marked accretion of products of chain elongation-desaturation, such as C&0:4, w-6 and C,,:,, w-3 occurred (Fig. 3C and D, respectively) resulting in accretion of w-6 and w-3 fatty acids (Fig. 3E and F, respectively) some 50-80-fold greater than for the essential precursors C!,,:,, w-6 and C,,:,, w -3. Accretion of w -9, saturated and total fatty acids also occurred during the last trimester (Fig. 3G, H and I, respectively) reflecting changes in brain weight (Fig. 1) and presumably myelination. Growing cerebellum contained remarkably lower levels of C1sL2, w-6 and ClsZ3, w-3 relative to levels of longer chain homologues (Fig. 4A-E). During the last trimester of cerebellum development accretion of all fatty acids reflected the accretion of total fatty acids in cerebellum (Fig. 4A, C, D, F, G and H) with the notable exception of the w -3 fatty acids (Fig. 4B and E). Accretion rates for C18:3, w-3 remained low in cerebellum during the normal gestational period (Fig. 4B and Table I). By solving the regression equations for fatty acid accretion at each week of gestation during the third trimester, mean accretion rates for each fatty acid can be computed (Table I). The sum of these means plus two standard errors provide estimates of the typical requirement for each fatty acid normally utilized per week for intrauterine synthesis of brain and cerebellum tissues.
125
Gestational Age (weeks) Fig. 3. Fatty acid content of infant brain expressed as a function of gestational age. Values of r and significance levels are indicated (df = 13). Totals for each fatty acid (mg) are computed where x = gestational age (weeks). A: total C,,:,, w-6 = -606.2 + 48.81x 1.269x’ + 0.0108x”. * = P < 0.05. B: total C,,:,, w-3 = 109.4 - 8.008~ + 0.1916xz 0.0014~~. C: total C,,:,, w-6 = 7049 - 599.4x + 16.59x1 - 0.145x’. *** = P < 0.001. D: total C,,;,, w-6 = 4621 + 410.9x - 12.05x’ + 0.1191x3. ** = P < 0.01. E: total w-6 fatty acids = 6269 - 521.6x + 14.09x’ - 0.1156x3. *** = P < 0.001. F: total w-3 fatty acids = - 9842 + 855.9x - 24.43~~ + 0.2321~‘. ** = P < 0.01. G: total w-9 fatty acids = 2983 - 255.9x + 7.134~~ - 0.0571~). ** = P < 0.01. H: total saturated fatty acids = 20,270 - 1739x + 48.73x z - 0.4188x”. ** = P < 0.01. I: total fatty acids = 34,060 -2998x + 86.34x’ - 0.7632~“. ** = P < 0.01.
1
126 TABLE I Accretion rates for fatty acids in infant brain and cerebellum Fatty acids
C Is:,> C,,:,, C,,:,, Total Total Total Total Total
w-6 w-3 w -6 w -6 w -3 w -9 saturates fatty acids**
Accretion rate (mg fatty acid/week)* Whole brain
Cerebellum
0.457 0.132 18.8 31.3 14.5 28.8 101 181
0.037 * 0.012 0.026 + 0.036 1.59 t 0.558 1.53 L 0.252 0.127 2 0.238 2.41 f 0.994 6.1 + 2.22 11.5 * 3.4 ~~___~ _ _ _ _ _ _ _ _
t 0.622 _+ 0.194 * 7.24 f 9.64 t 7.34 t 6.16 + 24.18 I 39.2
*From 26 weeks of gestation to term si + 2 SEZ. **Total fatty acids were calculated from individual tissue analyses.
Gestat~onal
Age
(weeks)
DISCUSSION
Development of infant brain occurs in association with a rapid rise in cellularity and declining water content 151. In the current study, this increase in cellular membranes is quantitatively reflected by increases in long-chain polyenoic fatty acid content in both brain and cerebellum. Absolute accretion rates of w -3 fatty acids, particularly C&, w-3, were greater in the antenatal period (Table I) relative to postnatal rates (to be published), suggesting that accretion of long-chain w-3 fatty acids during the third trimester may be particularly important in the normal rise in brain cellularity during this period of development. In the animal model specific fetal long-chain polyenoic fatty acids originate primarily from synthesis in maternal liver [14]. In the human, placental synthesis of CZoZ4, w -6 occurs [ 231 with concomitant fetal accretion of both o -6 and w-3 chain elongation-desaturation products (Figs. 3 and 4). However, it is not known whether these long-chain polyenoic components accrue as a result of placental transfer or fetal synthesis. In this regard, if microsomal chain elongation+esaturation of C!,,:,, w-6 and C&, w-3 occurs in infant brain or liver by enzyme systems analogous to those of mouse brain where at least 3 separate acyl-CoA elongating enzymes are involved [ 81, then developmental examination of chain elongating enzymes could have implications for the nutritional support of the low birth weight infant with fatty acids. Quantitation of the essential fatty acid composition of cerebrum and cerebellum clearly indicates how vulnerable the very low birth weight infant is in respect to essential fatty acid deficiency at birth. A premature AGA infant of 1300 g born at 30 weeks of gestation has a brain weight of = 170 g and cerebellum weight of == 8.5 g (Figs. 1 and 2). According to our present results the total w -6 and w -3 fatty acid content of these two organs is = 300 mg (Figs. 3 and 4). Analysis of the chemical composition of the developing fetus indicates that the total fat content of such an infant is approximately 28 g [20,21], and the estimated essential fatty acid content of adipose tissue is 1.8% [2,22], i.e. some 0.5 g. Thus, it is apparent that the reserve of essential fatty acids in this prematurely born infant is marginally greater than the essential fatty acid content and accretion rate in brain. Consequently, these vital structural components of brain can be provided from endogenous sources for only a very limited period. Fig. 4. Fatty acid content of infant cerebellum as a function of gestational age. Values of r and significance levels are indicated (df = 13). Totals for each fatty acid (mg) are como u t e d w h e r e Y = eestational see Iweeks). A: total C...,. w-6 = 14.67 - 1.296x + :-51 + 4.394x - 0.1238;’ + b.0372 x1 - 0:003x;. * = P < rj.O$. B: t&al C,,:,, w-3’ 0.0011~‘. * = P < 0.05. C: total C,,.., w-6 = -466.9+41.54x - 1.228~’ + 0.0123x”. ** = P < 0.01. D: total w-6 fatty acids = 208.7 - 20.11~ + 0.6224~’ -0.0058~‘. * = I’ < 0.05. E: total w-3 fatty acids = 157.4 - 15.21~ + 0.4827~~ -0.0049x’ . F: total w-9 fatty acids = -865.9 + 77.54~ - 2.301~’ + 0.0229x3. * = P < 0.06. G: total saturated fatty acids = -1842 + 164.7~ -4.879x’ + 0.0488~‘. ** = P < 0.01. H: total fatty acids = -2699 + 239.4x -7.039x1 + 0.0707~~. ** = P < 0.01.
128
The physician faces a quite different situation in the nutritional management of a more mature neonate. For example, if an infant is born after 35 weeks of gestation with a body weight of 2200 g (AGA) the essential fatty acid content can be similarly estimated at = 2.5 g of w -6 and w -3 fatty acids. At this developmental age the essential fatty acid content of the brain is = 0.45 g. Consequently, this infant is less at risk of essential fatty acid deficiency, and when under nutritional stress would be expected to maintain an essential fatty acid supply from endogenous sources for continued development of the central nervous system for a longer period of time. In emergency situations when only fat free intravenous alimentation is indicated, this knowledge of minimal essential fatty acid requirements necessary to preserve normal development of the central nervous system may prove invaluable. Furthermore, since growth and composition of tissues after birth are influenced by manipulations of dietary fatty acids, the present results suggest guide-lines for ‘natural’ or close to ‘optimal’ nutritional support with essential fatty acids for the premature human neonate. Current studies of energy metabolism in low birth weight infants fed mothers’ own milk indicate [17] that, for net energy intakes of 120 kcal/kg/ day, some 20 k&/kg/day of fat fed will be deposited in tissues (2.2 g/kg/ day). Based on quantitative fatty acid analysis of human milk [3] and third-trimester accretion rates (Fig. 3), it can be estimated that a significant quantity of w6 and o-3 fatty acids provided for tissue deposition at this level of net energy intake would be utilized in brain synthesis if postnatal accretion of these components is to occur in low birth weight infants (< 1500 g) at rates analogous to intrauterine accretion rates (Table I). Such calculations regarding utilization of fatty acids in tissue synthesis do not wholly consider tissue synthesis costs; however, they do indicate that a significant proportion of long-chain polyenoic fatty acids present in human milk may be of particular importance to nutriture of low birth weight neonates if brain synthesis and accretion of fatty acids in brain is to occur at rates analogous to those in the normally growing fetus.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the technical assistance of Mrs. S. Salciccioli. This study was supported by The Hospital for Sick Children Foundation and H.J. Heinz Foundation.
REFERENCES 1 Btinkov, S.M. and Gleeer, I. (1968): In: The Human Brain in Figures and Tables; A Quantitative Handbook, pp. 334-335 and 338-339. Basic Books, Plenum Press, New York. 2 Boersma, E.R. (1979): Changes in fatty acid composition of body fat before and after birth in Tanzania: an international comparative study. Br. Med. J., 1.850-853.
129
3 Chappell, J.E., Clandinin, M.T., Heim, T., Swyer, P.R. and Chance, G.W. (1980): Levels of long chain polyenoic fatty acids in early human milk, to be submitted. 4 Dobbing, J. (1972): Vulnerable periods of brain development. In: Lipids, Mainutrition and the Developing Brain, pp. 9-29. Ciba Foundation Symposia. Editors: K. Elliot and J. Knight. Elsevier, Amsterdam. 5 Dobbing, J. and Sands, J. (1973): The quantitative growth and development of human brain. Arch. Dis. Child., 48, 757-767. 6 Dubowitz, L.M.S., Dubowitz, V. and Goldberg, G. (1970): Clinical assessment of gestational age in the newborn infant. J. Pediatr., 77, l-10. 7 Folch, J., Lees, M. and Sloane-Stanley, G.H. (1957): A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem., 226, 497507. 8 Goldberg, I., Schechter, I. and Bloch, K. (1973): Fatty acyl-xxnzyme A elongation in brain of normal and quaking mice. Science, 182,497499. 9 Holman, R.T. (1973): Essential fatty acid deficiency in humans. In: Dietary Lipids and Postnatal Development, pp. 127-143. Editors: C. GaIli, G. Jacini and A. Pecile. Raven Press, New York. 10 Lamptey, M.S. and Walker, B.L. (1978): Physical and neurological development of the progeny of female rats fed an essential fatty acid-deficient diet during pregnancy and/orlactation. J. Nutr., 108,351-357. 11 Lamptey, MS. and Walker, B.L. (1978): Learning behaviour and brain lipid composition in rats subjected to essential fatty acid deficiency during gestation, lactation and growth. J. Nutr., 108,358-367. 12 Metcaife, L.D. and Schmitz, A.A. (1961): The rapid preparation of fatty acid esters for gas chromatographic analysis. Analyt. Chem., 33,363-364. 13 Miwa, T.K., Mikolajaczak, K.L., Earle, F.R. and Wolff, J.A. (1960): Gas chromatographic characterization of fatty acids. Identification of constants for mono and dicarboxylic methyl esters. Analyt. Chem., 32, 1739-1742. 14 Pascaud, M., Phan, H. and Renard, J.L. (1979): Transfert materno-foetai et captation des acides gras essentiels chez le rat. Ann. Biol. Anim. B&hem. Biophys., 19, 251256. 15 Steel. R.G.D. and Torrie. J.H. (1960): Principles and Procedures of Statistics. MC Graw-Hill, N e w Y o r k . 16 Tinoco, J., Babcock, R., Hincenbergs, I., Medwadowski, B., Miljanich,P. and Williams, M.A. (1979): Linolenic acid deficiency. Lipids, 14,166-173. 17 Verellen, G., Heim, T., Smith, J.M., Swyer, P., Atkinson, S.A. and Anderson, G.H. (1979): Partition of metabolisahle energy (ME) in AGA premature infants < 1300 g birth weight, age 14 weeks. Pediatr. Res., 13,409 (Ahstr.). 18 Walker, B.L. (1967): Maternal diet and brain fatty acids in young rats. Lipids, 2, 497600. 19 Walker, B.L. (1974): Convenient apparatus for extraction of tissue lipids. Lipids, 9, 938-940. 20 Widdowson, E.M. (1968): Growth and composition of the human foetus and newborn. In: Biology of Gestation, Vol. 2, pp. 148. Editor: N.S. Assah. Academic Press, New York. 21 Widdowson, E.M. and Dickerson, J.W.T. (1960): The effect of growth and function on the chemical composition of soft tissue. B&hem. J., 77, 30-33. 22 Widdowson, E.M., Dauncey, M.J., Gairdner, D.M.T., Jonxis, J.H.P. and PelikanFilipkora, M. (1975): Body fat of British and Dutch Infants. Br. Med. J., 1, 653-655. 23 Zimmermann, T., Winkler, L., Moller, U., Schubert, H. and Goetse, E. (1979): Synthesis of arachidonic acid in the human placenta in vitro. Biol. Neon., 36, 209-212.