Deficits in vascular function, including at the level of the microvasculature, are strongly associated with birth weight (low and high) and later cardiovascular risk in infants, children and young adults born small or prematurely (Clough & Norman 2011, Torrens et al. 2011), and maternal calorific intake and calorific composition have been shown to influence microvascular angiogenesis, vasculogenesis and barrier integrity and maturation of the foetal vascular system (Leach 2011). In addition, a number of studies describe the effects of maternal diet and other maternal stressors on uterine and umbilical blood flows and the size of the placental vascular bed and their potential to affect foetal vascularization, independent of offspring birth weight so confirming the critical role of the placenta in developmental programming (Redmer et al. 2009, Meyer et al. 2010, Reynolds & Caton 2012). While there have been few investigations of associations between overnutrition/obesity in pregnant women and cardiovascular risk in their children, a positive association between gestational weight gain and offspring obesity and systolic blood pressure has been reported in young children (Lawlor et al. 2004, Oken 2009) and young adults, where a greater gestational weight gain was associated with greater BMI and systolic blood pressure (Mamun et al. 2009). There are also an increasing number of animal studies examining the association between maternal overnourishment/obesity, foetal growth rate and placental vascularity (Ma et al. 2010), foetal organ development (George et al. 2010) and offspring organ and vascular function and a cardio-metabolic disease phenotype (Taylor et al. 2004, Samuelsson et al. 2008, Bruce et al. 2009, Torrens et al. 2012, Dong et al. 2013) as reviewed by Ainge et al. (2011).
The first report linking ‘intrauterine under-nutrition’ with later obesity and cardiovascular disease was described in an epidemiological study on adult offspring of female survivors of the Dutch famine of 1944. Ravelli et al. (1976) suggested that maternal undernutrition during early gestation was associated with the development of hypertension in the offspring in adult life (Roseboom et al. 1999, 2001). In later stages of gestation, however, under-nutrition gave rise to offspring with an increased adiposity and glucose intolerance when subjects were compared to well-nourished offspring (Law et al. 1992, Ravelli et al. 1999). Studies in other human cohorts have shown similar findings. Yajnik et al. (1995, 2003) pioneered the first study of ‘foetal origins’ in India where they tested the hypothesis that low birth weight was an independent predictor of glucose tolerance and insulin resistance variables in 4- and 8-year-old urban children born in the city of Pune, India. Children with a low birth weight were found to have a significantly high systolic blood pressure, plasma cholesterol and insulin, and were significantly heavier than non-low-birth weight children. A more recent study showed that offspring born after the Chinese famine (1959–1961) had a higher risk of developing hyperglycaemia and type-2 diabetes in adulthood. This risk was found to be higher in subjects who went on to adopt a Western dietary pattern (Li et al. 2010). Together these studies also suggest that an important driver for future cardio-metabolic risk was the ‘nutritional mismatch’ experienced by the offspring during their intrauterine and early post-natal life. This paradigm formed the basis of the predictive adaptive response hypothesis described by Gluckman and colleagues (Gluckman & Hanson 2004, Gluckman et al. 2008, Vickers & Sloboda 2012).
Oxidative stress resulting from compromised endogenous antioxidant defences is a common feature of cardio-metabolic disease and has emerged as a key link between birth weight and increased morbidity later in life as a consequence of sustained oxidative stress. Reduced levels of superoxide dismutase, catalase, glutathione and serum malondialdehyde have been observed when cord blood from small for gestational age (SGA) babies was compared to that of healthy children that were born appropriate for gestational age (AGA) (Gupta et al. 2004). Furthermore, there is evidence that the raised levels of biomarkers of oxidative stress in addition to insulin resistance persist in pre-pubescent life (Franco et al. 2007, Chiavaroli et al. 2009).
Several human studies have also provided evidence for impaired endothelium-dependent and endothelium-independent vasodilation in low birth weight individuals at birth, at 3 months of age, in later childhood and in early adult life (Goh et al. 2001, Norman & Martin 2003). In infants, vascular dysfunction was evidenced by a significant reduction in acetylcholine-induced vasodilation in the microvasculature and a reduced ability of the brachial artery to dilate in response to increased blood flow following occlusion (Martin et al. 2000). In a recent study in children aged 7–15 year, it was observed that non-obese SGA children similarly displayed a lower FMD compared with obese AGA children (Jouret et al. 2011).
In many countries, preterm birth is a more common cause of low birth weight than foetal growth restriction. Prematurely born infants display an increased risk for developing metabolic (insulin resistance) and cardiovascular disorders (elevated blood pressure and abnormal retinal vasculature) as adults (Kistner et al. 2005, Kerkhof et al. 2012). Interestingly, these individuals have been shown to have otherwise healthy endothelial function in childhood and early adulthood at both microvascular and macrovascular levels (Bonamy et al. 2005, 2007, Mikkola et al. 2007). Adolescents and young adults born premature have higher blood pressure, abnormal retinal vascularization and lower peripheral skin blood flow than control subjects (Kistner et al. 2002, Bonamy et al. 2005). In another study, adult women born preterm appear to have a 2.5-fold increased risk of developing hypertension compared with women born at term (Pouta et al. 2004). Interestingly, in those studies, SGA subjects were not different from AGA subjects, both within the premature and the term groups, suggesting an important role of preterm birth itself in the programming of cardiovascular function in later life. Although the mechanisms linking prematurity and later cardiovascular disorders remain unclear, it is evident that it is associated with impaired nephrogenesis and vasculogenesis (Gilbert et al. 2005, 2007) as well as oxidative stress (Yeung 2006), all of which may contribute to later end-organ microvascular dysfunction and the risk of developing cardio-metabolic disease. The extent to which the microvasculature is altered in other tissues in premature and low birth weight individuals has important implications for cardio-metabolic health, and remains to be clarified. For instance, muscle capillary density determines diffusion and tissue concentration of insulin, as described above. Functional and/or structural loss of muscle capillaries could therefore be a mechanism linking low birth weight and/or preterm birth to later insulin resistance.
Intrauterine undernutrition is associated with endothelial vasodilator dysfunction in the adult offspring (Lamireau et al. 2002, Taylor et al. 2004, Torrens et al. 2009). Animal models of maternal undernutrition support a role for programming of hypertension through altered endothelium-dependent dilations in resistance arteries in the offspring (Torrens et al. 2003) via altered NO signalling (Rodford et al. 2008, Torrens et al. 2009). Consistent with this, Lamireau et al. (2002) showed that maternal protein deprivation resulted in a 50% reduction of NO-dependent relaxation in resistance cerebral microvessels which was associated with reduced cGMP and soluble guanylate cyclase expression. Reduced NO levels in undernourished offspring have also been attributed to oxidative stress following an increased superoxide generation (Ceravolo et al. 2007) and reduced tetrahydrobiopterin (BH4), an important cofactor in nitric oxide synthase (NOS) activity and therefore in the conversion of L-Arg to NO, in resistance arteries (Franco et al. 2002, 2003, 2004). The administration of antioxidants (Franco et al. 2003) and exogenous BH4 (Franco et al. 2004) was found to reduce oxidative stress and blood pressure thus improving vascular function. In support of this, endothelial dysfunction and reduced antioxidant protection in resistance arteries in male offspring of protein-restricted rats were linked to the reduced expression of the antioxidant enzyme, haem oxygenase-1 (HO-1) (Rodford et al. 2008).
Strong correlations between endothelial function in adult offspring and micronutrient intake during pregnancy have also been found. In a protein-restricted model, low folate levels were linked to a reduced NO bioavailability and endothelial NOS mRNA expression (Torrens et al. 2006). Similarly, vitamin D deficiency during development and early life in rats was associated with a reduced EDHF-induced dilation and an increased myogenic tone in resistance arteries in adult offspring (Tare et al. 2011).
Structural alterations in small resistance arteries have also been implicated in mediating vascular dysfunction and hypertension in offspring exposed to intrauterine calorie restriction (Khorram et al. 2007a,2007a,2007c). For example, reduced microvessel numbers and branches found in the mesenteric and renal microcirculation as a result of decreased angiogenesis in offspring of calorie restricted dams are thought to contribute to an increased peripheral vascular resistance and hypertension in later life (Khorram et al. 2007c). Further, capillary rarefaction in skeletal muscle in young rats (7 and 28 days old) but not in foetal rats with programmed raised blood pressure associated with a maternal low-protein diet, suggesting an early post-natal disruption in normal microvascular development (Pladys et al. 2005). In the sheep, capillary rarefaction in skeletal muscle and a reduced capillary to myofibre ratio have also been linked to a reduction in myofibre density but not in size, in a muscle-specific manner, following maternal undernutrition (Costello et al. 2008).
Microvascular rarefaction associated with impaired angiogenesis or active disappearance of blood vessels has been reported in some but not in all maternal dietary restriction models in skeletal muscle and cerebral cortex (Pladys et al. 2005, Khorram et al. 2007a, Costello et al. 2008) and reviewed by Clough and Norman (2011). A reduced NO bioactivity or increased oxidative stress in early life that is associated with many dietary models could be postulated to impact on microvessel formation and survival through perturbation of angiogenic and/or apoptosis signalling pathways (Pladys et al. 2005, Cambonie et al. 2007, Neville et al. 2010). However, a reduction in VEGF and VEGFR protein expression seen in neonates appears reversed and/or normalized by adulthood (Khorram et al. 2007b). Placental IGF-I synthesis is also down-regulated in placental insufficiency and foetal growth restriction, and low serum concentrations of IGF-I in preterm infants have been associated with growth arrest of the microvasculature in the eye (Hellstrom et al. 2003) which is modifiable by nutritional intervention (Engstrom et al. 2005).
An imbalance between angiogenic and anti-angiogenic factors involved in regulation of number and function of foetal endothelial progenitor cells has also been suggested to be involved in developmental programming of hypertension and cardiovascular disease (Ligi et al. 2010). Altered angiogenic properties of cord blood endothelial colony-forming cells (ECFCs) have been in preterm neonates compared with those studied at term (Ligi et al. 2011). Similarly, subsets of circulating progenitor cells (CPCs) with robust angiogenesis-facilitating functions have been shown to be reduced in women with gestational diabetes (Acosta et al. 2011). However, the impact of these increases in CPCs does not appear to be translated into altered vascular function in the neonates, as endothelial function appears normal (Acosta et al. 2011). Thus, to what extent impaired angiogenic properties of CPCs and ECFCs and other factors may influence early vascular remodelling and/or angiogenesis has yet to be explored.
Nutrient restriction has also been shown to induce remodelling of the vascular extracellular matrix with increased collagen deposition and matrix metalloproteinase expression in the walls of both mesenteric arterioles and the aorta, perhaps driven by the overexpression of VEGF, in adult offspring (Khorram et al. 2007a). Other studies report a 30% reduction in venular basement membrane type IV collagen in 8-week-old male offspring of undernourished rat dams (Landgraf et al. 2007) which may also modify angiogenesis and/or apoptosis of microvessels. More recently, maternal undernutrition has been shown to alter offspring vascular expression of micro-RNAs (miRNAs) including those targeting genes involved with the regulation of collagen, elastin, enzymes involved in ECM remodelling and angiogenic factors (Khorram et al. 2010,). These in turn could regulate the expression of genes involved with angiogenesis and extracellular matrix remodelling. While these findings help us to understand the links between the early life environment and structural remodelling of the microvasculature, the cell-signalling mechanisms that underlie functional deficits have yet to be fully elucidated.
Animal studies also give insight into the effects of the timing of dietary intervention in pregnancy on the extent of vascular programming. Intrauterine growth restriction in late gestation led to region-specific programming of vascular dysfunction, but not to hypertension and obesity in 18-month-old female offspring. In these females, vascular dysfunction, as manifested by a reduction in EDHF-mediated relaxation and structural rarefaction, was seen in the uterine artery but not in the mesenteric, renal and femoral arteries (Mazzuca et al. 2010). In contrast, intrauterine growth restriction induced by restricting blood flow to the gravid uterus (day 14 of gestation) in rats resulted in growth-restricted female offspring that were hypertensive as adults and had an increased smooth muscle contractility and altered endothelium-dependent and -independent relaxation in the mesenteric microcirculation (Anderson et al. 2006). Further, the effects on the offspring were transgenerational. More recently, Torrens and others showed the vascular effects of foetal programming may have long-term consequences for the development of the next generation of offspring. In this study, offspring of rats that experienced intrauterine protein restriction exhibited a high blood pressure and endothelial dysfunction even in the absence of an additional dietary challenge (Torrens et al. 2008).
Consistent with the developmental origins of cardiovascular disease hypothesis, altered vascular biology is not confined to those who are small at birth. Epidemiological data from human cohorts have shown a positive correlation between increased birth weight (used as a surrogate for maternal overnutrition) and a high BMI and/or obesity in adulthood (Sen et al. 2012). In addition to weight gain, other studies have provided a strong link between maternal obesity and subsequent development of the metabolic syndrome in adulthood, independent of maternal diabetes (Boney et al. 2005). However, while an association between maternal overnutrition and foetal growth, adiposity and cardiovascular risk factors in well-nourished populations (Drake & Reynolds 2010, Freeman 2010) has been shown, there is little evidence as yet for the consequences of maternal overnutrition on offspring microvascular function. However, the maternal insulin resistance, which is induced by an increased maternal fat mass, is likely to affect not only foetal growth, but also microvascular structure and function. Increased maternal body fat, serum leptin and triglycerides are associated with altered placental vascular development and a 30% decrease in the number of mature blood vessels that stained positive for smooth muscle actin (Hayes et al. 2012) which these authors hypothesize may result in a reduced oxygenation of foetal tissue contributing to poorer outcomes. A number of recent studies have further shown that greater pre-pregnancy BMI and pregnancy BMI are associated with adverse retinal microvessel number and structure (Li et al. 2012) and that retinal arterioles and venules may be differentially associated with offspring growth in early life, and early adiposity may adversely affect the microcirculation (Tapp et al. 2013).
In a recent systematic review by Ainge et al. (2011) of studies on developmentally primed offspring of overnourished rodent dams, including ad libitum feeding of high-fat diets supplemented with sweetened condensed milk or high-fat and/or high sugar junk food/cafeteria diet who are then weaned onto a laboratory chow diet, offspring consistently exhibited hyperphagia, increased adiposity, hypertension, abnormal glucose homoeostasis and reduced insulin sensitivity, and endothelial dysfunction in resistance and conduit arteries that persists into adulthood (Samuelsson et al. 2008, Elahi et al. 2009). Furthermore, there is increasing evidence that these cardio-metabolic disease risk factors persist in spite of a healthier post-natal diet (Taylor et al. 2004, Armitage et al. 2008) and may be increased further by an adverse post-weaning diet and by age (Ford & Long 2011, Masuyama & Hiramatsu 2012, Torrens et al. 2012, Yan et al. 2013). Similarly, we (Torrens et al. 2012) have shown there to be significant effects of maternal diet on offspring body weight, systolic blood pressure and endothelium-dependent relaxation to ACh measured in the femoral artery. Having controlled for differences in maternal diet, post-weaning offspring diet significantly influenced systolic blood pressure and endothelium-dependent relaxation. There was also a significant interaction between maternal diet and offspring diet to influence NO bioavailability associated with an increase in ROS in the artery wall and insulin resistance (Bruce et al. 2009). In humans, an increased expression of ROS in the placenta has been suggested to be one of the initiating factors in neonatal insulin resistance and obesity in response to maternal overnutrition (Radaelli et al. 2003).
While the NO pathway has been implicated in the endothelial dysfunction observed in the offspring of overnourished mothers, this does not apply to all models or all arteries. In their model of maternal high fat, Khan et al. (2005) showed impaired dilatation of the mesenteric arteries to ACh, which was still present after blockade of NOS and COX pathways, leading them to suggest this was an EDHF-related adaptation. Interestingly, this did not appear to be the case in the femoral arteries of the same animals. Evidence for an up-regulation of EDHF-mediated relaxation is seen in small resistance arteries and arterioles from other animal models of diet-induced obesity (Chadha et al. 2010, Feher et al. 2010, Haddock et al. 2011), hypercholesterolaemia (Ashraf et al. 2007), hypertension (Zhang et al. 2011), diabetes (Leo et al. 2011) and in apolipoprotein E and low density lipoprotein receptor-deficient mice fed a high-fat diet in adult life (Wolfle et al. 2009). Further, Kelsall et al. (2012) have demonstrated that in addition to an altered ACh-mediated vasorelaxation in mesenteric arteries from offspring of high-fat-fed rat dams, the type and amount of maternal dietary fat differentially influenced vasomotor tone and de novo biosynthesis of arachidonic acid in vascular smooth muscle. The extent to which the foetal environment impacts on EDHF or other signalling pathways in the human microcirculation remains to be established. Further, the impact of low-grade inflammation in the maternal environment through increased expression of pro-inflammatory cytokine genes and ROS in the placenta (Radaelli et al. 2003, Zhu et al. 2010), and increased foetal serum levels of TNF-α and IL-6 (Stewart et al. 2007) on the microvasculature has yet to be elucidated.