Intrauterine growth restriction (IUGR) is associated with accelerated growth and increased adiposity in early life due to unknown mechanisms, which could include increased thyroid hormone (TH) action. We hypothesized that placental restriction (PR) of fetal growth would increase circulating TH concentrations and alter their response to fasting, and that these would relate to growth and body composition in the young lamb. PR reduced size at birth, increased fractional growth rates (FGRs) of soft and skeletal tissues up to 30 days of age, and slowed the ontogenic decrease in plasma total T3 and plasma total T3/T4. PR did not alter the abundance of plasma THs after short-term fasting. In general, plasma total T3 and total T3/T4 ratio correlated negatively, whereas plasma total T4 correlated positively with size at birth. Absolute growth rates of weight and crown-rump length correlated positively with plasma total T3 and total T4 between days 15 and 35. Current FGRs for weight and metatarsal length correlated positively with plasma total T3 between days 20 and 35. In conclusion, PR and small size at birth reduce plasma total T4 and increase plasma total T3 postnatally, whereas catch-up growth relates to increased abundance of the more bioactive forms of TH. Finally, greater soft tissue growth occurs in PR compared with control lambs at the same circulating TH concentrations. This suggests that PR and small size at birth may increase activation of T4 to T3 and sensitivity of soft tissues to TH, which may contribute to catch-up growth following IUGR.
- intrauterine growth restriction
- catch-up growth
- thyroid hormone
children who are born short or light at birth due to intrauterine growth restriction (IUGR) grow at an accelerated rate during infancy (termed “catch-up” growth) (29), but the physiological basis for this is unknown. Clinical IUGR is common, occurring in ∼10% of live births in Australia and up to 40% in some developing countries (45). Most IUGR infants (57–84%) undergo some catch-up growth, which begins as early as 2 wk and is largely complete by 5 mo of age for weight and head circumference and by 6 to 12 mo for height (1, 25, 29, 32, 34). This catch-up growth is associated with a reduced risk of morbidity and mortality in childhood (53) and increases the likelihood of target adult height (25, 29). However, it also predicts increased childhood and adult obesity (46) and may independently add to the risk of developing adult-onset diseases such as diabetes, hypertension, and other cardiovascular disease (12, 16, 19, 42). Understanding the mechanistic basis of catch-up growth after IUGR may clarify whether it could be causally related to later outcomes.
The association of catch-up growth with IUGR suggests that the underlying causes originate in an adverse intrauterine environment, which restricts growth and alters functional development (11, 18, 31), including that of characteristics influencing infant growth, such as appetite and the activity of major neuroendocrine and endocrine axes (13, 17, 21, 49, 55). One such candidate axis is that of hypothalamic-pituitary-thyroid axis (3, 41). Thyroid hormone (TH) acts via T3 and possibly other forms of TH to regulate growth and differentiation of major tissues, as well as fuel metabolism and metabolic efficiency (10, 48, 54). An important element of the hypothalamic-pituitary-thyroid axis and determinant of its activity are circulating levels of TH and the deiodination of T4 to T3, the more biologically active form (27).
Plasma T3 and T4 are reduced at birth in the IUGR infant (6, 28), but plasma T3 subsequently normalizes from 1 wk to 8 mo of age, whereas plasma T4 is reduced up to 2 mo of age before normalizing (28). This suggests that following IUGR, there may be normal TH production, but enhanced conversion of T4 to T3, which helps maintain TH bioavailability and action (28). Although the smaller IUGR infants in this study gained as much weight as normal infants over the period of study, suggestive of catch-up growth, the associations with circulating T3 or T4, were not assessed (28). It is therefore possible that in those infants undergoing catch-up, the conversion of T4 to T3 and plasma T3 levels are highest (28).
Limited data from other species also suggest a role for TH action in catch-up growth. In lambs with spontaneous fetal growth restriction and fed artificially, plasma T3 and T4 several hours after birth and plasma T3 at 11 days, correlated positively with birth weight, while plasma T3 after birth also correlated positively with subsequent weight gain during the first month of life (7). Whether plasma concentrations of THs beyond the first day of life are also related to size at birth, and subsequent postnatal growth in lambs with fetal growth restriction was not determined, however (7).
A major cause of IUGR is placental insufficiency, which reduces delivery of oxygen and nutrients to the fetus (reviewed in Ref. 38). We have shown that experimental restriction of placental and, hence, fetal growth in sheep and other species has similar growth, metabolic, and endocrine consequences for the fetus, to human IUGR (40). We have also recently shown that placental restriction reduces size at birth and increases postnatal growth in the first month of life in sheep (20). Placental restriction in sheep decreases plasma total T4 levels in the lamb in the first 24 h after birth (35), but whether plasma T3 is also altered and whether such changes persist and contribute to altered growth restriction is unknown. We therefore hypothesized that placental restriction (PR) of fetal growth in sheep would increase circulating TH concentrations and that these would relate to growth and body composition in the young lamb in the first month of life. Short-term fasting inhibits activation of T4 to T3 and reduces plasma total or free T3 in rats and humans in adaptation to fuel shortage (26, 30). We therefore hypothesized that the greater growth rate of PR lambs might be partially due to reduced responses of THs to fasting, such that TH levels and activity would be maintained to a greater extent between feeds in placentally restricted lambs compared with control lambs.
MATERIALS AND METHODS
Animals and Surgery.
All procedures were approved by the University of Adelaide Animal Ethics Committee (Animal Ethics Approval Number: M/1/97A). Placental growth was restricted (PR) in 45 Merino ewes by removal of the majority of visible endometrial caruncles (65–148) from the nonpregnant uterus (43). Six to ten caruncles were left in each horn of the bicornuate uterus. This reduces the number of placentomes formed in the subsequent pregnancy and hence placental size and function (39). Surgery was performed under aseptic conditions, and general anesthesia was induced by intravenous injection (9.9–14 mg/kg) of pentothal (Rhone Merieux, Australia), and maintained with halothane inhalation anesthetic (Zeneca Limited Australia) in oxygen. Postsurgery, all ewes received a 3-day course of intramuscular antibiotics (2 ml) (Ilium Penstrep-Troy Laboratories, Smithfield, New South Wales, Australia). After a 6-wk recovery period, the ewes entered a mating program, and pregnancies were confirmed by ultrasound. Ewes were housed in individual pens in animal holding rooms from approximately a week before giving birth, with a 12:12-h light-dark cycle, and fed lucerne chaff twice daily ad libitum, with water ad libitum. The control and placentally restricted ewes delivered 35 singleton lambs, of which 18 lambs were from control ewes, and 17 lambs were from the placentally restricted ewes (PR lambs). Placentae were collected and weighed ∼2 h after natural delivery to avoid disturbing neonatal and maternal behavior, and these data are only included where the placenta was intact. Lambs were housed in the pens with their mothers throughout the study. At 4 to 5 days of age postnatally, catheters were inserted into the femoral artery and vein under general anesthesia of halothane inhalation anesthetic and aseptic techniques. All lambs received an intramuscular injection (2 ml) of antibiotics (Ilium Penstrep-Troy Laboratories) before surgery and then for 3 days postsurgery. Patency of the catheters in the lambs was maintained by filling the catheters with heparinized saline (500 U/ml) daily for 3 days and then every second day.
Body weight and various parameters of size were measured at birth and subsequently every 5 days up to 45 days of age. The following parameters were measured to characterize growth: total body weight (soft tissue and skeletal tissue growth), crown-rump length (CRL; vertebral column growth), shoulder height, tibia and metatarsal lengths, and abdominal circumference (liver and gut). Body mass index was calculated as weight/CRL2 (kg/m2), and ponderal index was calculated as weight/CRL3 (kg/m3). The absolute growth rate (AGR) of each lamb was determined by linear regression analysis of a parameter vs. postnatal age from 0–45 days to obtain the slope of the line of best fit. Neonatal fractional growth rate (FGR) was calculated as the AGR for the period of growth (0–45 days) divided by the value of the parameter at birth. Current FGRs at each age (current FGR) were calculated as the absolute growth rate for the period of growth (0–45 days) divided by the value of the parameter at that age. FGRs better reflect the anabolic effort that the individual lamb is exerting for growth, as it indicates the amount of tissue gained relative to the size of that lamb and the tissues producing that growth. Furthermore, an increased fractional growth rate indicates catch-up growth. Neonatal and current FGRs for all parameters of growth at various ages were correlated with plasma total T3 and T4 concentrations at the corresponding ages. At 43 ± 2 days of age, a subset of the lambs (nine controls and nine PR) was killed by intravenous administration of an overdose of barbiturate (pentobarbital sodium, Lethabarb). The fat depots that could be completely dissected were weighed (retroperitoneal, perirenal, and omental fat depots). Visceral fat weight was calculated as the sum of these three fat depots. Relative visceral fat weight was calculated as a proportion of body weight.
Plasma THs (Free and total T3 and T4).
Fasted arterial blood samples were taken at 30 days of age, and plasma was removed for determination of TH concentrations in 18 control and 17 PR lambs. In a subset of these animals (9 control: 5 male, 4 female; 9 PR: 4 male, 5 female), additional fed (days 5, 15, 20, 35, 40, 45) and fasted (day 8) arterial blood samples were collected for determination of plasma TH concentrations. Fed samples at day 5 were collected ∼24 h postsurgery. Before collection of fasting samples, lambs were fasted for 3 h by placing them into a pen beside the mother to prevent suckling and to minimize stress. Free T3 and T4 (FT3, FT4), and total T3 and T4 (TT3, TT4) were determined in plasma by specific radioimmunoassay (RIA-gnost FT3, FT4, TT3, TT4 CIS bio International, France). The detection limits for the assays of FT4 and TT4 were 0.5 pg/ml and 2.5 ng/ml, respectively, whereas the detection limits for FT3 and TT3 were 2.0–4.25 pg/ml and 0.1 ng/ml, respectively. The intra-assay coefficients of variation for FT3, TT3, FT4, and TT4 were 1.9%, 1.6%, 5.4%, and 6.2%, respectively.
Data were analyzed using the SPSS 13.0 software package for Windows. Two-way ANOVA was used to test for effects of gender and interactions with PR. Where gender effects were not significant, the effect of PR on parameters was subsequently assessed by 1-way ANOVA. The effect of PR on TH concentrations in fed plasma was analyzed using repeated-measures ANOVA (six levels). Comparisons between control and PR lambs at each age were made by using 1-way ANOVA. Associations between parameters were assessed by Pearson correlation or multiple linear regression analysis. Statistical significance was assumed at P < 0.05, and data are expressed as means ± SE.
Effect of PR on size at birth and postnatal growth.
PR reduced placental weight (−39%) and size at birth, with greater reductions in soft tissues: weight (−23%), CRL (−7%), abdominal circumference (−10%), shoulder height (−7%), tibia (−7%), and metatarsal lengths (−5%), with reduced body mass index (−12%), and relative sparing of head width (−5%) (P < 0.05 for all) (Table 1). PR did not alter AGRs (data not shown) but increased neonatal FGRs (AGR relative to size at birth) for weight (+24%), tibia (+15%), and metatarsal (+18%) lengths, hindlimb (+23%), and abdominal circumferences (+19%) between birth and 45 days (P < 0.05 for all). PR also increased current FGRs for weight (days 5 to 30), crown-rump length (days 5 to 30), abdominal circumference (days 5 to 25), and metatarsal length (days 5 to 15 and day 25) (P < 0.05 for all) (Fig. 1). Body weight before postmortem tended to be higher in male than female lambs (P = 0.06), but it did not differ between control and PR lambs (control male: 21.4 ± 0.7 kg, PR male: 20.2 ± 0.8 kg, control female: 16.3 ± 0.8 kg, PR female 16.0 ± 0.7 kg, P = 0.33). PR increased absolute perirenal fat mass (control: 92.8 ± 13.9 g, PR: 133.3 ± 10.7 g, P = 0.035) and relative perirenal fat mass (control: 0.54 ± 0.04%, PR: 0.74 ± 0.06%, P = 0.022). PR tended to increase absolute visceral fat mass (control: 309.6 ± 36.6 g, PR: 406.8 ± 38.5 g, P = 0.086), and increased relative visceral fat mass (control: 1.64 ± 0.18%, PR: 2.24 ± 0.21%, P = 0.04).
Effect of PR on circulating TH in the fed state.
In the fed state (days 5 to 40), plasma total T4 did not change with age and was not altered by PR (Fig. 2). Plasma total T3 decreased with age (P < 0.0001) and was reduced at days 35 and 40 compared with that at day 5 in the fed state (P < 0.05 for both) (Fig. 2). In the fed state, PR did not alter plasma total T3 overall (Fig. 2). Plasma total T3/T4 ratio declined with age (P < 0.0001) and was reduced at days 20, 25, 35, and 40 compared with that at day 5 in the fed state (P < 0.05 for all) (Fig. 2). PR did not alter the plasma total T3/T4 ratio overall but increased plasma total T3/T4 ratio at day 20 (P < 0.05) in the fed state (Fig. 2). There were no differences between males and females in plasma total T3 and T4 concentrations or the total T3/T4 ratio in the fed state nor did any interactions between sex, PR, and age influence these parameters (data not shown).
Effect of PR on the circulating TH response to fasting.
Plasma total T3 and T4 concentrations and the ratio of T3 to T4 on days 8 and 30 after short-term fasting did not differ between PR and control lambs (P > 0.05 for all, Fig. 2).
PR did not alter plasma-free T3 (control: 9.7 ± 0.3 pmol/l; PR 10.0 ± 0.4 pmol/l) or T4 (control: 21.7 ± 1.4 pmol/l; PR 23.7 ± 1.4 pmol/l) concentrations and the free T3/T4 (control: 0.35 ± 0.16; PR 0.36 ± 0.18), free T3/total T3 (control: 6.5 ± 0.4; PR 6.8 ± 0.3), and free T4/total T4 (control: 0.31 ± 0.02; PR 0.29 ± 0.02) ratio at 30 days of age. There were no differences between males and females in plasma-free and total T3 and T4 concentrations or free and total T3/T4 ratios in the fasted state, nor were there significant interactions between sex and PR (data not shown).
Associations between circulating THs and size at birth in the fed and fasted state.
In control lambs, plasma total T4 concentrations in the fed state throughout the study did not correlate with size at birth (Fig. 3), whereas those in the fasted state at days 8 and 30 correlated positively with size at birth in terms of weight, CRL, and shoulder height (Fig. 3) and tibia and metatarsal lengths (data not shown). Similarly, in PR lambs, plasma total T4 concentrations in the fed state generally did not correlate with size at birth (Fig. 3), whereas those in the fasted state at day 8 correlated positively with size at birth in terms of CRL and shoulder height (Fig. 3).
In control lambs, plasma total T3 concentrations in the fed state during the first month of life (days 5, 15, 25) usually correlated negatively with size at birth, in terms of weight, CRL, shoulder height (Fig. 3), and metatarsal length (data not shown). In PR lambs, plasma total T3 in the fed state during the first month of life also correlated negatively with size at birth, in terms of weight, CRL, and shoulder height (Fig. 3), as well as metatarsal and tibia lengths (data not shown) but predominantly at days 15, 20, and 25 (Fig. 3). In the fasted state, plasma T3 concentrations during the first month of life did not correlate with size at birth in either control or PR lambs, except for PR lambs at day 8, in which T3 concentration was positively related to shoulder height at birth (Fig. 3).
In control lambs, the plasma total T3/T4 ratio in the fed state during the first month of life did not correlate with size at birth (data not shown). In PR lambs, the plasma total T3/T4 ratio in the fed state during the first month of life (days 5 and 25) correlated negatively with size at birth in terms of weight (r = −0.91, n = 4, P < 0.05; r = −0.69, n = 8, P < 0.05), CRL (r = −0.92, n = 4, P < 0.05; r = −0.74, n = 8, P < 0.05), metatarsal (r = −0.91, n = 4, P < 0.05; r = −0.64, n = 8, P < 0.05) and radius/ulna length (r = −0.97, n = 4, P < 0.05; r = −0.73, n = 8, P < 0.05), skull length (r = −0.88, n = 4, P < 0.05; r = −0.71, n = 8, P < 0.05), abdominal (r = −0.91, n = 4, P < 0.05; r = −0.66, n = 8, P < 0.05) and thoracic circumference (r = −0.92, n = 4, P < 0.05; r = −0.82, n = 8, P < 0.05). In control or PR lambs, plasma total T3/T4 ratios in the fasted state at days 8 and 30 did not correlate with any measure of size at birth (data not shown). Free T3 and free T4 concentrations in the fasted state at 30 days of age were not related to size at birth. In all lambs, plasma-free T3 to free T4 ratio correlated positively with size at birth for abdominal circumference (r = 0.29, n = 34, P = 0.03) and negatively with placental weight (r = −0.49, n = 34, P = 0.02). In control lambs, plasma-free T3 to free T4 ratio correlated positively with size at birth for weight (r = 0.36, n = 22, P = 0.05), abdominal (r = 0.45, n = 22, P = 0.02), and thoracic (r = 0.44, n = 22, P = 0.02) circumferences. Plasma-free T3 to free T4 ratio did not correlate with size at birth in PR lambs.
Associations between circulating THs and postnatal growth in the fed and fasted state.
AGR for all lambs in terms of weight correlated positively with plasma total T3 concentrations in fed lambs at day 35 (r = 0.39), and plasma total T4 concentrations at days 30 (fasted; r = 0.37), 35 (fed; r = 0.48), and 40 (fed; r = 0.45) (P < 0.05 for all). AGR for all lambs in terms of CRL correlated positively with plasma total T3 concentrations in fed lambs at days 15 (r = 0.48), 20 (r = 0.51), and 35 (r = 0.42), and plasma total T4 concentrations in fed lambs at day 20 (r = 0.64) (P < 0.05 for all).
In PR lambs, AGR in terms of tibia length correlated positively with plasma total T3 concentrations in fed lambs at day 35 (r = 0.75, n = 9, P < 0.05). In PR lambs, absolute growth rate in terms of abdominal circumference correlated positively with plasma total T3 and T4 concentrations in the fasted state at day 8 (r = 0.76, n = 8, P < 0.05, for both) and with that of T3 at day 30 (r = 0.60, n = 9, P < 0.05).
Neonatal FGRs did not correlate with plasma total T4 concentrations during the first month of life in either control or PR lambs in the fed or fasted states (data not shown). In control and PR lambs, neonatal FGR in terms of weight correlated positively with plasma-free T4 concentrations in fasted lambs at day 30 (controls r = 0.43, P < 0.05, PR r = 0.46, P < 0.05) (Fig. 4).
In control and PR lambs, neonatal FGRs for weight (control: r = 0.92, P = 0.0005; PR: r = 0.80, P = 0.025) and CRL (control: r = 0.88, P = 0.0005; PR: r = 0.67, P = 0.05) correlated positively with plasma total T3 in fed lambs at day 15 (Fig. 5). In PR, but not control lambs, neonatal FGR in terms of weight correlated positively with plasma total and free T3 concentrations in fasted lambs at day 30 (controls r = 0.23, NS, PR r = 0.57, P < 0.05; controls r = 0.23, NS, PR r = 0.57, P < 0.05, respectively) (Fig. 4).
Current FGR for any parameters of size generally did not correlate with plasma total T4 in either control or PR lambs in the fed or fasted states throughout the first month of life (data not shown). However, in PR lambs in the fed state at day 20, current FGR for CRL correlated positively with plasma total T4 (r = 0.92, n = 7, P < 0.005), whereas current FGR for metatarsal length correlated positively with plasma total T4 (r = 0.66, n = 7, P < 0.05) (Fig. 6).
In control lambs, current FGR for weight correlated positively with plasma total T3 concentrations in fed lambs at days 15 (r = 0.61, n = 9, P < 0.05), 25 (r = 0.63, n = 9, P < 0.05), 35 (r = 0.69, n = 9, P < 0.05), and 40 (r = 0.81, n = 9, P < 0.05) (Fig. 6). Similarly, in control lambs, current FGR for CRL correlated positively with plasma total T3 concentrations in fed lambs at day 5 (r = 0.69, n = 8, P < 0.05), 15 (r = 0.78, n = 9, P < 0.05), and 25 (r = 0.75, n = 9, P < 0.05) (Fig. 6). In control lambs, current FGR for metatarsal length (control: r = 0.66, P = 0.05; PR: r = 0.15, NS) correlated positively with plasma total T3 concentrations in fed lambs at day 15 (Fig. 6).
In PR lambs, current FGR for weight correlated positively with plasma total T3 concentrations in fed lambs at day 15, but negatively with plasma total T3 concentrations in fed lambs at day 40 (Fig. 6). In PR lambs, current FGR for CRL correlated positively with plasma total T3 concentrations in fed lambs at day 20 (r = 0.69, n = 7, P < 0.05) (Fig. 6). Similarly, in PR lambs, current FGRs for metatarsal length correlated positively with plasma total T3 concentrations in fed lambs at day 15 (r = 0.42) and 25 (r = 0.44) (P < 0.05 for all). In PR lambs, current FGRs for tibia length correlated positively with plasma total T3 in fed lambs at day 35 (r = 0.74, n = 7, P < 0.05) (Fig. 6).
This study has shown that in the first month of life in sheep, PR alters TH abundance, between 15 and 25 days of age during the period of catch-up growth. Previous studies showed that, plasma T3 and T4 were reduced by IUGR in the first day of life in sheep (7, 28), which we now show does not persist. Rather, there is normal plasma T4, increased plasma T3 and an increased ratio of plasma T3 to T4 in early postnatal life after IUGR, possibly due to increased TH activation. Because this increase in circulating T3 is of delayed onset, between 15 and 25 days of life, which is mostly after the period of maximal catch-up growth, it may not have a primary role in the earliest onset catch-up growth. However, subsequent catch-up growth throughout the first month of life is correlated with and may partly depend on circulating total or free T3 and free T4. Because we measured free T3 and free T4 using assays designed for human plasma, because of the lack of assays validated for the sheep, it is possible that not all interference from circulating TH binding proteins was removed. Nevertheless, the correlations between growth rate and free TH concentrations in the present study suggest that this did provide a good measure of bioactive TH.
Plasma T3 and the ratio of plasma total T3 to T4 increase in the first 8 h of life in the sheep and other species (3, 7, 28), reflecting, in part, an increase in TSH, which stimulates T4 and T3 secretion from the thyroid, as well as the loss of the placenta with its high deiodinase type III (DIII) activity, which inactivates T3 and T4 (4). Subsequently, in sheep, plasma T3 and the ratio of plasma T3 to T4 increase to a maximum at 1 to 2 wk of age and then decline (7). Our current findings are consistent with this ontogeny, as plasma T4 did not change, whereas plasma T3 and the ratio of total T3 to T4 were maximal in samples collected between 5 and 15 days of age, then decreased subsequently in the young lamb. A broadly similar pattern occurs in the human infant, in which serum-free and total T3 increase to maximal levels at 50 to 80 days of age before decreasing (28). These changes in circulating TH levels and the relative amounts of T3 and T4 presumably reflect the secretion of T4 and T3 by the thyroid under TSH stimulation, and the production of T3 in extrathyroidal tissues (28). The relative abundance of different forms of TH is regulated by 5′ deiodination catalyzed by deiodinase type I (DI) and type II (DII), with the former, in particular contributing to circulating T3 (reviewed in Ref. 4). Together with DIII, these selenodeiodinases also inactivate T3 and T4 and help determine their levels in blood (reviewed in Ref. 4). We still know little of how these particular determinants of T4 and T3 activation and inactivation change and are regulated during development in sheep or humans, but an adverse prenatal environment, which restricts fetal growth, may persistently affect these and hence peripheral THs.
Although plasma total T4 levels in the young lamb in the fed state were unaffected by placental restriction or size at birth, they did decrease with size at birth in the fasted state. In lambs exhibiting spontaneous fetal growth restriction, plasma total T4 and T3 levels in the first few hours of life and at 11 days of age are reduced and correlate positively with birth weight and with subsequent postnatal growth in terms of weight, during the first month of life (7). Therefore, a mild deficit in circulating T4 may persist for several weeks after significant IUGR, because of reduced production or increased activation and metabolism. Our findings provide support for the latter mechanism, as plasma T3 increased with decreasing size at birth at various ages in control and/or placentally restricted lambs, implying that being small at birth enhances activation of T4 to T3. Also, consistent with increased activation of T4 to T3 after fetal growth restriction, we observed an increased plasma T3/T4 ratio with decreasing size at birth in PR lambs during the first month of life. In the human IUGR infant, serum total T4 is reduced and serum total T3 increases in the first 2 mo of life, suggesting an increased total T3 to T4 ratio, which resembles the pattern observed here and in previous studies of experimental and spontaneous IUGR in the neonatal sheep (28). It may be that in the young lamb and human infant, restriction of fetal growth increases deiodinase activity, contributing to the observed increases in plasma T3 and the plasma T3 to T4 ratio.
Placental restriction alters the metabolic environment and endocrine state of the fetus, but the particular changes and factors that initiate programming of altered production and activation of TH postnatally and the mechanisms responsible are unknown (4). In the rat, induction of perinatal hypothyroidism in mothers and their offspring alters the hypothalamic-pituitary-thyroid axis, producing hypothyroidism in offspring, which persists into adult life, with reduced plasma T4 and T3, despite normal plasma TSH (41). In contrast, perinatal hypothyroidism in the pig increases plasma total and free T3 during the first day of life, but whether this persists was not determined (3). As PR in the sheep also causes fetal hypothyroidism (23, 38, 39), this endocrine perturbation may alter TH production and activation postnatally, as occurs in these other species, albeit by unknown mechanisms. Other neuroendocrine consequences of PR could also contribute to enhanced activation of TH in the young lamb. Factors known to promote DI activity and/or activation of T4 to T3 include T3 itself, growth hormone (GH), and glucocorticoids, which have effects that appear to vary with developmental stage (14, 51, 52). In humans, low birth weight is characterized by signs of GH resistance and increased glucocorticoid activity in postnatal life (9, 33), which would promote TH activation. Similarly, we have previously shown that the male PR sheep is GH resistant, while the females are GH deficient (20). Spontaneous IUGR in sheep is characterized by increased circulating cortisol in the first 48 h of life, although whether this persists is unclear (7).
In the current study, postnatal growth in the first month of life was related to circulating TH abundance, particularly after 15 days of age in the young lamb, consistent with the requirement of neonatal growth for TH (8, 15, 54, 56). FGRs for soft and skeletal tissues were predicted by plasma-free T4 and free and total T3, but not plasma total T4, suggesting that they are driven in part by the more biologically active forms of TH (27). Interestingly, for catch-up growth in terms of weight, a higher rate of neonatal or current fractional growth was seen at a given plasma concentration of TH in the PR lambs compared with the control lambs. One possibility is that the PR lambs are more “sensitive” than normal to growth-promoting actions of TH actions, which could be brought about by increased tissue expression of TH receptors (THRs) or increased activation of T4 via altered expression of deiodinases (4, 48). Certainly, increased expression of THRs in skeletal muscle has been observed in the very young pig in response to experimentally induced postnatal hypothyroidism (24). Whether any of these changes occur in other tissues and persist and impact on growth is unclear (24). Increased T3 abundance may also contribute to the altered body composition of the PR lamb, which had increased visceral adiposity at 1 mo of age. The twin or triplet neonatal lamb that has experienced fetal growth restriction is also relatively fat compared with singleton lambs at any given body weight (2, 22). Furthermore, in humans, low birth weight predicts reduced skeletal tissue or lean tissue mass (46) and increased relative central adiposity in later life (37). This suggests a greater proportion of tissue accretion in adipose tissue in early postnatal life following IUGR; in which increased T3 production and action may have a role, as this promotes differentiation of preadipocytes into adipocytes and the accumulation of triglyceride (47).
Although there were consistently higher rates of neonatal and current FGRs for weight in PR lambs, at any given plasma TH concentration, this was not the case for growth of skeletal tissues. This implies that the sensitivity of skeletal tissue growth to THs did not differ between control and PR lambs and suggests that the primary determinant of increased growth of long bones or the vertebral column in the young PR lamb, is increased abundance of free T3 and T4 and of total T3 in circulation, rather than any changes in determinants of local TH action in the growth plate and elsewhere within skeletal tissues (44). Increased abundance and action of THs may also influence postnatal growth after IUGR indirectly, by promoting synthesis and secretion of insulin-like growth factor-I (5) or by inhibiting the activity of 11β-hydroxysteroid dehydrogenase-1 and local production of cortisol from cortisone and hence its inhibitory actions on growth (8, 36, 50).
In conclusion, placental restriction in the sheep reduced size at birth and increased postnatal growth rates in terms of weight and long bones in lambs. Plasma TH was associated with absolute and fractional growth of soft and skeletal tissue in lambs during early postnatal life, particularly in those undergoing catch-up growth. Thyroid hormone has long been known to be essential for infant growth, and this study has now shown that the increased rates of growth seen in postnatal life after IUGR may be partly due to the increased abundance of TH and possibly increased sensitivity of tissues to TH.
We thank Dr. Tim Butler, Anne Jurisevic, and Dr Michael Adams for technical assistance with surgery. This project was funded by the National Health and Medical Research Council of Australia.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society