Abstract:
57 Prenatal nutritional influences on growth and development of ruminants P.L. Greenwood1 and A.W. Bell2 1 2 NSW Agriculture Beef Industry Centre, University of New England, Armidale NSW 2351 Department of Animal Science, Cornell University, Ithaca NY 14853_4801, USA paul.greenwood@agric.nsw.gov.au Summary Intrauterine growth retardation (IUGR) results from inadequate nutrient supply to the foetus due to placental insufficiency and/or low maternal circulating metabolite concentrations during late gestation if nutrient intake and body reserves are limited. In our recent study of postnatal consequences of IUGR, growthretarded newborn lambs tended to be hypoglycaemic and exhibited sluggish postnatal engagement of the growth hormone (GH)/insulinlike growth factor (IGF) system. When artificially reared in an optimum environment, low birth weight lambs grew at rates matching those of normal lambs. However, increased fatness at any given weight resulted, apparently related to high energ y intakes soon after birth, low maintenance energ y requirements, and limited capacity for bone and muscle growth. These growth characteristics were accompanied by higher plasma levels of GH and leptin, and lower levels of IGF1 during the first week or two of postnatal life, and higher levels of insulin during subsequent growth to 20 kg. Emerging evidence suggests that in sheep, as in rodents, foetal programming of postnatal cardiovascular and metabolic dysfunctions is associated with IUGR and may be mediated partly by foetal overexposure to cortisol. Similar postnatal responses can be elicited by maternal undernutrition or cortisol treatment in earlymid pregnancy without changes in foetal or placental growth. Keywords: sheep, cattle, foetus, placenta, birth weight, nutrition retardation (IUGR). Excessive foetal growth due to maternal nutrition also increases perinatal mortality. Emerging evidence suggests that foetal metabolic disturbance can lead to programming of increased predisposition to various disease syndromes during later postnatal life, possibly with ramifications for longterm health and productivity of livestock. This review focuses on the causes and consequences of IUGR in ruminant livestock. Our own recent studies on early postnatal metabolic development and capacity for growth of key tissues in lambs suffering severe, natural IUGR are summarized. Other, recent studies on incipient or actual pathophysiological consequences of prenatal nutritional insufficiency and IUGR in neonatal and older sheep, which include experimental evidence for the concept of foetal programming, are also discussed. Normal conceptus metabolism and growth Patterns of prenatal growth Patterns of foetal and placental growth in the normal and growthretarded sheep conceptus are illustrated in Figure 1. In sheep, as in other placental mammals, post embryonic growth becomes quantitatively significant only after midgestation. However, this is preceded by rapid hyperplastic growth of the placenta which attains all or most of its mass of dry tissue, protein, and DNA by midgestation (Ehrhardt and Bell 1995). Foetal growth then follows its familiar, flattened sigmoid pattern during the latter half of gestation as it proceeds from an early exponential phase through a rapid, linear phase, and then, as term approaches, begins to diminish in rate. In most species, there is little or no increase in placental weight during this period; the ovine placenta actually diminishes in weight, mostly due to loss of extracellular water (Ehrhardt and Bell 1995). However, the placenta undergoes extensive tissue remodelling after midgestation, including major proliferative growth of the umbilical vasculature (Teasdale 1976), Introduction The Australian environment can result in prolonged periods of severe nutritional constraints for ruminant livestock during gestation and early postnatal life, primarily due to drought and/or regional climatic characteristics. High levels of mortality and morbidity in low birthweight offspring remain a major problem, particularly in sheep flocks, despite decades of research on the multifaceted aetiology of intrauterine growth Recent Advances in Animal Nutrition in Australia, Volume 14 (2003) 58 Greenwood, P.L. and Bell, A.W. which is associated with a progressive increase in its functional capacity. Relations between placental size and function, and implications for foetal growth are discussed below. (Kennaugh et al. 1987). The remaining 40% are rapidly catabolized, accounting for at least 30% of the oxidative requirements in the wellnourished sheep foetus (Faichney and White 1987) or, in the case of glutamate and serine, taken up and metabolised by the placenta (Battaglia and Regnault 2001). Gestational changes in conceptus metabolism The manyfold increase in foetal mass from mid to late gestation is accompanied by increased absolute rates of uterine and umbilical uptake of oxygen and nutrients and of urea export by conceptus tissues, and of foetal wholebody protein synthesis in sheep and cattle (Bell et al. 1986; Reynolds et al. 1986; Kennaugh et al. 1987; Bell et al. 1989; Ferrell 1991). When expressed on a weightspecific basis these rates are considerably greater in mid than in late gestation, concomitant with greater relative rates of growth in the immature foetus. In sheep, the gestational decline in weightspecific foetal wholebody metabolic rates is associated with allometric growth patterns of metabolically active vital organs, such as the liver, versus that of less active skeletal tissues (Bell et al. 1987a), as well as a decline in the weightspecific rate of foetal hepatic oxygen consumption (Vatnick and Bell 1992). Figure 1 Patterns of foetal and placental growth in the normal (--) and growth_retarded (_ _) sheep conceptus. Adapted from the data of Ehrhardt and Bell (1995) and Greenwood et al. (2000a). Reproduced with permission of Elsevier Science B.V., Amsterdam (Bell et al. 2003). Intrauterine growth retardation Placental size and nutrient transport capacity Placental weight and associated capacity for maternal foetal nutrient transfer are powerful determinants of foetal growth during late gestation in all species studied. This has been most persuasively demonstrated by controlled manipulation of placental size and/or functional capacity using premating carunclectomy (Alexander 1964), heatinduced placental stunting (Alexander and Williams 1971), or uteroplacental vascular embolisation (Creasy et al. 1972). Natural variations in foetal weight due to varying litter size in prolific ewes are strongly correlated with placental mass per foetus (Rhind et al. 1980; Greenwood et al. 2000a). Recently, the quite profound growth retardation of foetuses in overfed, primiparous ewes has been attributed to a primary reduction in placental growth (Wallace et al. 2000). Placental weight and birth weight are also highly correlated in cattle (Anthony et al. 1986; Echternkamp 1993; Zhang et al. 1999). The probably common aetiology of IUGR in experimentallyinduced and natural cases of placental insufficiency is illustrated by the similar patterns of association between foetal and placental weights in pregnant ewes with varying conceptus weights due to carunclectomy, heat stress, litter size, and overfeeding of primiparous dams (Figure 2). In each case, severe growth retardation was associated with chronic foetal General features of foetal metabolism and its regulation Foetal macronutrient requirements and metabolism in sheep and cattle have been quantitatively described in terms of umbilical exchanges of oxygen, nutrients, and metabolites (see Bell et al. 2003). During late pregnancy in these species, 3540 % of foetal energy is taken up as glucose and its foetalplacental metabolite lactate, and a further 55% is taken up as free amino acids. In contrast to its importance as an energy source in the maternal ruminant, umbilical uptake of acetate is estimated to account for only 510% of foetal energ y consumption. In ruminants, placental capacity for transfer of longchain, nonesterified fatty acids (NEFA) and ketoacids is even more limited (see Bell and Ehrhardt 2002), making these maternal substrates trivial contributors to foetal metabolism. Almost all of the nitrogen acquired by the foetus is in the form of amino acids, but a small net umbilical uptake of ammonia is derived from placental deamination of amino acids during the latter half of gestation (Holzman et al. 1977; Bell et al. 1989). About 60% of these amino acids are used for tissue protein synthesis, which accounts for ~18% of foetal energy expenditure Prenatal nutritional influences on growth and development of ruminants 59 hypoxaemia and hypoglycaemia during late gestation (Creasy et al. 1972; Harding et al. 1985; Bell et al. 1987b; Wallace et al. 2002). A detailed assessment of influences on placental transport of nutrients is beyond the scope of this review, but is provided in Bell et al. (2003). 4000 3000 Fetal weight, g 2000 1000 0 0 100 200 300 400 500 Placental weight, g Figure 2 Relation between foetal and placental weights in ewes representing different models of placental insufficiency during late pregnancy. Variation in placental weight was achieved by premating carunclectomy ( l; Owens et al. 1986), chronic heat treatment (�; Bell et al. 1987b), natural variation in litter size (p; Greenwood et al. 2000a), and overfeeding of adolescent ewes (r; Wallace et al. 2000). Reproduced with permission from the Society for Reproduction and Fertility (Greenwood and Bell 2003). Maternal nutrition Maternal nutrition influences growth of the foetus and size of the newborn either directly as a result of the adequacy of nutrient intake and circulating substrate concentrations, or indirectly due to effects on the capacity of the placenta to transport nutrients to the foetus. Chronic and acute nutritional effects on foetal growth in sheep occur mainly during the final two months of pregnancy, when foetal nutrient requirements increase rapidly (Wallace 1948; Mellor 1983). These effects can be substantial and have been demonstrated using an in vivo technique to measure curved crown rump length and thoracic girth circumference. Commencing at 112 to 120 days of gestation, acute nutritional restriction of ewes reduced foetal growth rate by 30 to 47% within 3 days, including some foetuses that eventually reached growth stasis (Mellor and Matheson 1979; Mellor and Murray 1981). When ewes that had been severely undernourished for 9 or 16 days were realimented to normal levels, there was an immediate increase in foetal growth rate, but not in foetuses of ewes undernourished for 21 days (Mellor and Murray 1982b). Similarly, foetal intravenous nutritional supplementation overcame experimentally induced foetal growth retardation during late gestation (Charlton and Johengen 1987). Chronic, moderate undernutrition of ewes between 90 d and 140 d of pregnancy resulted in a progressive decline in foetal growth rate, with foetuses of undernourished ewes being 22% lighter than those of wellnourished ewes at 142 d (Mellor and Murray 1982a). The adverse effects of chronic undernutrition throughout pregnancy or of severely restricted nutrition during late gestation on foetal growth and birth weight can be variable. Moderating influences include maternal body condition (McNeill et al. 1999) and plane of nutrition during late pregnancy (Oddy and Holst 1991) when growth of the foetus is normally constrained. Nevertheless, a high plane of nutrition during the final four weeks of pregnancy could not totally compensate for the effect on birth weight of severe undernutrition of ewes from mating to four weeks prepartum, indicating that following prolonged foetal growth retardation, an extended period of nutritional rehabilitation is also required to normalise birth weight (McClymont and Lambourne 1958). Effects of adverse nutrition during early to mid pregnancy on foetal growth (Everitt 1964) or birth weight (Nordby et al. 1987) have been demonstrated. However, these required extremely severe maternal undernutrition, to the extent that some ewes died (Everitt 1965), or prolongation of feed restriction from one month prior to breeding until 100 days of pregnancy (Nordby et al. 1987). In general, maternal nutritional restriction during early (Parr et al. 1986; Krausgrill et al. 1999) and/or mid pregnancy (Oddy and Holst 1991; McCrabb et al. 1992; Fogarty et al. 1992; Cronje and Adams 2002; Jopson et al. 2002) has only small, if a n y, effect on weight of the foetus or newborn if adequate nutriment is restored during the final 2 months or so of pregnancy. In contrast to the limited effects of moderate nutritional restriction during early to mid pregnancy on birth weight, shearing during early to mid gestation enhances mobilisation of maternal body tissues (Jopson et al. 2002) and has increased the birth weight of single or twin lambs by up to 17%, although results have been inconsistent (Morris, et al. 2000; Jopson et al. 2002; Kenyon et al. 2002a; Revell et al. 2000, 2002). This has led to the proposal that the foetal growth response to shearing occurs in ewes that would otherwise give birth to a low birth weight newborn, and the ewe must have adequate maternal reserves and/or be fed adequately to support increased foetal growth (Kenyon et al. 2002b). In cattle, severe nutritional restriction for at least the last half to onethird of pregnancy is required to reduce foetal growth (Holland and Odde 1992). Birth weight was unaffected by nutritional restriction of heifers from mating to 140 days gestation (Cooper et al. 1998) or of mature cows for the second trimester (Freetly et al. 2000). However, significant reductions in birth weight were caused by prolonged underfeeding of heifers from weaning until parturition (Wiltbank 60 Greenwood, P.L. and Bell, A.W. et al. 1965), and underfeeding of heifers and cows during the second and third trimesters (Freetly et al. 2000; Hennessy et al. 2002), or during late pregnancy only (Hight, 1966; Tudor 1972; Bellows and Short 1978; Kroker and Cummins 1979). The effect of nutritional restriction on birth weight was more pronounced in heifers than cows when the period of restriction encompassed mid and late gestation (Hennessy et al. 2002) rather than late gestation only (Tudor 1972). Interestingly, birth weight of calves of Hereford dams sired by doublemuscled Piedmontese bulls was more affected by restricted nutrition during mid and late pregnancy than those sired by Wagyu bulls (Hennessy et al. 2002), indicating that foetal growth capacity can interact with available nutrition in determining whether foetal growth is retarded. When assessed within parity and sirebreed, nutritional restriction resulted in reduced birth weights of Piedmontesesired calves from heifers and cows, but only of Wagyusired calves from heifers. Effects of foetal growth potential, or foetal nutrient demand, on the nutritional reserves of pregnant cows were also evident (Greenwood et al. 2002b). Dams mobilized more muscle to support growth of male compared to female foetuses and tended to mobilize more muscle to support growth of Piedmontesesired compared to the Wagyusired foetuses, while heifers mobilized less fat and muscle to support foetal growth than cows. Because placental growth precedes foetal growth on a weight specific basis (Ehrhardt and Bell 1995), residual effects of nutrition during early and mid pregnancy on subsequent foetal growth may be mediated, at least in part, by effects on placental size. This has stimulated interest in understanding how nutrition may be used during early to mid pregnancy to increase placental capacity for nutrient transport in sheep (Davis et al. 1981; McCrabb et al. 1992; Clarke et al., 1998; Heasman et al. 1998; Cooper et al. 1998; Wallace et al. 1999b) and cattle (Cooper et al. 1998; Perry et al. 1999) prior to the period of maximal foetal growth potential during late pregnancy. However, effects of nutrition on placental growth during early to mid pregnancy are highly variable, and may be influenced by a range of factors that uncouple the normally tight association between placental and foetal weights, including nutritional status of the dam prior to mating (Kelly 1992). For example, evidence supports the proposition that ewe fatness during early gestation influences the placental growth response to nutrition (Bell and Ehrhardt 2000). Ewes that were fatter during early pregnancy responded to restricted nutrition during early to mid pregnancy with increased placental size, while placental size was reduced following undernutrition of ewes that were thinner during early pregnancy (McCrabb et al. 1992). These results appear consistent with retarded placental growth in overfed pregnant adolescent ewes (Wallace et al. 1996, 1999a) being overcome by restricting nutrition of these ewes from 50 to 100 days of pregnancy, during which time enhanced placental growth was associated with maternal body tissue mobilization (Wallace et al. 1999b). Furthermore, ewe body condition during early to mid gestation was inversely related to placental weight when maternal nutrient requirements were met (Greenwood et al. 2000a). Taken overall, the above findings emphasise that practices which affect the dams capacity to mobilize peripheral body tissues or partition nutrients towards the gravid uterus may influence birth weight by altering placental development during early to mid gestation and/ or the supply of nutrients available to the foetus during late gestation. For example, cold stress of ewes during late gestation increased maternal glucose, glycerol and nonesterified fatty acid concentrations and foetal glucose concentration, and increased birth weight by 15% (Thompson et al. 1982). Similarly, following isolation stress of ewes for 1 h on 10 occasions during late pregnancy, birth weight of lambs was increased by 12% (Roussel and Hemsworth 2002). Coordination of foetal metabolism and growth The mechanisms relating nutrient supply to expression of endocrine and local regulatory factors, and thence tissue metabolism and growth, can be illustrated by integration of the present knowledge on IUGR, whether caused by placental insufficiency, maternal undernutrition, or insulininduced maternal hypoglycaemia. Effects on the local expression of trophic factors and the cellular growth of skeletal muscle will serve as an example of tissue responses to an altered extracellular milieu. The putative relationships discussed below are schematically represented in Figure 3. Placental insufficiency during late gestation is generally characterized by foetal hypoxaemia and hypoglycaemia, whether caused by surgical reduction (carunclectomy; Harding et al. 1 985), placental embolisation (Creasy et al. 1972), maternal heat stress (Bell et al. 1987b), or overfeeding of adolescent ewes (Wallace et al. 2002). Associated endocrine changes include decreased foetal plasma concentrations of insulin (Robinson et al. 1980) and IGF1 and 2 (Owens et al. 1994), and increased concentrations of cortisol (Phillips et al. 1996). All of these changes can be elicited by maternal undernutrition or insulininduced hypoglycaemia, implicating foetal glycaemia as an important primary signal (Mellor et al. 1977; Osgerby et al. 2002). However, it must be recognized that hypoxaemia may reinforce these responses through its stimulation of foetal adrenal secretion of cortisol and catecholamines, and the inhibitory influence of the latter on foetal insulin secretion. It seems likely that hypoinsulinaemia is a primary, coordinating mediator of the numerous metabolic and trophic consequences of reduced foetal nutrient supply. Disruption of foetal pancreatic insulin secretion has a potent, negative effect on foetal growth (Fowden 1995), associated with decreased foetal tissue uptake and Prenatal nutritional influences on growth and development of ruminants 61 metabolism of glucose (Fowden and Hay 1988), decreased uptake of amino acids and increased proteolysis (Carver et al. 1997), and reduced circulating levels of IGF1 (Gluckman et al. 1987). However, although circulating IGF1 may be of increasing importance during late gestation, it is likely that local tissue expression and actions of this and other growth factors are more significant mediators of tissue growth responses to altered nutrient supply. For example, foetal muscle strongly expresses IGF1 throughout gestation (Dickson et al. 1991; Lee et al. 1993) and disruption of the IGF1 gene causes lethal abnormalities in muscle development (Liu et al. 1993), consistent with the extensive evidence for the role of IGF1 in regulation of myogenesis (Florini et al. 1996). It therefore seems likely that the reduced mitotic activity of myosatellite cells and growth of skeletal muscle in acutely undernourished or placentally growthretarded sheep foetuses (Greenwood et al. 1999) was mediated, at least partly, by reduced local expression of IGF1, possibly caused by elevated plasma levels of cortisol (Li et al. 2002). Finally, although this section has focused on IUGR to illustrate aspects of the coordination of nutrient supply with growth in the foetus, it should be recognised that even in optimally fed, healthy, animals the foetal growth is constrained by placental capacity for nutrient transfer during late pregnancy. This phenomenon ensures that the unborn animals demands upon its dams nutrient reserves are not excessive, and reduces the possibility of birth injury to itself and its mother. The capacity for increased growth in response to increased nutrient supply was demonstrated by the almost 20% increase in birth weight of singleton lambs that had been infused directly with glucose for the last 30 days of gestation in ewes that were extremely well fed (Stevens et al. 1990). Postnatal consequences of altered conceptus metabolism and growth We recently compared postnatal growth, body composition, tissue development, metabolites and hormones, and gene expression in normally grown and MATERNAL UNDERNUTRITION PLACENTAL INSUFFICIENCY p Foetal glucose p Foetal O2 p Foetal insulin n Foetal catecholamines n Foetal cortisol p Foetal IGF-1 (local and systemic) p Foetal tissue uptake of glucose, AA p Foetal tissue protein synthesis, mitosis, and cell differentiation INTRAUTERINE GROWTH RETARDATION Figure 3 Schematic outline of some important factors linking maternal undernutrition and placental insufficiency to intrauterine growth retardation. Reproduced with permission of Elsevier Science B.V., Amsterdam (Bell et al. 2003). 62 Greenwood, P.L. and Bell, A.W. severely growthretarded male Suffolk x (Finnsheep x Dorset) lambs at birth and during postnatal growth to a nominal live weight (LW) of 20 kg (Greenwood et al. 2002a; Greenwood and Bell 2003). Wellgrown (birth weight >4.3 kg) and growthretarded (birth weight <2.9 kg) lambs were removed from their dams at birth and reared artificially on sheepmilk replacer as described by Greenwood et al. (1998). The following section deals with these and other influences of metabolism and growth during foetal life on postnatal metabolism, growth and body composition, with some emphasis on muscle development and growth. Interactions between prenatal and earlypostnatal development and their potential consequences for longerterm growth and development are also highlighted. Metabolic and endocrine characteristics at birth Transition from prenatal to postnatal life is characterized by an abrupt increase in the supply of nutrients and changes in glucose and lipid metabolism compared to prenatal life (Girard et al. 1992). Associated with this increase is a shift in the quality of the supply of nutrients from primarily g lucose and amino acids, t o less carbohydrate and more fat. Immediately postpartum, catabolism of brown adipose tissue occurs to support thermoregulation of the newborn outside of the uterine environment (Alexander 1979). Plasma glucose concentration in the newborn increases rapidly, and induction of expression of genes for key regulatory enzymes important in gluconeogenesis and lipid metabolism occurs during transition to postnatal life (Girard et al. 1992; 1997). Survival of newborns is inversely related to birth weight, until weights that result in dystocia are reached (Alexander 1974). A major contributing factor to postpartum mortality is inadequate energy stores to meet the requirements of the growthretarded newborn for heat production via shivering in skeletal muscle and nonshivering thermogenesis in brown adipose tissue (Alexander 1979). Data for lambs sampled before feeding and within 2 h of birth are summarised in Table 1 (Greenwood et al. 2002a). The moderately elevated levels of plasma urea nitrogen in growthretarded lambs could have been due to greater rates of amino acid catabolism and/or lesser capacity for renal clearance of urea, both of which are foetal characteristics and might be regarded as signs of immaturity. The small lambs also tended to be more hypoglycaemic than their normal counterparts, possibly extending from the chronic hypoglycaemia that is typical of lategestation foetuses suffering placental insufficiency (Bell et al. 1999). However, the most striking feature of these observations is the apparent immaturity of the somatotropic axis in the growth retarded lambs. This is indicated by very high levels of growth hormone (GH) and low levels of insulinlike growth factor (IGF)1 more reminiscent of the late gestation foetus than of the normal, wellgrown lamb immediately after birth (Gluckman et al. 1999). It is notable that hepatic expression of the gene for the acid labile subunit (ALS), which is GHdependent and is greatly increased at or soon after birth in normal lambs (Rhoads et al. 2000a), was reduced in growthretarded newborn lambs (Rhoads et al. 2000b). An early postnatal reduction in the hepatic synthesis and secretion of ALS would delay the normal postnatal shift in size of circulating IGF complexes from 50 kDa to 150 kDa (Butler and Gluckman 1986) and the consequent major increases in halflife and concentration of circulating IGF1. Other indices of hepatic GH responsiveness, including expression of mRNA for the GH receptor, IGF1, and IGF binding protein (IGFBP)3 were not significantly affected by birth weight (Rhoads et al. 2000b). It is notable that reduced hepatic expression of both ALS and IGF1 was discernible as early as 130 d of gestation in growthretarded foetuses, despite the much lower absolute levels of expression of these genes in foetal versus neonatal lambs (Rhoads et al. 2000b). These data are consistent with decreases in foetal plasma IGF1 that were highly correlated with decreases in placental weight and apparent delivery of glucose and Table 1 Plasma concentrations of metabolites and hormones in normally grown and severely growth_retarded newborn lambs. Variable Normally grown (n = 4) 4.89 � 0.21 Growth_retarded (n = 4) 2.24 � 0.26 Significance of difference (P) Birth weight (kg) Plasma concentration Glucose (mmol/L) Urea N (mmol/L) Insulin (�g/L) Growth hormone (�g/L) IGF_1 (�g/L) Leptin (�g/L) 2.63 6.39 0.13 10.8 158 3.8 � � � � � � 0.95 0.32 0.06 4.3 22 0.3 1.42 8.31 0.09 49.1 36 4.1 � � � � � � 0.23 0.25 0.02 17.0 7 0.3 ns <0.01 ns <0.05 <0.001 ns Values are means � SEM; ns, not significant Data from Ehrhardt et al. (2001) and Greenwood et al. (2002a) Prenatal nutritional influences on growth and development of ruminants 63 oxygen in carunclectomized ewes during late pregnancy (Owens et al. 1994), given that, in both cases, foetal growth retardation was due to placental insufficiency. The endocrine mediation of altered development of the GH/IGF system is unclear. A logical candidate for this role might be cortisol, plasma concentration of which is elevated in the placentally retarded foetus (Phillips et al. 1996). However, treatment with cortisol appears to advance rather than retard the development of GHdependent hepatic expression of IGF1 in the late gestation sheep foetus (Fowden et al. 1998). Plasma leptin concentrations were similarly low in small and normally grown newborn lambs (Table 1; Ehrhardt et al. 2001), consistent with their low and similar relative masses of adipose tissue and total body lipid (Greenwood et al. 1998). Postnatal metabolism and growth Most of the data discussed in this section, dealing with effects of size at birth on plasma concentrations of metabolites and hormones in neonatal lambs, are summarized in Table 2 and described in detail elsewhere (Greenwood et al. 2002a). Postnatal changes in superficial indices of carbohydrate and protein metabolism were little affected by birth weight in small and normal lambs that were artificially reared with ad libitum access to milk replacer. The very high concentrations of plasma GH in small, newborn lambs decreased markedly within two days of birth but remained significantly higher than levels in normal lambs for about two weeks. During the same period, plasma IGF1 increased steadily in both groups but remained significantly lower in the small lambs (Greenwood et al. 2002a). These observations suggest that the apparent immaturity of the GH/IGF axis in growthretarded newborn lambs persists for several weeks after birth. Interestingly, only during this early postnatal phase did the absolute growth rates of low birth weight lambs (248 g/d) lag significantly behind those of normal birth weight lambs ( 353 g/d) (Greenwood et al. 1998). Thereafter, during rapid growth from about 2 weeks of age to slaughter at 20 kg (attained at 6.5 to 8 weeks of age), plasma IGF 1 concentrations were persistently higher but GH concentrations were not different in low versus normal birth weight lambs (Table 2). This study did not examine the consequences of low birth weight after weaning. However, plasma GH concentrations tended to be higher during adolescence (~132 days of age) and adulthood (~378 days of age) in low birth weight male lambs from carunclectomized ewes compared to lambs of normal birth weight and were negatively correlated with indices of birth size (Gatford et al. 2002). Plasma insulin concentrations increased rapidly during the early postnatal period in small lambs feeding ad libitum, consistent with their very high levels of energy intake. Then, from about two weeks of age until slaughter at 20 kg, plasma insulin concentrations were persistently higher in low compared with normal birth weight lambs (Table 2). We speculate that this relative hyperinsulinaemia may be due to the predisposition of growthretarded neonates to develop insulin resistance (Hales et al. 1996). Plasma leptin concentrations were somewhat higher in rapidly fattening, low birth weight lambs during the first week post partum, but not thereafter (Ehrhardt et al. 2001), despite the fact that at any subsequent liveweight up to 20 kg these lambs were significantly fatter than their normal birth weight counterparts (Greenwood et al. 1998). These findings also suggest that relatively low levels of plasma leptin during the immediate postpartum period may support high weightspecific levels of feed intake to enable rapid accretion of energy (lipid) stores to enhance later survival. This appears to be particularly relevant to the survival of the very small newborn. The adverse longerterm consequences of greater fatness and a degree of insulin resistance (see Hales et al. 1996) in small newborn animals may, therefore, arise due to adaptations to enhance survival during early Table 2 Plasma concentrations of metabolites and hor