Stress-stunted placentas can typically fulfill the relatively modest fetal nutrient demands in early and mid-gestation, but exponential fetal growth during late gestation pushes O₂ and nutrient requirements beyond the capacity of the stunted placenta (Limesand et al., 2013; Macko et al., 2013). As the fetus continues to grow, nutrient deficits progressively worsen. Natural sheep models for IUGR produce up to 50% reductions in fetal blood glucose near term (Saker et al., 1999; Limesand et al., 2007; Yates et al., 2016; Edwards et al., 2020). Endogenous hepatic glucose production is engaged in the IUGR fetus but only partially compensates for its hypoglycemia (Fowden and Silver, 1995; Limesand et al., 2007; Thorn et al., 2009; Thorn et al., 2013). In response to limited nutrient availability, the fetus engages its own nutrient repartitioning adaptations that prioritize vital nervous and endocrine tissues over others, particularly skeletal muscle (Poudel et al., 2015). Greater circulating lactate concentrations observed when IUGR fetal sheep were experimentally made hyperglycemic were consistent with a shift in muscle glucose metabolism from oxidative phosphorylation to anerobic glycolysis (Thorn et al., 2013; Lacey et al., 2021). Such shift coincides with and is perhaps necessitated by the hypoxemic state of the fetus (Regnault et al., 2003; Brown et al., 2015). However, less oxidative phosphorylation diminishes energy status by reducing production of ATP (Cree-Green et al., 2018). Despite this, IUGR fetuses exhibited greater circulating CO₂ due to compromised placental gas transfer, which can affect acid-base balance (Macko et al., 2013; Lacey et al., 2021). Disruptions in lipid homeostasis included 10%–25% lower circulating cholesterol, which were reported in IUGR human and rodent fetuses at term despite markedly higher precursor concentrations and slower clearance rates (Cadaret et al., 2019a; Pecks et al., 2019). This indicates that deficits were due at least partially to impaired de novo cholesterol synthesis by the IUGR fetal liver rather than solely due to deficient placental transport. Conversely, ∼15% reductions in circulating triglycerides coincided with less placental fatty acid transporter expression and greater disparity in maternofetal triglyceride gradients in IUGR fetal models (Meyer et al., 2010; Cadaret et al., 2019a; Xu et al., 2019). Diminished placental amino acid transporter expression together with less effective Na⁺/K⁺-ATPase support is reflected in circulating amino acid profiles in IUGR fetuses (Stremming et al., 2020; Rosario et al., 2021). Brown et al. (2012) observed reduced placental transport of isoleucine, leucine, phenylalanine, tryptophan, methionine, and tyrosine, which notably lowered fetal circulating arginine and methionine and muscle protein synthesis patterns in IUGR fetal sheep. Interestingly, concentrations of some amino acids were actually increased in these fetuses, which was presumably a byproduct of reduced protein synthesis (Armstrong, 1973). Even electrolyte homeostasis can be disrupted by placental insufficiency, as fetal blood concentrations of Na⁺, K⁺, Ca⁺⁺, and Cl⁻ were increased in IUGR fetal sheep (Jansson et al., 2006; Bacchetta et al., 2009; Beer et al., 2021; Lacey et al., 2021).

Nutrient and O₂ paucities stimulate fetal stress responses that include systemic inflammatory and adrenergic components (Yates et al., 2012a; Berbets et al., 2021). Hypoxemia, and to a lesser extent hypoglycemia, stimulate secretion of the catecholamines norepinephrine and epinephrine from the fetal adrenal medulla (Gardner et al., 2002; Ly et al., 2011; Yates et al., 2012a). Circulating norepinephrine (the primary fetal adrenal catecholamine) was elevated by as much as 8-fold in IUGR fetal sheep, and hypercatecholaminemia was among the earliest indicators of placental insufficiency (Macko et al., 2013; Chang et al., 2019a). Catecholamines are most associated with physiological mechanisms aimed at immediate survival, and sustained exposure of tissues can disrupt β adrenergic programming, the details and implications of which have been reviewed elsewhere (Yates et al., 2011; Posont et al., 2017; Gibbs and Yates, 2021). Perhaps the most consequential effect of fetal hypercatecholaminemia is suppressed insulin activity. Basal circulating insulin concentrations were often modestly decreased in IUGR fetal sheep, but glucose-stimulated insulin secretion was almost completely suppressed (Leos et al., 2010; Macko et al., 2013; Cadaret et al., 2019b). In contrast, maternal nutrient restriction-induced IUGR did not affect insulin concentrations (Edwards et al., 2020), which indicates that hypoxemia and hypercatecholaminemia are the suppressors of insulin secretion. Not surprisingly, circulating IGF-1 concentrations were likewise reduced in IUGR fetal sheep during mid- and late-gestation (Thorn et al., 2009; Macko et al., 2013; Rozance et al., 2018). As hypoxemia stimulates the adrenal medulla, hypoglycemic conditions stimulate the adrenal cortex to secrete cortisol, a steroid stress hormone associated with changes in intermediary metabolism (Thorn et al., 2013; Jonker et al., 2018). This has been documented across many IUGR models and among several species (Sutherland et al., 2012; Li et al., 2013; Thorn et al., 2013), and it may ultimately lead to reduced sensitivity of the cortisol axis (Iwata et al., 2019). Despite the anti-inflammatory effects of cortisol, IUGR fetuses exhibited elevated circulating inflammatory cytokines, including tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), and IL-6 (Cadaret et al., 2019b; Zhang et al., 2021), and lower concentrations of anti-inflammatory IL-10 and IL-12 (Huang et al., 2019). Cytokines mediate inflammatory responses to pathogens, reactive oxygen species, and toxins, but they are also responsive to hypoxemia and other physiological stressors (Lacy and Stow, 2011; Hicks and Yates, 2021). They are released in greatest volume from circulating white blood cells and resident tissue macrophages and mast cells, and they act on tissues throughout the body (Tracey and Cerami, 1993; Lacy and Stow, 2011). Elevated inflammatory cytokines in IUGR fetuses can increase catabolism of skeletal muscle and affect mobilization of nutrient stores (Hicks and Yates, 2021). The concurrent appearance of systemic inflammation and hypercatecholaminemia in near-term IUGR fetuses is somewhat paradoxical, as macrophagic release of TNFα and IL-6 is suppressed by adrenergic stimulation under normal conditions (Donnelly et al., 2010; Papandreou et al., 2016). Nonetheless, the combined heightened inflammatory and adrenergic tones mediate many of the changes in metabolism and growth that IUGR fetuses exhibit (Macko et al., 2013; Hicks and Yates, 2021).

The bodyweights of IUGR fetal sheep and pigs were reduced by as much as 55% near term (Tree et al., 2016; Cadaret et al., 2019c; Tang and Xiong, 2022), and IUGR-born offspring remained lighter well into the neonatal period (Gibbs et al., 2020; Shoji et al., 2020). However, growth restriction is not equivalent among all fetal tissues. In fact, sustained nutrient insufficiency and the resulting fetal stress response yields hallmark asymmetric growth by disproportionally slowing muscle accretion relative to cranial and skeletal growth (Lapillonne et al., 1997). This was reflected in reduced muscle protein accretion rates, muscle morphometrics, and body length-to-mass ratios of IUGR fetal sheep (Yates et al., 2016; Rozance et al., 2018; Cadaret et al., 2019b). In human fetuses for whom adiposity is naturally high, disproportionally slower fat deposition was also apparent from ultrasound diagnosis of IUGR (Padoan et al., 2004; Ikenoue et al., 2021). The fetal hypoxemia-hypercatecholaminemia-hypoinsulinemia cascade appears instrumental in muscle growth restriction, as experimental induction of each factor individually produced some degree of IUGR in sheep (Philipps et al., 1991; Bassett and Hanson, 1998; Rozance et al., 2018). Hypoxemia-induced inflammation also limited muscle mass by increasing protein catabolism, reducing protein accretion, and altering muscle stem cell function (Li W. et al., 2009; Wang et al., 2014; Madaro et al., 2018; Posont et al., 2022). Less severe reductions in body length, head circumference, and cannon bone length observed in IUGR fetal lambs reflected the modest nature of structural growth restriction (Rozance et al., 2018; Cadaret et al., 2019c; Sandoval et al., 2020). As with bodyweight, however, these mild deficits in structural metrics persisted postnatal (Gibbs et al., 2020; Shoji et al., 2020). Ultimately, most IUGR-born offspring undergo a period of catch-up growth that diminishes or eliminates their weight disparity (Dulloo et al., 2006). However, this compensatory weight gain results from greater fat accumulation and does not reflect recovery of muscle mass (Ong et al., 2000; Gibbs et al., 2020). Consequently, IUGR-born adolescents and adults are more likely to develop high body mass indices (Fagerberg et al., 2004; Zinkhan et al., 2018). For humans, altered body composition was associated with a higher risk for metabolic and cardiovascular dysfunction later in life (Ong et al., 2000; Fagerberg et al., 2004). In food animals, IUGR-altered body composition resulted in smaller and less valuable carcasses (Greenwood et al., 2005; Greenwood and Bell, 2019).

Preferential reappropriation of fetal nutrients from muscle and other peripheral soft tissues to vital brain and endocrine tissues is reflected in blood flow patterns. In IUGR fetal sheep, blood flow was maintained or increased in every region of the brain, and pancreatic and adrenal blood flow increased by up to 2-fold (Poudel et al., 2015). Conversely, blood flow to the hindlimb, which is about 45% skeletal muscle (Hicks et al., 2021), was reduced by half (Poudel et al., 2015; Hicks et al., 2021). In a baboon model of IUGR, less hindlimb blood flow was associated with decreased size and distensibility of the external iliac and femoral arteries, which was sustained into adulthood (Kuo et al., 2018). In contrast, lower ultrasound-estimated resistance in the carotid artery of human IUGR fetuses reflected the brain-sparing characteristic of asymmetric growth restriction (Groenenberg et al., 1989). Preferential delivery of blood and nutrients maintained brain weights across sheep models for IUGR (Yates et al., 2016; Rozance et al., 2018; Wallace et al., 2020; Posont et al., 2021) even when heart, lung, and liver weights were reduced (Hyatt et al., 2007; Cadaret et al., 2019b; Wallace et al., 2020). Not surprisingly, weights of individual hindlimb muscles relative to fetal weight or hindlimb length were reduced in IUGR fetal sheep by up to 40% (Rozance et al., 2018; Lacey et al., 2021).

Impaired myoblast function and slower protein synthesis each contributed to reduced IUGR muscle mass (Yates et al., 2014; Soto et al., 2017; Rozance et al., 2018; Felicioni et al., 2020), which persisted after birth (De Blasio et al., 2007a; Gosby et al., 2009; Gibbs et al., 2020). Myoblasts are the myogenic stem cells that facilitate muscle fiber hypertrophy in late gestation and after birth when fiber numbers have become largely static (Maier et al., 1992; Wilson et al., 1992). Myoblasts undergo rate-limiting proliferation and differentiation steps before fusing with existing muscle fibers to increase the myonuclear content/protein capacity of the fiber (Allen et al., 1979; Yates et al., 2012b). Histological and ex vivo assessments indicated that differentiation capacity was impaired in myoblasts from IUGR fetal lambs (Yates et al., 2014; Yates et al., 2016; Soto et al., 2017; Posont et al., 2022). Myoblast proliferation was also typically reduced by IUGR intrauterine conditions, although proliferation was increased in one model induced by maternofetal inflammation (Soto et al., 2017; Cadaret et al., 2019a; Posont et al., 2022). Myoblast dysfunction is reflected in expression of key myogenic transcription factors, which hint at mechanisms for IUGR myoblast deficits. Specifically, differentiation factors myoD and myogenin were reduced in IUGR fetal myoblasts near term, whether assessed in vivo or ex vivo (Yates et al., 2014; Chang et al., 2019b; Rozance et al., 2021; Posont et al., 2022). Reduced myoblast proliferation, differentiation, and fusion slowed myonuclear accumulation in all muscle fiber types, which in IUGR fetal lambs restricted cross-sectional fiber area by as much as 60% (Yates et al., 2016; Chang et al., 2019b). Smaller muscle fibers have also been documented in IUGR piglets and rats (Cadaret et al., 2019a; Felicioni et al., 2020). Although some variation exists among IUGR models, species, and even specific muscles, disproportional reduction of muscle and other soft tissues as a means of sparing brain and structural growth is a well-conserved phenotype. Table 1 illustrates the distinct size reductions of specific tissues and organs that yield asymmetric body composition.

Limiting skeletal muscle growth contributes to fetal nutrient sparing necessitated by placental insufficiency (Brown, 2014). Additionally, the IUGR fetus alters tissue-specific utilization, metabolism, and storage of nutrients (Rozance and Wolfsdorf, 2019; Yates et al., 2019; Zhang et al., 2021). As with growth restriction, these changes disproportionally affect muscle. Several IUGR fetal sheep models found that muscle-specific glucose oxidation rates were reduced by up to 80% near term, which along with reduced hepatic oxidative metabolism produced a ∼50% reduction in whole-fetus glucose oxidation (Peterside et al., 2003; Limesand et al., 2007; Brown et al., 2015; Cadaret et al., 2019b). This deficit was observed under normal resting conditions and experimental hyperinsulinemia, and it persisted well after birth (Gibbs et al., 2021; Posont et al., 2021; Cadaret et al., 2022). Impaired glucose oxidation also occurred despite normal rates of glucose uptake by muscle, which has been observed before (Rozance et al., 2018; Cadaret et al., 2019c) and after birth (Gibbs et al., 2021; Posont et al., 2021; Cadaret et al., 2022). Although IUGR muscle expressed less of the insulin-dependent glucose transporter Glut4 in some studies (De Blasio et al., 2012; Duan et al., 2016; Yates et al., 2019; Jones et al., 2022), glucose uptake may have been rescued by an increase in the insulin-independent transporter Glut1 (Brown et al., 2015; Yates et al., 2019). Moreover, it should be noted that some studies found no changes in IUGR muscle content of Glut1 or Glut4 (Limesand et al., 2007; Garg et al., 2009). Not surprisingly, Glut1 expression in the brain was increased by over 60% in IUGR fetal lambs and rats as well as in IUGR-born neonatal rats (Sadiq et al., 1999; Limesand et al., 2007), which is consistent with brain sparing during chronic hypoglycemia (Gibbs and Yates, 2021).

Reduced skeletal muscle glucose oxidation rates coincided with lower proportions of slow oxidative (type I) fibers relative to intermediate (type IIa) and fast glycolytic (type IIx) fibers in hindlimb muscles of IUGR fetal sheep and loin muscles of IUGR fetal pigs (Wang et al., 2013; Yates et al., 2016; Stremming et al., 2022). Furthermore, IUGR muscle exhibited reduced activity of pyruvate dehydrogenase and citrate synthase enzymes that generate Krebs cycle intermediates, as well as increased expression of (inhibitory) pyruvate dehydrogenase kinase and impaired function of Electron Transport Chain Complex I (Brown et al., 2015; Pendleton et al., 2019; Stremming et al., 2022). These observations, combined with reduced O₂ utilization, normal glucose utilization, and greater lactate dehydrogenase B content (Brown et al., 2015; Pendleton et al., 2019), indicated that much of the reduction in glucose oxidative phosphorylation was replaced by greater glycolytic lactate production. This is evident in elevated circulating lactate concentrations, which were as much as 3-fold greater in IUGR fetuses, particularly during experimental hyperglycemia or hyperinsulinemia (Limesand et al., 2007; Davis et al., 2021; Camacho et al., 2022). Lactate-O₂ quotient was also 2-fold greater, meaning that more lactate was produced from each mole of O₂ consumed (Rozance et al., 2018). Like other IUGR pathologies, hyperlactatemia appears to be driven primarily by hypoxemia-induced hypercatecholaminemia. In fetal sheep made experimentally hypoxemic (but not hypoglycemic) for 9 days, the 3-fold increase in circulating norepinephrine resulted in 20% less glucose oxidation, 3.5-fold greater circulating lactate, and 2-fold greater fetal lactate production (Jones et al., 2022). This model also demonstrated that hyperlactatemia may indicate condition severity, as circulating lactate concentrations were elevated by robust fetal hypercatecholaminemia (3-fold or greater increase) but not by a more modest 1.4-fold increase in norepinephrine (Jones et al., 2019; Rozance et al., 2021; Jones et al., 2022). Catecholamine-induced hyperlactatemia also persisted in IUGR-born neonates (Chen et al., 2010). Some studies have reported normal blood lactate in IUGR fetuses (Wai et al., 2018; Cadaret et al., 2019b; Pendleton et al., 2020), but this does not necessarily mean that lactate was produced at normal rates. Rather, IUGR fetuses were shown to engage hepatic gluconeogenesis that converts lactate and other substrates to glucose, a mechanism that is largely idle under normal intrauterine conditions (Limesand et al., 2007). This was reflected in elevated expression of gluconeogenic enzymes such as phosphoenolpyruvate carboxykinase and glucose 6-phosphatase and their transcriptional promoters in IUGR sheep, pigs, and rodents in late gestation and after birth (Peterside et al., 2003; Vuguin et al., 2004; Limesand et al., 2007; Thorn et al., 2009; Brown et al., 2015; Ying et al., 2017). Hepatic gluconeogenesis allows the IUGR fetus to utilize greater skeletal muscle lactate production to partially offset poor glucose supply via the Cori cycle (Soeters et al., 2021).

The IUGR fetus’s diminished protein supply affects utilization of amino acids for tissue accretion and for energy production. Foundational work by researchers at the University of Colorado School of Medicine found that hindlimb protein accretion was reduced by 55% in IUGR fetal sheep (Rozance et al., 2018; Wai et al., 2018). This coincided with a comparable reduction in total amino acid uptake rates by the hindlimb, although rates among individual amino acids were not affected uniformly (Rozance et al., 2018). For example, hindlimb uptake rates of the essential branched-chain amino acids leucine, valine, and isoleucine, were reduced by up to 73% in IUGR fetal sheep, whereas alanine, glutamine, and glycine were actually secreted from the hindlimb, despite slightly less protein breakdown (Chang et al., 2019a). These differential fluxes are presumably an attempt at metabolic thrift, as branched-chain amino acids can be converted into alanine, glutamine, and glycine, which are substrates for hepatic gluconeogenesis (Chang et al., 2019a). Indeed, expression of enzymes that facilitate this conversion was greater for IUGR fetal rats and sheep (Kloesz et al., 2001; Chang et al., 2019a). Interestingly, long-term infusion of essential amino acids into IUGR fetal sheep did not consistently increase protein accretion, indicating that amino acid utilization was restricted by more than just the short supply (Wai et al., 2018). Together, these studies show definitively that protein accretion in the IUGR fetus is impeded primarily by slower protein synthesis and not by greater protein breakdown. Like amino acid uptake, the impact of IUGR on individual amino acid oxidation rates is not uniform. For example, IUGR fetal sheep oxidized 47% less threonine (Anderson et al., 1997) but 2-fold more lysine (Limesand et al., 2009). Leucine oxidation rates were reduced by 34% in one early study of IUGR fetal sheep (Ross et al., 1996) but were normal in others (Rozance et al., 2018; Wai et al., 2018). Moreover, leucine oxidation was not affected in fetal sheep made chronically hypoxemic (Rozance et al., 2021), hypoglycemic (Carver et al., 1997), hyperglucagonemic (Cilvik et al., 2021), or hyperinsulinemic (Brown et al., 2016) late in gestation, but it was markedly reduced by elevated circulating IGF-1 (Stremming et al., 2021). A greater amount of leucine was oxidized when normal fetuses were experimentally administered excess amino acids (Rozance et al., 2009; Maliszewski et al., 2012), but this effect was blunted in IUGR fetuses (Brown et al., 2012). Amino acids and glucose are normally competitive oxidative substrates (Brown et al., 2017). However, the apparent lack of compensatory amino acid oxidation in IUGR fetuses despite substantial deficits in glucose oxidation may have been the result of mitochondrial deficits. Indeed, IUGR fetal skeletal muscle had less citrate synthase, which is an indicator of intact and functional mitochondria (Stremming et al., 2022). It also expressed less BCAT1 and BCAT2, the transaminase enzymes that convert leucine and isoleucine into the Krebs cycle intermediate succinate, and less ALT, the enzyme that converts alanine to pyruvate (Chang et al., 2019a). The coincident deficits in glucose and amino acid oxidation ultimately diminished ATP content in IUGR fetal muscle (Stremming et al., 2020).

The impact of IUGR on lipid homeostasis is perhaps less clear. Elevated triglycerides and non-esterified (or free) fatty acids (NEFA) in cord blood at birth is a hallmark of IUGR in humans (Youssef et al., 2021; Chassen et al., 2022) and has been used for decades as a clinical indicator of fetal stress (Elphick et al., 1978). High blood lipid concentrations in late gestation and after birth have also been documented in piglets with spontaneous IUGR (i.e., litter runts) (Li et al., 2018) and in several IUGR sheep models (Wallace et al., 2014; Wallace et al., 2020; Posont et al., 2021; Zhang et al., 2021), although we are aware of at least one study that found reduced circulating NEFA in young IUGR-born lambs (Duffield et al., 2009). The impact of IUGR on circulating cholesterol is similarly inconsistent and may depend in part upon the lipoprotein to which cholesterol is bound. For example, Youssef et al. (2021) observed 23% greater total cholesterol but 20% less HDL-C in cord blood from IUGR babies, whereas Pecks et al. (2012) reported reductions in both total cholesterol and HDL-C. Circulating cholesterol was elevated in several IUGR fetal sheep studies and correlated tightly with fetal weights (Meyer et al., 2010; Zywicki et al., 2016; Sun et al., 2018; Zhang et al., 2021). As neonates, however, IUGR-born lambs had normal or even reduced circulating cholesterol, although HDL-C was still elevated (Wallace et al., 2014; Posont et al., 2021). Changes in intracellular lipids within IUGR tissues may also differ between prenatal and postnatal stages. In fetal sheep, for example, IUGR due to nutrient restriction or fetal crowding increased blood NEFA and triglycerides by as much as 50% but reduced liver triglycerides and hepatic lipase, the enzyme responsible for their hydrolysis (Meyer et al., 2010; Zywicki et al., 2016; Sun et al., 2018; Zhang et al., 2021). In IUGR fetal guinea pigs, hepatic utilization of palmitate (the most common saturated fatty acid in animals and plants) was reduced by about 50% (Detmer et al., 1992). Conversely, IUGR-born neonatal goats had greater liver accumulation of NEFA, triglycerides, and cholesterol (Liu et al., 2022), and IUGR-born runt piglets had more expansive hepatic lipid droplets, particularly after undergoing catch-up growth (Wang et al., 2022). IUGR-induced changes in lipid metabolism were largely tissue-specific, as were changes in de novo synthesis. Shortly after birth, IUGR infants exhibited greater rates of whole-body triglyceride mobilization and fatty acid oxidation (Patel and Kalhan, 1992). Moreover, gene expression for the key β-oxidation facilitator PPARα was increased in liver tissue from IUGR-born runt piglets (Wang et al., 2022) and in cardiac muscle from IUGR-born adult guinea pigs (Botting et al., 2018). Gene expression for CPT1 and HADHA, which play rate-limiting roles for fatty acid β-oxidation, was reduced in liver tissues of IUGR-born rats (Lane et al., 2001a). However, skeletal muscle from these rats had greater CPT1 and HADHA expression along with greater triglyceride content (Lane et al., 2001b). Cardiac muscle from IUGR fetal sheep also expressed less mRNA for CPT1 and for enzymes associated with β-oxidation and esterification of fatty acids (Drake et al., 2022). This coincided with greater circulating acylcarnitines, which are indicative of impaired or incomplete fatty acid oxidation (Drake et al., 2022). Fatty acid synthesis rates were reduced in liver and lung tissues of IUGR rat fetuses, but not in brain tissue (Vileisis et al., 1982). Interestingly, maternofetal O₂ supplementation did not recover fatty acid synthesis in these fetuses, which indicates that the deficit was not a product of hypoxemia (Vileisis, 1985). As adults, fatty acid synthesis was normal in liver and muscle from IUGR-born rats but was elevated in adipose tissues (Yee et al., 2016). Disruptions in lipid homeostasis almost certainly contribute to greater adiposity in IUGR-born offspring (Ong et al., 2000; Zinkhan et al., 2018). Indeed, IUGR-born lambs exhibited markedly greater insulin sensitivity for fat deposition (De Blasio et al., 2007b). They also produced muted increases in circulating NEFA in response to epinephrine challenge, indicating a reduced capacity for lipid mobilization (Harwell et al., 1990; Chen et al., 2010).

Deficient insulin production and secretion is a key contributor to poor growth and metabolic dysfunction in IUGR fetuses and offspring (Limesand et al., 2006; Xing et al., 2019). Large reductions in basal circulating insulin and complete inhibition of nutrient-stimulated insulin secretion observed in IUGR fetal sheep (Limesand et al., 2007; Cadaret et al., 2019b) were the direct result of fetal hypoxemia and hypoglycemia. To illustrate, when normoxemia and euglycemia were experimentally restored for a 5-day period in IUGR fetal sheep, basal insulin concentrations normalized and glucose-stimulated insulin secretion was rescued (Camacho et al., 2022). Conversely, acute or chronic experimental hypoxemia or hypoglycemia in otherwise uncompromised fetal sheep reduced insulin secretion (Rozance et al., 2006; Yates et al., 2012a; Benjamin et al., 2017). Basal hypoinsulinemia was also resolved postnatal, as IUGR-born offspring were no longer hypoxemic or hypoglycemic after birth (Posont et al., 2021; Cadaret et al., 2022). The suppressive effects of these conditions, particularly hypoxemia, on pancreatic β cell function are mediated in large part by adrenergic signaling. In fact, blocking the adrenergic response to IUGR conditions via pharmaceutical antagonists or adrenal demedullation not only rescued insulin secretion, but in most cases revealed a compensatory enhancement in β cell stimulus-secretion coupling (Leos et al., 2010; Yates et al., 2012a; Macko et al., 2013; Macko et al., 2016). Comparable enhancements in β cell function were observed following chronic norepinephrine infusion into otherwise normal fetal sheep or rats (Chen et al., 2014; Chen et al., 2017; Li et al., 2021). Despite greater sensitivity of β cells, IUGR pancreatic islets are poorly developed. Specific islet deficits are reviewed in detail elsewhere (Boehmer et al., 2017) but include smaller size with fewer β cells/islet, less production and sensitivity to growth factors, and poor vascularity (Limesand et al., 2005; Styrud et al., 2005; Limesand et al., 2006; Ham et al., 2009; Rozance et al., 2015).

Inflammatory cytokines are dynamic regulators of myoblast function and thus play a complex role in postnatal muscle growth (Bodell et al., 2009; Alvarez et al., 2020). In vitro studies found that proliferation rates increased when myoblasts were incubated for short periods with low or moderate concentrations of TNFα, IL-1β, or IL-6 (Otis et al., 2014; Alvarez et al., 2020). For TNFα and IL-6 incubations, greater proliferation was coincident with impaired differentiation (Langen et al., 2001; Alvarez et al., 2020; Posont et al., 2022). Indeed, exposure to either cytokine substantially reduced the early differentiation transcription factor, myoD, and modestly reduced the later differentiation factor, myogenin (Alvarez et al., 2020). However, nuclear expression of the myoblast-specific proliferation factor, pax7, was also reduced by these incubations (Alvarez et al., 2020). Furthermore, high physiological concentrations of IL-6 actually diminished myoblast proliferation, and incubation for longer periods of time dampened the inhibitory effect on differentiation (Steyn et al., 2019). These idiosyncratic outcomes were associated with the respective up or downregulation of IL6R by moderate or substantial IL-6 exposure (Steyn et al., 2019). Moreover, IL6R inhibition diminished not only the effects of IL-6 but those of TNFα and IL-1β as well (Alvarez et al., 2020). Nevertheless, all cytokine effects were lost when myoblasts were incubated with NFκB inhibitors (Otis et al., 2014). Disruption of the balance between myoblast proliferation and differentiation clearly reduces the capacity for muscle hypertrophy. To illustrate, intramuscular IL-6 infusion in young rats decreased muscle growth by about 10% over 2 weeks (Bodell et al., 2009). This coincided with greater gene expression for cytokine signaling components, including TNFα, TNFR1, and atrogin-1 (Haddad et al., 2005; Bodell et al., 2009). In TNFα-infused mice, an observed increase in muscle catabolism was attributed to greater ubiquitin-dependent proteolysis (Llovera et al., 1997). The combination of greater relative muscle catabolism and decreased myoblast-facilitated muscle hypertrophy associated with experimentally-elevated cytokine exposure was comparable to the phenotype observed in the IUGR fetus (Yates et al., 2014; Soto et al., 2017; Rozance et al., 2018; Posont et al., 2022).

Like inflammation, oxidative stress impairs skeletal muscle hypertrophy in similar but independent fashion (Jang et al., 2020). Addition of even modest concentrations of reactive oxygen species to myoblast incubations increased proliferation (Sciancalepore et al., 2012) and inhibited differentiation (Langen et al., 2002; Sandiford et al., 2014), regardless of the presence of inflammatory cytokines. When primary mouse myoblasts were engineered to overproduce H₂O₂, only about half as many stained positive for myogenin, expressed myosin heavy chain protein, or fused with other myoblasts (Sandiford et al., 2014). Furthermore, accumulation of reactive oxygen species both directly and indirectly resulted in muscle catabolism (Jang et al., 2020). In addition to directly damaging cellular proteins, elevated reactive oxygen species in antioxidant-deficient mice upregulated cysteine proteases, which increased protein breakdown by the ubiquitin-proteasome and autophagy-lysosome systems (Muller et al., 2006; Jang et al., 2020). These mice were characterized by a reduction in muscle mass of up to 50%, along with poor exercise performance (Muller et al., 2006).

Sustained exposure to elevated inflammatory cytokines or reactive oxygen species dysregulates skeletal muscle glucose metabolism and impairs insulin signaling and secretion. Brief incubations with high physiological IL-6 concentrations increased glucose uptake, oxidation, and incorporation into glycogen and also reduced lactate production in primary human and rodent muscle, as well as in culture-derived myotubes (Al-Khalili et al., 2006; Glund et al., 2007; Gray et al., 2009; Cadaret et al., 2017). Glucose uptake was stimulated at much lower IL-6 concentrations when its receptor was added to the incubation in concert (Gray et al., 2009). In primary rat muscle and myotubes, TNFα similarly elevated glucose uptake and oxidation (Yamasaki et al., 1996; Cadaret et al., 2017). Moreover, greater glucose oxidation was observed in vivo when healthy young adults were infused with moderate amounts of IL-6 over several hours (Carey et al., 2006). Some of these studies indicated that cytokine regulation of muscle glucose utilization was independent of insulin and its major pathway components, IRS1 and Akt (Yamasaki et al., 1996; Glund et al., 2007; Gray et al., 2009; Cadaret et al., 2017). Rather, cytokines appeared to stimulate translocation of sequestered Glut4 glucose transporters to the membrane by activating the kinase enzyme, AMPK (Al-Khalili et al., 2006; Carey et al., 2006). In fact, longer periods of cytokine exposure generally disrupted insulin signaling (Rotter et al., 2003; Al-Khalili et al., 2006; Cadaret et al., 2017), although the direct effects on glucose utilization remained. This was demonstrated particularly well by in vitro incubation of myoblast cell lines with TNFα, which increased basal glucose uptake by almost 3-fold but reduced insulin-stimulated glucose uptake by about 75% (Yamasaki et al., 1996; Grzelkowska-Kowalczyk and Wieteska-Skrzeczynska, 2006; Remels et al., 2015). Infusion of TNFα into mice resulted in a 50% reduction of insulin-stimulated hindlimb glucose uptake but did not affect resting glucose uptake rates (Youd et al., 2000). The primary target of cytokines within the insulin signaling pathway appears to be the phospho-activation of Akt, a hub that facilitates myriad insulin effects (Ji et al., 2022). Incubation of primary rat muscle with moderate concentrations of TNFα or IL-6 for as little as 2 h reduced insulin-stimulated Akt phosphorylation by more than 50% (Cadaret et al., 2017). Similar deficits were observed in culture-derived myotubes after 6-day incubation with TNFα (Grzelkowska-Kowalczyk and Wieteska-Skrzeczynska, 2006) and in primary bovine adipocytes after 12 h (Du et al., 2022). In addition to disrupting insulin signaling, cytokines also impair insulin stimulus-secretion coupling in pancreatic islets, although this effect is quite dependent upon duration of exposure. In fact, when mice were injected with IL-6, glucose-stimulated insulin secretion was increased over the first 45 min, but this effect was gone by 1 h (Ellingsgaard et al., 2011). Incubation of primary mouse islets with IL-1β alone or in combination with IFNγ increased glucose-stimulated insulin secretion for the first few hours but reduced it after 20 h (Andersson et al., 2005). Likewise, 48-h incubation with TNFα alone or together with IL-1β and IFNγ had no effect on basal insulin secretion but completely inhibited glucose-stimulated insulin secretion in primary rat islets and in INS-1 rat β cells (Hadjivassiliou et al., 1998; Zhang et al., 2022). This combination of cytokines also reduced insulin content in primary human islets after 48 h (Hadjivassiliou et al., 1998). The 40% reduction in glucose-stimulated insulin secretion and 25% reduction in pyruvate-stimulated insulin secretion observed in INS-1 rat β cells after 24 h with IL-1β and IFNγ coincided with reduced indicators of oxidative metabolism, the facilitating mechanism for β cell stimulus-secretion coupling (Rozance et al., 2007; Barlow et al., 2018).

Like cytokines, in vitro exposure of primary rat muscle to reactive oxygen species for less than 1 h increased glucose uptake by up to 48% and glycogen synthase activity by about 20% by enhancing canonical insulin signaling (Kim et al., 2006; Higaki et al., 2008). In fact, concurrent exposure to insulin and reactive oxygen species for 20 min had an additive effect on glucose uptake (Higaki et al., 2008). Short-term oxidative stress reflects the normal microenvironment of muscle and is not particularly harmful, but sustained exposure becomes deleterious (Higaki et al., 2008; Diamond-Stanic et al., 2011). Incubation of rat muscle with high concentrations of reactive oxygen species ceased stimulating basal glucose uptake after 6 h (Diamond-Stanic et al., 2011). Insulin-stimulated glucose uptake and Akt phosphorylation were hindered by reactive oxygen species after only 2 h (Diamond-Stanic et al., 2011). The adverse effects of longer exposure on insulin signaling components were mediated in part by p38 MAPK and were evident even at modest reactive oxygen species concentrations (Archuleta et al., 2009; Diamond-Stanic et al., 2011). Interestingly, reactive oxygen species did not appear to affect function of primary rat or human islets after 48 h of incubation (Hadjivassiliou et al., 1998).

Inflammatory cytokines help to regulate lipid homeostasis and metabolism by skeletal muscle in coordination with their effects on muscle glucose utilization. Infusion of IL-6 into healthy individuals for 3 hours increased total fatty acid oxidation rates during and for several hours after the infusion period (van Hall et al., 2003). In primary rat muscle, 1-h incubations with high IL-6 concentrations increased palmitate oxidation but not deposition, whereas 1-h incubation with TNFα increased deposition but not oxidation (Bruce and Dyck, 2004). Conversely, 3-h incubation of culture-derived human myotubes with IL-6 increased palmitate uptake and oxidation rates (Al-Khalili et al., 2006). Greater oxidation occurred in the presence and absence of insulin and was mediated primarily by AMPK pathways (Bruce and Dyck, 2004; Al-Khalili et al., 2006). To facilitate greater lipid utilization by muscle, cytokines concurrently mobilize lipid deposits from adipose and other tissues. This was demonstrated by infusions of moderate or high concentrations of IL-6 into humans, both of which elevated circulating NEFA and triglycerides (van Hall et al., 2003). Moreover, blocking IL-6 signaling by administering a long-lasting IL6R antagonist reduced indicators of fatty acid mobilization in adult men during various degrees of physical activity and across a range of body mass indices (Trinh et al., 2021). In primary rat muscle, IL-6 stimulated lipid mobilization in part by disrupting insulin’s lipogenic effects (Bruce and Dyck, 2004). Unlike IL-6, TNFα had no impact on endogenous or exogenous skeletal muscle fatty acid oxidation (Bruce and Dyck, 2004). However, several in vitro studies have shown that even modestly elevated TNFα concentrations stimulated lipolysis (Ryden et al., 2004; Lee and Fried, 2012; Du et al., 2022). This was recapitulated by 7-day TNFα infusion into mice, which elevated their circulating triglycerides and NEFA concentrations (Li L. et al., 2009). Moreover, both glycerol and NEFA were released from adipocytes at greater rates when stimulated with TNFα in vitro (Green et al., 2004; Ryden et al., 2004). TNFα-mediated lipolysis was facilitated in large part by downregulation of perilipin, a protein that mediates hormone-sensitive lipase activity in lipid droplets (Morin et al., 1995; Wu et al., 2004). Studies in rats and cultured adipocytes found that TNFα also reduced activity and expression of lipoprotein lipase, an enzyme that breaks down circulating lipoproteins for deposition into adipocytes (Morin et al., 1995; Wu et al., 2004). The increased production of reactive oxygen species in cytokine-stimulated adipocytes further contributed to lipolysis (García et al., 2021), an effect that was dampened by concurrent inhibition of superoxide production in culture (Krawczyk et al., 2012; Issa et al., 2018). The impact of reactive oxygen species on adipocyte lipolysis coincided with phospho-activation of hormone sensitive lipase (Krawczyk et al., 2012; Issa et al., 2018).

Benefits of ω-3 PUFA supplementation during critical windows for fetal development on growth and metabolic outcomes have been documented in humans and animals. Meta-analyses of global health records and clinical trials indicated that dietary ω-3 PUFA supplementation over the 2nd half of gestation was associated with reductions of up to 73% in perinatal mortality rates as well as fewer neonatal intensive care stays (Saccone et al., 2015; Saccone et al., 2016; Ali et al., 2017; Middleton et al., 2018). The supplements were also associated with lower incidence of IUGR in many world populations (Cetin et al., 2002; Agostoni et al., 2005; Fares et al., 2015; Saccone et al., 2015; Gholami et al., 2020). In one prime example, DHA supplementation in primigravidae women in Mexico reduced IUGR rates by half (Ramakrishnan et al., 2010). Better pregnancy outcomes coincided with positive effects on uteroplacental tissues, as clinical trials found that ω-3 PUFA supplements increased uterine and umbilical blood flow and reduced placental apoptosis (Wietrak et al., 2015; Ali et al., 2017). Moreover, dietary ω-3 PUFA supplementation in pregnant rats reduced indicators of placental oxidative stress and increased placental and fetal mass (Jones et al., 2013). Even direct infusion of EPA into the bloodstream of IUGR fetal sheep benefitted placental function, as maternofetal glucose and O₂ gradients were improved (Beer et al., 2021). Importantly, maternal ω-3 PUFA supplements directly benefit the fetus as well as the placenta. Clinical trials found that IUGR fetuses and newborns were deficient in ω-3 PUFA due to impaired de novo production and that maternal supplementation was effective in increasing circulating concentrations in the fetus/newborn (Cetin et al., 2002; Agostoni et al., 2005; Llanos et al., 2005; Fares et al., 2015; Wietrak et al., 2015). These findings were recapitulated in rats, where IUGR-born neonates were found to be DHA-deficient, but maternal supplementation increased circulating DHA concentrations in these offspring by 3.25-fold (Morand et al., 1981; Joss-Moore et al., 2010). In pigs, maternal ω-3 PUFA supplementation increased DHA and EPA and reduced ω-6:ω-3 PUFA ratios in fetal blood and skeletal muscle near term, which coincided with greater blood glucose, less cholesterol, and a reduction in the incidence of IUGR from 22% to 14% (Heras-Molina et al., 2021; Heras-Molina et al., 2022). These benefits persisted after birth, as weaning-aged pigs born to ω-3 PUFA-supplemented sows had less circulating total cholesterol, HDL-C, and LDL-C as well as greater total ω-3 PUFA content and lower ω-6:ω-3 PUFA ratios in adipose, muscle, and liver tissues (Heras-Molina et al., 2020). When pregnant rodents carrying IUGR pregnancies were supplemented DHA, birthweight and neonatal growth of their offspring were improved without increased adiposity (Bagley et al., 2013; Velten et al., 2014). Offspring from dams supplemented DHA or ω-3 PUFA-rich fish oil had greater circulating adiponectin, smaller adipocytes, and adipose tissue that expressed more PPARγ, adiponectin, and adiponectin receptors R1 and R2 (Bringhenti et al., 2011; Bagley et al., 2013; Ghnaimawi et al., 2022). Supplementing pregnant cows with ω-3 PUFA (delivered as Ca²⁺ salts for ruminal protection) during late gestation increased circulating DHA in the dam by 5-fold, and calves born to supplemented cows exhibited 6%–10% better average daily gain in the feedlot (Marques et al., 2017). At harvest, these calves produced marginally larger carcasses and loin muscles with ∼10% greater marbling (Marques et al., 2017). In sheep, maternal supplementation of ω-3 PUFA Ca²⁺ salts for the final trimester of gestation had no adverse effects on dam or fetus and increased blood concentrations of EPA and DHA in the ewe by ∼40% and in the newborn lamb by ∼50% (Coleman et al., 2018; Rosa Velazquez et al., 2020; Rosa-Velazquez et al., 2022). Greater fetal ω-3 PUFA availability appeared to be particularly beneficial to IUGR muscle and pancreatic islets. In mice, protein content for the insulin receptor and the myogenic transcription factor, myoD, as well as several genes regulating myogenesis were upregulated in skeletal muscle of newborn pups born to ω-3 PUFA-supplemented dams (Ghnaimawi et al., 2022). Moreover, 5-day infusion of EPA directly into IUGR fetal sheep improved growth of several muscles, restored normal fiber type proportions in the semitendinosus, and modestly improved deficits in muscle glucose uptake and oxidation (Lacey, 2021; Lacey et al., 2021). It also recovered about 50% of the deficit in glucose-stimulated insulin secretion (Lacey, 2021; Lacey et al., 2021). These benefits coincided with anti-inflammatory indicators, as EPA infusion ameliorated the elevated circulating TNFα concentrations and reduced total circulating white blood cells observed in IUGR fetal sheep (Lacey, 2021; Lacey et al., 2021). Likewise, maternal ω-3 PUFA supplementation reduced oxidative stress in brain tissues of fetal rats and resolved the heightened macrophage invasion observed in lung tissues of IUGR newborn mice pups (Velten et al., 2014).

Postnatal supplementation of ω-3 PUFA to IUGR-born offspring is also effective in improving metabolic deficits. For example, IUGR-born rats exhibited 50% reductions in circulating triglycerides and 33% reductions in blood urea nitrogen when nursing foster dams that were fed diets high in ω-3 PUFA (i.e., supplemented to offspring via milk) (Voggel et al., 2022). Likewise, directly feeding ω-3 PUFA-enriched diets or supplementing ω-3 PUFA-rich fish oil to IUGR-born juvenile rats partially resolved their hyperlipidemia and reduced their adiposity (Wyrwoll et al., 2006; Bringhenti et al., 2011; Chen et al., 2016; Chicco et al., 2016). It also decreased adipocyte size and leptin secretion, ameliorated inflammatory markers in adipose tissues, and resolved elevated TNFα in circulation (Wyrwoll et al., 2006; Bringhenti et al., 2011; Mark et al., 2014). Postnatal fish oil supplementation improved HOMA-estimated insulin sensitivity, glucose tolerance, and markers of systemic inflammation in IUGR-born rats (Bringhenti et al., 2011; Chen et al., 2016).

As of this writing, fish oil and other ω-3 PUFA supplements are considered safe for pregnant women by the American Pregnancy Association, which is supported by several Cochrane meta-analyses (Kar et al., 2016; Middleton et al., 2018; Wieland, 2019). Challenges regarding the use of ω-3 PUFA as dietary supplements in livestock include post-ingestive feedback (Freidin et al., 2011), bioavailability in ruminants (Chikunya et al., 2004), and meat sensory traits when fed at high concentrations (Burnett et al., 2020). Reduced palatability and ruminal microbiome changes associated with fish oil can cause animals to eat less when too much is included in a dietary ration, and livestock studies have indicated that the threshold limit is around 2% of the diet. Indeed, inclusion of fish oil at 1% of the diet in dairy cows did not affect ad libitum intake, but inclusion at 2% and 3% reduced intake by up to 11% and 32%, respectively (Donovan et al., 2000; Whitlock et al., 2002; Alizadeh et al., 2012; Kairenius et al., 2018). Although milk yield was increased by dietary fish oil at 1% and 2% in these cows, the large drop in dietary intake at 3% caused milk yield to fall (Donovan et al., 2000). Calves fed milk replacer enriched with 0.5% fish oil maintained their intake from birth to 2 months of age, which facilitated a modest increase in their average daily gain (Melendez et al., 2022). Lambs fed Ca²⁺ salts of ω-3 PUFA at 1.5% also maintained dry matter intake (Carranza Martin et al., 2018). In Angus feedlot steers, dry matter intake was reduced by dietary inclusion of fish oil at 2.4% but not at 0.8% or 1.6% (Shingfield et al., 2010). Limits on dietary inclusion could be problematic for ruminant livestock, as bioavailability of ω-3 PUFA is reduced by the markedly high rates of microbial biohydrogenation (i.e., saturation of previously unsaturated fatty acids) in the rumen (Jenkins et al., 2008). To illustrate, 93% of EPA and DHA from unprotected sources were absorbed unmodified across the intestinal wall of the rat (Chen et al., 1985), but as little as 7% were absorbed unmodified in cows (Dewhurst and Moloney, 2013). Similar outcomes were created by the rumen microbiome in sheep, where biohydrogenation rates were upward of 75% (Chikunya et al., 2004; Jenkins et al., 2008). Processing techniques for ω-3 PUFA-rich feed ingredients like flaxseed increased bioavailability of EPA by 30% and total ω-3 PUFA by 23% in lambs (Kronberg et al., 2012). In young goats, heat treatment of linseed oil diets increased DHA bioavailability by 62% and total ω-3 PUFA bioavailability by 19%, which corresponded with greater liver and adipose tissue concentrations (Wang et al., 2019). Freeze-drying of a microalgae-based supplement reduced ruminal biohydrogenation of EPA in lambs from 80% to about 45% (Vítor et al., 2021). Supplements utilizing Ca²⁺ salts of ω-3 PUFA helped maintain or increase bioavailability through ruminal passage without producing off-target effects such as reduced intake and milk fat (Castañeda-Gutiérrez et al., 2007; Oyebade et al., 2020). It is worth noting that even unprotected dietary sources still deliver meaningful amounts of absorbable ω-3 PUFA in ruminants. For example, inclusion of 1.1% fish oil in a silage-based diet for dairy cows increased the amount of EPA and DHA bypassing the rumen microbiome by 2-fold and 3-fold, respectively, despite 78% and 83% biohydrogenation rates (Kairenius et al., 2018). In meat animals, ω-3 PUFA supplementation strategies must consider the effects on product appearance and shelf life. Like all fatty acids, ω-3 PUFA are prone to oxidation that can discolor the surface of meat products (Burnett et al., 2020). Steers fed diets with 3% fish oil had 1.5-fold–2-fold greater carcass EPA and DHA content at harvest (Vatansever et al., 2000). Hamburger patties from these steers reached unacceptable discoloration 1–3 days sooner than normal and cooked sirloin scored about 11% lower in overall liking, despite no reduction in flavor score (Vatansever et al., 2000). It should be noted that most of the individual fatty acids measured in this study were increased in meat from fish oil-supplemented animals and that linseed oil supplementation in the same study increased ω-3 PUFA content without the same adverse sensory effects. Thus, adverse sensory traits may be attributable more to the use of a high volume of fish oil as a source than to the increase in ω-3 PUFA per se. Indeed, palatability and tolerability of German sausage products were not affected by enrichment with ∼1% purified EPA, DHA, and α-linolenic acid (Köhler et al., 2017). Nevertheless, fish products are the most common source for ω-3 PUFA supplements in livestock (Burnett et al., 2020), and lamb cuts from animals fed diets with 9% fish meal for 6 weeks were scored ∼9% lower in juiciness by a trained sensory panel, although flavor, aroma, and palatability were not affected (Ponnampalam et al., 2002). Likewise, meat from lambs fed diets enriched with 1.5% fish oil for 6 weeks was scored 14% lower in overall palatability, despite no reduction in individual scores for flavor, aroma, or juiciness and no issues with discoloration or shelf life (Ponnampalam et al., 2001; Ponnampalam et al., 2002). Because of the potential for reduced meat sensory traits, investigation into the benefit of withdrawing ω-3 PUFA supplements prior to harvest may be warranted.