Dose response effects of postnatal hydrocortisone on growth and growth factors in the neonatal rat

Steroids 140 (2018) 1–10

Contents lists available at ScienceDirect

Steroids journal homepage: www.elsevier.com/locate/steroids

Dose response effects of postnatal hydrocortisone on growth and growth factors in the neonatal rat

T

Maria A. Abrantesa,b,c,1, Arwin M. Valenciaa,b,d,2, Fayez Bany-Mohammedb, Jacob V. Arandae,f, ⁎ Kay D. Beharrya,b,e,f,3, a

Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Long Beach Memorial Medical Center, Long Beach, CA 90806, USA Division of Neonatal-Perinatal Medicine, Department of Pediatrics, University of California, Irvine Medical Center, Orange, CA 92868, USA c Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Kaiser Permanente, Anaheim, CA 92806, USA d Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Saddleback Memorial Medical Center, Laguna Hills, CA 92653, USA e Department of Pediatrics, Division of Neonatal-Perinatal Medicine, State University of New York, Downstate Medical Center, Brooklyn, NY 11203, USA f Department of Ophthalmology, Division of Neonatal-Perinatal Medicine, State University of New York, Downstate Medical Center, Brooklyn, NY 11203, USA b

A R T I C LE I N FO

A B S T R A C T

Keywords: Hydrocortisone Growth hormone Insulin Insulin-like growth factor-I Leptin Postnatal growth Rats

Background and purpose: Hydrocortisone (HC), at different dosages, is used in critically ill newborns for lung stability, blood pressure support, and prevention of chronic lung disease (CLD). Its long-term effects on postnatal growth are not well studied. We hypothesized that early exposure to high doses of HC adversely affects growth, growth factors, metabolic hormones, and neurological outcomes, persisting in adulthood. Experimental design: Rat pups received a single daily intramuscular dose of HC (1 mg/kg/day, 5 mg/kg/day, or 10 mg/kg/day on days 3, 4 & 5 postnatal age (P3, P4, P5). Age-matched controls received equivalent volume saline. Body weight, linear growth, and neurological outcomes were monitored. Animals were sacrificed at P21, P45, and P70 for blood glucose, insulin, IGF-I, GH, leptin, and corticosterone levels. Liver mRNA expression of IGFs and IGFBPs were determined at P21 and P70. Memory and learning abilities were tested using the Morris water maze test at P70. Results: HC suppressed body weight and length at P12, P21 and P45, but by P70 there was catchup overgrowth in the 5 and 10 mg/kg/day groups. At P70 blood insulin, IGF-I, GH, and leptin levels were low, whereas blood glucose, and liver IGFs and IGFBPs were high in the high dose groups. High HC also caused delayed memory and learning abilities at P70. Conclusions: These data demonstrate that while higher doses of HC may be required for hemodynamic stability and prevention of CLD, these doses may result in growth deficits, as well as neurological and metabolic sequelae in adulthood.

1. Introduction Extremely low birth weight preterm infants with respiratory distress often show evidence of adrenal insufficiency during the first week of life [1]. Refractory hypotension is frequent in very low-birth weight infants, whose hypothalamic–pituitaryadrenal axis has been suggested to be

immature [2]. In general, sicker and smaller infants have relatively low cortisol levels while in the neonatal intensive care unit [3]. Severe and prolonged hypotension is associated with increased mortality and central nervous system morbidity in critically ill preterm infants [4,5]. Attempts have been made to normalize blood pressure (BP), cardiac output (CO), and organ perfusion. In most hypotensive infants, cautious volume administration and the early use of low to medium dosages of

⁎ Corresponding author at: Department of Pediatrics & Ophthalmology, Neonatal-Perinatal Medicine Clinical & Translational Research Labs, State University of New York, Downstate Medical Center, 450 Clarkson Avenue, Box 49, Brooklyn, NY 11203, USA. E-mail addresses: [email protected] (F. Bany-Mohammed), [email protected] (J.V. Aranda), [email protected] (K.D. Beharry). 1 Present address: Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Kaiser Permanente, Anaheim, CA 92806, USA. 2 Present address: Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Saddleback Memorial Medical Center, Laguna Hills, CA 92653, USA. 3 Present address: Department of Pediatrics, Division of Neonatal-Perinatal Medicine, State University of New York, Downstate Medical Center, Brooklyn, NY 11203, USA; and Department of Ophthalmology, Division of Neonatal-Perinatal Medicine, State University of New York, Downstate Medical Center, Brooklyn, NY 11203, USA.

https://doi.org/10.1016/j.steroids.2018.08.003 Received 3 July 2018; Received in revised form 8 August 2018; Accepted 13 August 2018 Available online 22 August 2018 0039-128X/ © 2018 Elsevier Inc. All rights reserved.

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

12 litters at P3 and were treated with a daily IM dose of 1 mg/kg/day HC (physiologic), 5 mg/kg/day (anti-inflammatory), or 10 mg/kg/day (stress) on P3, P4 & P5 (36 pups per treatment). A control group received equivalent volume sterile normal saline, IM. Total body weight and linear growth (nose to tail) were measured before treatment on P3, and post-treatment on P12, P21, P45, and P70. Neurological tests were performed before treatment, and at P12 and P21 post treatment. Food and water intake were determined from P21 (time of weaning from the dam) to P45 by gender. Food intake was determined by weighing the pellets using an electronic weighing balance, at baseline and at the end of each week. Water intake was determined by measuring the volume of water at baseline using a graduated cylinder, at baseline and at the same time each day. Pups were euthanized at P21, P45 and P70 (12 pups/treatment/age group). Prior to sacrifice at P70, rats were tested for memory and learning abilities using the Morris water maze test. At sacrifice, blood was collected for analysis of glucose, insulin, IGF-I, GH, leptin and corticosterone. Liver samples were analyzed at P21 and P70 for mRNA expression of IGF-I, IGF-II, IGFBP-1 and IGFBP-3. The brain, heart, lungs, liver, kidneys, pancreas, and spleen were removed and weighed.

dopamine or epinephrine are effective in stabilizing the cardiovascular status and renal function. However, a subgroup of hypotensive preterm infants, do not respond to vasopressor inotropes. Critically ill preterm infants with vasopressor resistance are administered hydrocortisone (HC) to improve BP and renal function [6–9]. In these HC-treated infants, BP increases by 2 h after the first dose, and vasopressor inotrope requirement decreases by 6–12 h of treatment [7,9]. In addition to administering HC for physiologic deficiency, HC at different dosages is used in critically ill newborn infants for lung stability and prevention of chronic lung disease (CLD). While physiologic doses of 1 mg/kg may improve survival without bronchopulmonary dysplasia [10]; neurodevelopmental impairments at 2 years of age are not reduced [11–13]. Higher doses of 5 mg/kg appear to be more beneficial without adverse neurological outcomes [14–16], but may have detrimental effects on postnatal growth and cardiovascular outcomes in adult life. There is a relative paucity of information regarding the effects of postnatal HC on anthropometric growth in preterm infants, compared to the numerous reports on brain volume and weight. A recent study by Tijsseling et al. [17] showed significant alterations in growth patterns for weight, height, and head circumference of preterm infants exposed to GCs. These growth impairments of HC, particularly in low birth weight for gestational age neonates may further predispose to metabolic syndrome in adulthood [18,19]. The mechanism(s) underlying the effects of early postnatal exposure to HC on postnatal growth may involve alterations in growth factors such as insulin-like growth factor (IGF)-I, insulin, and growth hormone (GH). IGFs are pro-insulin-like polypeptides which markedly stimulate cell division and differentiation; and are major stimuli for pre- and postnatal growth [20]. The 3 peptide hormones in the IGF family are insulin, IGF-1 and IGF-2 and have 50% of their amino acids in common [21]. IGFs bind to IGF receptors (IGFR-1 and -2) and are modulated by six IGF-binding proteins (IGFBP-1 to -6) which are synthesized in the liver [22,23]. IGF deficiency has been shown to cause marked fetal growth retardation [24]. In the brain, IGF-1 is a potent inducer of oligodendrocyte development which is responsible for the synthesis of myelin [25,26]. Both IGF-I and IGF binding protein-3 (IGFBP-3) regulate early postnatal growth. Low serum IGF-I levels, measured in preterm infants, is associated with impaired growth, and development of retinopathy of prematurity as well as other complications [27–29]. IGF-I is produced in the liver, a major target for GH, which induces the production of IGF-I in the liver [30]. Insulin is a regulator of GH receptor and IGFBP-1, an endogenous inhibitor of IGF action. Serum cortisol is directly related to IGFBP-1 [31], suggesting that higher cortisol levels secondary to exogenous GC administration, may suppress IGF-1 and growth through induction of IGFBP-1. We therefore conducted a series of experiments to test the hypothesis that high doses of HC have lasting negative effects on the somatic growth, growth factors, metabolic hormones, and neurological outcomes which may persist into adulthood.

2.2. Sample collection At euthanasia, blood samples were collected for glucose levels using a glucometer. Blood samples were also collected for plasma and serum in appropriate tubes for insulin, IGF-I, GH, leptin, and corticosterone levels. For plasma samples, blood was collected in tubes containing EDTA and centrifuged at 3000 rpm for 20 min at 4 °C. The resulting plasma was aspirated and transferred to eppendorf tubes and frozen at −20 °C until assay. For serum samples, blood was collected in tubes containing no preservative and placed on ice to clot for 30 min. The samples were centrifuged at 3000 rpm for 20 min at 4 °C, and the resulting serum was aspirated and transferred to eppendorf tubes and frozen at −20 °C until assay. Animals were not fasted prior to blood sample collection for insulin and glucose measurements. Organ weights (brain, heart, lungs, liver, kidneys, spleen, and pancreas) were determined, and liver samples (100 mg) were collected, placed in tubes containing 1.0 mL TriZol reagent and snap frozen at −80 °C until analysis of IGF-I, IGF-II, IGFBP-1 and IGFBP-3 mRNA expression. 2.3. Assay of serum IGF-I, GH & leptin levels Levels of IGF-1, GH, and letptin in serum samples were determined using commercially available enzyme immunoassay kits from R&D Systems (Minneapolis, MN, USA) according to the manufacturer’s protocol. 2.4. Assay of plasma insulin Insulin levels in the plasma were determined using enzyme immunoassay kits purchased from Cayman Chemicals (Ann Arbor, MI, USA) according to the manufacturer’s protocol.

2. Materials and methods 2.1. Experimental design

2.5. Assay of plasma corticosterone All experiments were approved by the Institutional Animal Care and Use Committee, Long Beach Memorial Medical Center, Long Beach, CA. Animals were cared for according to the guidelines outlined by the Guide for the Care and Use of Laboratory Animals (National Research Council). Euthanasia of the animals was conducted according to the guidelines of the American Veterinary Medical Association (AVMA Panel). Certified infection-free timed-pregnant Sprague Dawley rats (200–300-gram body weight) were purchased from Charles River (Hollister, CA) and allowed to deliver spontaneously at term (22 days gestation). The pregnant rats (n = 12) were housed in individual cages and allowed to stabilize for 48 h under controlled environmental conditions with free access to food and water. Rat pups were pooled from

Plasma corticosterone levels were determined using enzyme immunoassay kits purchased from Enzo Life Sciences (Farmingdale, NY, USA) according to the manufacturer’s protocol. 2.6. Isolation of total RNA Total cellular RNA in the liver was extracted by homogenization using a polytron homogenizer (Brinkman Instruments, Inc., Westbury, N.J.). The homogenates were centrifuged at 12,000 rpm for 20 min at 4 °C. The supernatant was collected and kept at room temperature for 5 min after which 200 μl chloroform per milliliter was added. The 2

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

and slightly moving ahead. The expected response was rhythmic flexion and extension. The tactile placing reflex was tested by holding the rat with the forepaws touching the edge of a table. The expected response was flexion followed by extension of paws. The negative reflex was tested by placing the rat with its head downward on a 30° slope. The expected response was turning to face up the slope within 60 s. The free-fall righting reflex was tested by dropping the rat back down from 20 cm on cotton wool. The expected response was landing on all four feet. Responses were recorded as 0 (no response), 1 (mild response), or 2 (full response).

mixture was shaken vigorously for 15 s and centrifuged at 12,000 rpm for 20 min at 4 °C. The aqueous phase was removed and the RNA was precipitated with isopropanol and collected by centrifugation at 12,000 rpm for 10 min at 4 °C. The supernatant was discarded and the RNA pellet was washed with 75% ethanol, centrifuged at 8000 rpm at 4 °C for 5 min, air-dried and solubilized in diethylpyrocarbonate (DEPC)-treated water. The integrity of the RNA was determined by gel electrophoresis in 1% agarose gel stained with ethidium bromide (EtBr). The purity of the RNA was assessed by the ratio of absorbance at 260 nm and 280 nm. The total RNA concentration was estimated by spectrophotometric measurements at 260 nm assuming 40 μg of RNA per milliliter equals one absorbance unit. The total RNA yield was diluted in DEPC-treated water to 1 μg/μL total RNA for all samples.

2.10. Morris water maze test for memory and learning Rats are natural swimmers. Free swim trials have proven to be an excellent tool to differentiate the involvement of hippocampal mineralcorticoid (MR) and glucocorticoid (GR) receptors. The advantage of the Morris water maze (MWM) test is that there are no local clues or scent trails. The Morris water maze test is the most widely used test for learning and memory in rodents. It consisted of a round tank (pool) of water about 2 feet deep with a clear plexiglass escape platform ½″ above the water (temperature of 26–27 °C). The rats were tested individually and placed in various areas of the tank with the tail-end lower and the head pointing toward the pool-side to minimize bias [35–37]. The time to reach the platform was recorded on 6 trials. The time limit per trial was 90 s to obtain good learning. With each subsequent trial the rats progressively become more efficient at locating the platform by learning the location of the platform.

2.7. Reverse transcriptase-polymerase chain reaction (RT-PCR) Two μg total RNA was reversely transcribed to complementary DNA (cDNA) using Muloney murine reverse transcriptase. Amplification of cDNA was performed using specific sense and anti-sense primers for rat GAPDH, IGF-I, IGF-II, IGFBP-1 and IGFBP-3 with AmpliTaq DNA polymerase (Perkin Elmer, Norwalk, CT). The sense and anti-sense primers were prepared by Life Technologies. The sense and anti-sense primer sequences GAPDH were 5′-CCT TCC TGC GCA TGG AGT CCT GG-3′ and 5′-GGA GCA ATG ATC TTG ATC TTC-3′, respectively. The sense and anti-sense primer sequences for IGF-I were 5′-TCC GCT GAA GCC TAC AAA GT-3′ and 5′-TTC CTC AAG CAG CAA AGG AT-3′, respectively (NIH Genebank #M15481). The sense and anti-sense primer sequences for IGF-I receptor were 5′-TCC GCT GAA GCC TAC AAA GT3′ and 5′-TTC CTC AAG CAG CAA AGG AT-3′, respectively (NIH Genebank #M27293). The sense and anti-sense primer sequences for IGF-II were 5′-CCC AGC GAG ACT CTG TGC GGA and GGA AGT ACG GCC TGA GAG GTA (NM_001190163). The sense and anti-sense primer sequences for IGFBP-I were 5′-CGG TTC TCA GCA TGA AGA GG-3′ and 5′-TGC TTT CTG TTG AGC GGC AC-3′, respectively [32]. The sense and anti-sense primer sequences for IGFBP-3 were 5′-CAG CAA CCT GAG TGC CTA CC-3′ and 5′-CTG TCT CCC GCT TAG ACT CG-3′, respectively [33]. The PCR cycle profile, which was carried out using a model 480 DNA thermal cycler (Perkin Elmer Cetus). Samples were denatured for 2 min at 94° following which they were amplified for 35 cycles using a profile of 94 °C for 30 s, 60 °C for 40 s and 72 °C for 40 s, then 72 °C for 2 min.

2.11. Statistical analysis A test for normality was first conducted using the Bartlett’s test. Normally distributed data were analyzed using one-way analysis of variance (ANOVA) with Dunnet’s post-hoc tests for comparison of the treated groups versus control. Non-normally distributed data was analyzed using Kruskall Wallis test with Dunn’s multiple comparison test. Data for neurological development were analyzed using the Fisher exact test. All data are expressed as mean ± SEM. A p-value of less than 0.05 was considered significant. Statistical analyses were accomplished with the use of SPSS, version 21; and all graphs were prepared with the use of GraphPad Prism, version 7. 3. Results

2.8. Densitometric scanning 3.1. Effect on somatic growth Gel electrophoresis of the PCR products was performed on 1.5% agarose gels stained with EtBr. The intensities of the bands were measured with the use of a GelDoc 1000 Darkroom Imager and Molecular Analyst software (BioRad Laboratories, Hercules, CA). The PCR fragments were identified according to their molecular mass using a DNA mass ladder (Perkin Elmer, Norwalk, CT). The amount of DNA in each specimen was quantitated by the integrated density of the product bands within a closed rectangle, which was then normalized to the density of the GAPDH bands. The data are expressed as mean IGF-I, IGFII, IGFBP-1 and IGFBP-3 to GAPDH ratio ± SEM (n = 4 samples/ group).

Pooling and randomization resulted in comparable body weight and length among all the treatment groups at P0. Fig. 1 shows the dose–response effect of early postnatal HC on total body weight. At P12, one week post treatment, mean total body weights were lower in all treatment groups compared to the placebo saline group. This effect remained sustained until P45. By P70, the groups treated with 5 and 10 mg/kg/day exhibited catchup overgrowth. Fig. 2 represents the effect of HC on body length. Similar reductions were noted in all treated groups at P12. Lower body length persisted in the groups exposed to 10 mg/kg/day at P21 and P45. Similar to body weight, by P70, mean body length was higher in the groups treated with the high doses. Rats are weaned from their mothers at P21 and around this age they generally consume solid food. At P21, only the kidney weights were smaller in all the treated groups. By P45, the brain, liver and kidneys were smaller in the treated groups. By P70, there were significant elevations in lung weights in the groups treated with 5 and 10 mg/kg/day (Table 1).

2.9. Physical and neurodevelopment testing Rats were assessed for neurological development using reflex-stimulus-responses as previously described [34]. The palmar grasp reflex was tested by gently stroking the palm of the forepaw with a thin probe. The expected response was flexing of the digits to grasp the probe. The plantar grasp reflex was tested by stroking the palm of the hindpaw with a thin probe. The expected response was flexing of the digits to grasp the probe. For the hopping reaction reflex (forepaws and hindpaws) the rats were held with forepaws or hindpaws touching the floor

3.2. Effect on food and water intake From P21 to P45, the overall food intake (grams) was 3

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

Fig. 1. Dose-response effects of postnatal hydrocortisone (HC) on total body weight at P12, P21, P45, and P70. HC-treated animals received intramuscular injections of HC (1, 5, or 10 mg/kg/day) on P3, P4 and P5. Age-matched littermates received equivalent volumes of sterile normal saline. Data are expressed as mean ± SEM (n = 36 pups/group).

152.5 ± 6.6) and water (1 mg/kg/day: 331.1 ± 13.5, p < 0.01; 5 mg/kg/day: 392.6 ± 18.4, p < 0.01; 10 mg/kg/day: 333.8 ± 17.7, p < 0.01 versus saline: 263.6 ± 11.8) compared to saline-treated rats. By contrast, HC-treated female rats consumed less food (1 mg/kg: 162.4 ± 9.9; 5 mg/kg/day: 154.0 ± 10.1; 10 mg/kg/day: 119.1 ± 5.4 (p < 0.01) versus saline: 176.3 ± 9.8) although significance was achieved only for the group treated with the high dose of 10 mg/kg/day. Similarly, water intake was lower in all HC-treated females (1 mg/kg: 297.5 ± 14.9; 5 mg/kg/day: 291.9 ± 16.9; 10 mg/

168.3 ± 8.89, 174.0 ± 10.1, 180.3 ± 10.5, and 155.1 ± 9.8 for the saline, and 1, 5, and 10 mg/kg/day groups, respectively, and was comparable among the groups. Water intake (mL) was 286.2 ± 13.9, 315.1 ± 15.1, 346.2 ± 116.6 (p < 0.05), and 285.0 ± 16.4, respectively. Water intake was significantly higher in the 5 mg/kg/day group compared to controls. Analysis by gender revealed that all HCtreated male rats consumed significantly greater amounts of food (1 mg/kg/day: 194.6 ± 10.0, p < 0.01; 5 mg/kg/day: 202.9 ± 9.5, p < 0.01; 10 mg/kg/day: 191.1 ± 9.6, p < 0.01 versus saline:

Fig. 2. Dose-response effects of postnatal hydrocortisone (HC) on body length (nose to tail) at P12, P21, P45, and P70. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 36 pups/group). 4

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

Table 1 Effect of HC on Organ Weights. Saline

Table 3 Neurological Test at P21 (Weaning).

HC (1 mg/kg)

HC (5 mg/kg)

HC (10 mg/kg)

P21 (weaning): Brain 1.46 ± 0.07 Heart 0.3 ± 0.03 Lungs 0.47 ± 0.04 Liver 1.98 ± 0.22 Kidneys 0.73 ± 0.06

1.43 ± 0.05 0.26 ± 0.05 0.49 ± 0.06 1.54 ± 0.3 0.5 ± 0.07*

1.3 ± 0.13 0.25 ± 0.03 0.51 ± 0.04 1.5 ± 0.27 0.5 ± 0.06*

1.3 ± 0.09 0.28 ± 0.03 0.5 ± 0.04 1.5 ± 0.1 0.52 ± 0.07*

P45 (Adolescent): Brain 1.98 ± 0.16 Heart 0.96 ± 0.17 Lungs 1.26 ± 0.24 Liver 11.1 ± 2.1 Kidneys 2.6 ± 0.6

1.68 ± 0.15** 0.83 ± 0.17 1.11 ± 0.2 8.6 ± 1.6** 1.6 ± 0.2**

1.78 ± 0.11* 0.85 ± 0.16 1.19 ± 0.18 8.5 ± 1.4** 1.9 ± 0.4**

1.72 ± 0.17** 0.81 ± 0.13 1.13 ± 0.18 8.2 ± 1.1** 1.7 ± 0.3**

P70 (Adult): Brain 1.78 ± 0.18 Heart 1.1 ± 0.25 Lungs 1.18 ± 0.24 Liver 11.5 ± 0.24 Kidneys 2.4 ± 0.65

1.8 ± 0.15 1.1 ± 0.25 1.27 ± 0.25 10.8 ± 2.3 2.4 ± 0.58

1.8 ± 0.12 1.2 ± 0.285 1.54 ± 0.33* 12.7 ± 3.2 2.8 ± 0.75

1.82 ± 0.09 1.26 ± 0.32 1.53 ± 0.35* 11.9 ± 3.1 2.7 ± 0.61

Saline

HC (1 mg/kg)

HC (5 mg/kg)

HC (10 mg/kg)

Palmar grasp: 0 0 1 0 2 100%

0 5% 95%

0 5% 95%

0 15% 75%

Plantar grasp: 0 0 1 0 2 100%

0 25%** 75%**

5% 20%* 80%*

0 20%* 80%*

Negative 0 1 2

10% 11% 89%

5% 10% 90%

5% 10% 80%*

0 0 100%

0 0 100%

0 0 100%

geotaxis: 0 0 100%

Freefall righting: 0 0 1 0 2 100%

** p < 0.01 (Fisher’s exact test); n = 36 pups per group.

** p < 0.01 (Fisher’s exact test); n = 12 pups per group.

kg/day: 236.3 ± 9.7, p < 0.01 versus saline: 306.2 ± 16.0) with statistical difference achieved only for the group treated with 10 mg/ kg/day.

3.3. Effect on neurological and memory outcomes Neurological tests were conducted at P3 before treatment and at P12 and P21 post treatment. The Morris water maze test for memory and learning was conducted only at P70 in the adult rats. There were no differences in neurological tests among the groups at P3. However, at P12, low-dose HC (1 mg/kg/day) caused a lower percentage of animals to receive a full negative geotaxis reflex score (Table 2). At P21, the plantar grasp reflex was retarded in all treated groups and the negative geotaxis reflex remained retarded in the 10 mg/kg/day group (Table 3). The results of the Morris water test are presented in Fig. 3. Animals treated with low HC (1 mg/kg/day) took the longest time to reach the platform on the second trial. However, by the fourth to sixth trials animals exposed to 5 and 10 mg/kg/day remained consistently slow to achieve the platform.

Fig. 3. Dose-response effects of postnatal hydrocortisone (HC) on memory and learning abilities assessed by the Morris water maze test at P70. Animals were tested for a total of 6 trials. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 12 pups/group).

3.4. Effect on glucose and insulin levels Glucose and insulin levels are presented in Fig. 4. At P21 and P70 glucose levels were consistently higher in the 5 and 10 mg/kg/day groups. Plasma insulin levels were higher with the 1 mg/kg/day dose and lower with the 5 mg/kg/day dose at P21. By P70, glucose levels remained higher and insulin level markedly lower in the 5- and 10-mg treated groups.

Table 2 Neurological Test at P12 (7 days post treatment). Saline

HC (1 mg/kg)

HC (5 mg/kg)

HC (10 mg/kg)

Palmar grasp: 0 0 1 0 2 100%

0 7% 93%

0 6% 94%

0 6% 94%

Plantar grasp: 0 3 1 0% 2 97%

0 17% 83%

0 13% 87%

0 20%* 80%*

Negative 0 1 2

0 13% 87%

0 20% 80%

0 43%** 57%**

3% 6% 91%

0 0 100%

0 3% 97%

geotaxis: 0 10% 90%

Freefall righting: 0 0 1 0 2 100%

3.5. Effect on serum IGF-I and GH levels Serum IGF-I and GH levels are presented in Fig. 5. At P21, IGF-I levels were higher in the groups treated with the 1 and 10 mg/kg/day doses, and GH were lower with the 5 and 10 mg/kg/day doses. By P70, all doses resulted in lower IGF-I levels, but GH was higher in response to the 1 and 5 mg/kg/day doses, and lower in response to the 10 mg/kg/ day dose. 3.6. Effect on liver IGF-I and IGF-II mRNA expression Liver IGF-I and -II mRNA expression is presented in Fig. 6. At P21, only the high dose of 10 mg/kg/day resulted in a 2-fold upregulation of liver IGF-I mRNA. In contrast, liver IGF-II mRNA expression was lower with the all doses. At P70, both the 5 and 10 mg/kg/day doses resulted in higher liver IGF-I and –II mRNA expression.

** p < 0.01(Fisher’s exact test); n = 36 pups per group. 5

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

Fig. 4. Dose-response effects of postnatal hydrocortisone (HC) on blood glucose (upper panel) and plasma insulin (lower panel) levels at P21 and P70. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 12 samples/group).

3.7. Effect on liver IGFBP-1 and IGFBP-3 mRNA expression

3.8. Effect on leptin and corticosterone levels

Liver IGFBP-1 and IGFBP-3 mRNA expression is presented in Fig. 7. At P21, IGFBP-1 and IGFBP-3 mRNA expression were downregulated with the 1 and 5 mg/kg/day doses and upregulated with 10 mg/kg/day dose. By P70, all HC doses resulted in sustained upregulation of IGFBP1 and IGFBP-3 mRNA expression in the liver.

The effect of HC on leptin and corticosterone levels are presented in Fig. 8. At P21, all HC doses caused a decline in leptin levels and increase in corticosterone levels. At P70, the suppressive effect on leptin remained sustained, but the effect on corticosterone subsided and all treated groups were comparable with the placebo saline control.

Fig. 5. Dose-response effects of postnatal hydrocortisone (HC) on serum IGF-I (upper panel) and serum GH (lower panel) levels at P21 and P70. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 12 samples/group). 6

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

Fig. 6. Dose-response effects of postnatal hydrocortisone (HC) on mRNA expression of IGF-I (upper panel) and IGF-II (lower panel) in the liver at P21 and P70. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 4 samples/group).

4. Discussion

periods. However, during adulthood, there was catchup overgrowth in the groups that received high HC doses; 2) high doses of HC caused sustained neurodevelopmental deficits evidenced by delayed memory and learning abilities, as well as retarded plantar grasp and negative geotaxis reflexes persisting until adulthood; and 3) high doses of HC caused sustained suppressive effects on insulin and glucose regulation, metabolic hormones, and growth factors which may account for the body overgrowth at P70. These findings support our hypothesis and suggest that exposure to supra-physiological GCs at a critical time of

The present study tested the hypothesis that early exposure to high doses of HC have lasting negative effects on somatic growth, metabolic hormones, and neurological outcomes persisting into adulthood. We therefore examined the dose–response effects of early postnatal HC over the suckling, weanling, adolescent and adult period of treated rats. The major findings of this study are: 1) early postnatal HC at all doses caused significant growth deficits during the weanling and adolescent

Fig. 7. Dose-response effects of postnatal hydrocortisone (HC) on mRNA expression of IGFBP-1 (upper panel) and IGFBP-3 (lower panel) in the liver at P21 and P70. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 4 samples/group). 7

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

Fig. 8. Dose-response effects of postnatal hydrocortisone (HC) on serum leptin (upper panel) and plasma corticosterone (lower panel) levels at P21 and P70. Groups are as described in Fig. 1. Data are expressed as mean ± SEM (n = 12 samples/group).

velocity in infancy was positively associated with greater waist circumference in adulthood, indicating larger visceral obesity [50], a predictor of cardiovascular disease [51]. The GH/IGF-I/insulin system, particularly, IGF-I, is an important regulator of postnatal growth and development [52]. IGF-I is induced by pituitary GH in the liver to promote growth, along with its binding proteins [53], and accounts for about 75% of all circulating IGFs [54]. Together with its receptor, it is expressed in almost all tissues for autocrine/paracrine purposes [55]. Although IGF-I is predominantly produced by the liver, studies have shown that liver-derived IGF-I is not required for postnatal growth, suggesting that local production of IGF-I may be more important than liver-derived circulating IGF-I for body growth [56]. In this study, HC increased the mRNA expression of IGF-I, but the levels were lower in the serum. This may suggest increased binding to IGFBPs thus making free IGF-I unavailable in the serum. IGF1 availability is tightly regulated by its binding proteins which increase IGF-1 half-life from minutes to hours, and shuttles IGF-I to specific target tissues [57]. Approximately 90% of IGF-1 is bound to IGFBP-3, the primary hepatic-derived IGFBP [58]. In rats, the fetal serum profile, characterized by high IGF-II and IGFBP-2, is replaced around the third week of life by the adult-type profile of high IGF-I and IGFBP-3, with a dramatic reduction in IGF-II and IGFBP-2 [59]. Studies have shown that elevating cortisol to prepartum levels results in downregulation of IGF-I [60] and IGF-II [61] in fetal muscle. Although in our study, HC at all doses elevated plasma corticosterone levels at P21 with concurrent downregulation in liver IGF-II expression, supporting previous findings, our data further showed that high doses of HC results in higher liver mRNA expression of IGF-I and II with no appreciative alterations in corticosterone levels. IGF-II originates in the liver and while it regulates growth in-utero, it is associated with obesity and diabetes in adults, and is inversely correlated with IGFBP-1 [62]. Here we showed lower liver IGF-II and higher IGFBP-1 with the high HC dose of 10 mg/kg/day, confirming previous reports. High mRNA expression of IGFs in the liver concurrent with elevated liver IGFBP-3 mRNA and lower serum IGF-I may be responsible for the catchup overgrowth noted in the high dose HC groups at P70, and may be an indicator of metabolic dysfunction [63] and obesity [64].

development in preterm infants may have lasting negative effects, thus raising concerns for early origins of adult diseases such as obesity, diabetes, hypertension, and cardiovascular disease. Reports of significant neurodevelopmental effect of dexamethasone have led to the use of the less potent HC as an alternative for blood pressure support and CLD treatment/prevention in preterm infants [38–40]. HC has been shown to influence fetal and postnatal growth and growth factors [17,41,42], and targets the neonatal brain [43]. While most clinical studies report no immediate and short-term adverse events of HC on neurological outcomes, there are no long-term followup studies. Furthermore, information regarding long-term effects on growth and factors that regulate growth is lacking. Our study showed that despite significant weight impairment, there was noticeable brain weight sparing with all doses. Regardless, there were distinct differences in the neurological outcomes and memory and learning abilities. In rats, the cerebellum plays an important role in the regulation of complex movement patterns at P14 [44]. The negative geotaxis reflex is attributed to vestibular function and cerebellar development [34], which has been shown to be negatively influenced by GCs [45], supporting our findings. The plantar grasp reflex is involved in sensory thresholds and signaling from the brainstem [46]. Our results suggest abnormalities in cerebellar and brain stem function with HC, which may also account for the lasting deficits in recognition memory and learning abilities at P70. Interestingly, the high doses of HC resulted in higher body weight and length at P70 suggesting catch-up growth or even overweight/ obesity. This effect of HC on early deficits in weight accretion followed by increased growth velocity was recently demonstrated in preterm infants receiving postnatal GCs for CLD and monitored until 4 years of age [17]. We now report similar growth trajectories in adulthood that is likely related to alterations in growth factors and metabolic hormones. It is well known that catchup growth after a period of growth restriction is an important risk factor for early origins of adult diseases, including obesity, diabetes, hypertension, metabolic disorders, and cardiovascular disease [47–50]. The growth impairment from exogenous GCs appears to be maximal during the initial months of treatment [51]. In a follow-up study of 3778 patients from 2 to 31 years of age, height 8

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

Physiologic cortisol induces IGFBP-1 which downregulates IGF-I bioactivity and subsequently, growth [62]. In our study, liver IGFBP-1 expression was elevated with HC, but with no subsequent reduction in IGF-I. This may suggest that these endocrine relationships may be absent with supraphysiological levels of GCs. GCs are known regulators of insulin, carbohydrates, lipids, and protein metabolism. Although GCs counteract insulin, they act synergistically to promote lipid storage [65,66]. GCs exposure is also a well-established risk factor for insulin resistance, impairment of glucose tolerance and development of diabetes [67]. In our study, insulin was elevated with the lowest dose, but declined with the two higher doses at P70, with concurrent elevations in glucose. This low insulin response to high HC doses suggest beta cell failure as previously shown [68]. Leptin is produced by adipose tissue and plays important roles in energy expenditure, insulin sensitivity, and maintain normal body weight by limiting food intake. Dysregulation of the leptin feedback system can lead to obesity. The decline in leptin levels at P70 may add to the catchup overgrowth phenomenon and adverse metabolic outcomes in the high dose groups. One limitation is that we did not measure thyroid hormones (THs) which play a role in growth and maturation of the central nervous system (neurogenesis, gliogenesis, myelogenesis), and may influence the GH-IGF axis [69]. TH deficiency in preterm infants is associated with neurodevelopmental deficits [70]. While caution is necessary when extrapolating data from animals to the human situation, there are still developmental similarities between species, that can provide valuable information. In contrast to humans, rodents experience their brain growth spurt after birth, and at P10, the rodent brain is roughly equivalent to that of the full term human brain [71], therefore, at P3-P5, when the HC doses were administered, the rat pup’s brain corresponded to that of a 26–28 week preterm infant [72]. Therefore, given these developmental parallels and common underlying mechanisms, the results of this investigation show that early postnatal exposure to high doses of HC can alter the developmental trajectories of certain organs such as the brain and liver to result in adverse neurodevelopmental outcomes and disturbances in the metabolic hormones that regulate growth, glucose homeostasis, and weight balance, with the potential for further adverse events in adult life.

[9] S. Noori, P. Friedlich, P. Wong, M. Ebrahimi, B. Siassi, I. Seri, Hemodynamic changes after low-dosage hydrocortisone administration in vasopressor-treated preterm and term neonates, Pediatrics 118 (2006) 1456–1466. [10] O. Baud, L. Maury, F. Lebail, D. Ramful, F. El Moussawi, C. Nicaise, V. ZupanSimunek, A. Coursol, A. Beuchée, P. Bolot, P. Andrini, D. Mohamed, C. Alberti, PREMILOC trial study group. Effect of early low-dose hydrocortisone on survival without bronchopulmonary dysplasia in extremely preterm infants (PREMILOC): a double-blind, placebo-controlled, multicentre, randomised trial, Lancet 387 (2016) 1827–1836. [11] O. Baud, C. Trousson, V. Biran, E. Leroy, D. Mohamed, C. Alberti, PREMILOC Trial Group. Association between early low-dose hydrocortisone therapy in extremely preterm neonates and neurodevelopmental outcomes at 2 Years of Age, JAMA 317 (2017) 1329–1337. [12] K.L. Watterberg, J.S. Gerdes, C.H. Cole, S.W. Aucott, E.H. Thilo, M.C. Mammel, R.J. Couser, J.S. Garland, H.J. Rozycki, C.L. Leach, C. Backstrom, M.L. Shaffer, Prophylaxis of early adrenal insufficiency to prevent bronchopulmonary dysplasia: a multicenter trial, Pediatrics 114 (2004) 1649–1657. [13] K.L. Watterberg, M.L. Shaffer, M.J. Mishefske, C.L. Leach, M.C. Mammel, R.J. Couser, S. Abbasi, C.H. Cole, S.W. Aucott, E.H. Thilo, H.J. Rozycki, C.B. Lacy, Growth and neurodevelopmental outcomes after early low-dose hydrocortisone treatment in extremely low birth weight infants, Pediatrics 120 (2007) 40–48. [14] S.E. de Jong, F. Groenendaal, F. van Bel, K.J. Rademaker, Pulmonary effects of neonatal hydrocortisone treatment in ventilator-dependent preterm infants, Int. J. Pediatr. 2011 (2011) 783893. [15] N.A. Parikh, K.A. Kennedy, R.E. Lasky, J.E. Tyson, Neurodevelopmental outcomes of extremely preterm infants randomized to stress dose hydrocortisone, PLoS One 10 (2015) e0137051. [16] L.W. Doyle, R.A. Ehrenkranz, H.L. Halliday, Postnatal hydrocortisone for preventing or treating bronchopulmonary dysplasia in preterm infants: a systematic review, Neonatology 98 (2010) 111–117. [17] D. Tijsseling, M. Ter Wolbeek, J.B. Derks, W.P. de Vrie, C.J. Heijnen, F. van Bel, E.J.H. Mulder, Neonatal corticosteroid therapy affects growth patterns in early infancy, PLoS One 13 (2018) e0192162. [18] K.J. Kersbergen, L.S. de Vries, B.J. van Kooij, I. Išgum, K.J. Rademaker, F. van Bel, P.S. Hüppi, J. Dubois, F. Groenendaal, M.J. Benders, Hydrocortisone treatment for bronchopulmonary dysplasia and brain volumes in preterm infants, J. Pediatr. 163 (2013) 666–671. [19] D.J. Barker, The developmental origins of adult disease, J. Am. Coll. Nutr. 23 (2004) 588S–595S. [20] M.J. Finken, M.G. Keijzer-Veen, F.W. Dekker, M. Frolich, E.T. Hille, J.A. Romijn, J.M. Wit, Preterm birth and later insulin resistance: effects of birth weight and postnatal growth in a population based longitudinal study from birth into adult life, Diabetologia 49 (2006) 478–485. [21] R. Holmes, Fetal and maternal plasma IGF and binding proteins in pregnant with or retarded fetal growth, Early Hum. Dev. 49 (1997) 7–17. [22] D. Le Roith, Seminars in medicine of the Beth Israel Deaconess Medical Center. Insulin-like growth factors, New Engl. J. Med. 336 (1997) 633–640. [23] J. Baker, Role of insulin and IGF in embryonic and postnatal growth, Cell 75 (1993) 73–82. [24] A.N. Brooks, D.M. Hagan, D.C. Howe, Neuroendocrine regulation of pituitaryadrenal function during fetal life, Eur. J. Endocrinol. 135 (1996) 153–165. [25] F.A. McMorris, T.M. Smith, IGF-1/somatomedin C, a potent inducer of oligodendrocyte development, Proc. Natl. Acad. Sci. U.S.A. 83 (1986) 822–826. [26] P. Ye, A.J. D’Ercole, Insulin-like growth factor actions during development of neural stem cells and progenitors in the central nervous system, J. Neurosci. Res. 83 (2006) 1–6. [27] A. Hellström, D. Ley, I. Hansen-Pupp, B. Hallberg, C. Löfqvist, L. van Marter, M. van Weissenbruch, L.A. Ramenghi, K. Beardsall, D. Dunger, A.L. Hård, L.E. Smith, Insulin-like growth factor 1 has multisystem effects on foetal and preterm infant development, Acta Paediatr. 105 (2016) 576–586. [28] E. Engström, A. Niklasson, K.A. Wikland, U. Ewald, A. Hellström A,, The role of maternal factors, postnatal nutrition, weight gain, and gender in regulation of serum IGF-I among preterm infants, Pediatr. Res. 57 (2005) 605–610. [29] A. Hellström, E. Engström, A.L. Hard, K. Albertsson-Wikland, B. Carlsson, A. Niklasson, C. Lofqvist, E. Svensson, S. Holm, U. Ewald, G. Holmström, L.E. Smith, Postnatal serum insulin-like growth factor I deficiency is associated with retinopathy of prematurity and other complications of premature birth, Pediatrics 112 (2003) 1016–1020. [30] M. Cornblath, M.L. Parker, S.H. Reisner, A.E. Forbes, W.H. Daughaday, Secretion and metabolism of growth hormone in premature and full-term infants, J. Clin. Endocrinol. 25 (1965) 209–218. [31] S. Cianfarani, IGF-1, BP-1 are related to cortisol in human cord blood, Euro. J. Endocrinol. 138 (1998) 524–529. [32] F.A. Hills, R. Crawford, S. Harding, The effect of labor on maternal and fetal levels of IGFBP-1, Am. J. Obstet. Gynecol. 171 (1994) 1292–1295. [33] R. Mewar, F.A. McMorris, Expression of insulin-like growth factor binding protein messenger RNAs in developing rat oligodendrocytes and astrocytes, J. Neurosci. Res. 50 (1997) 721–728. [34] A. Gramsenbergen, E.J.H. Mulder, The influence of betamethasone and dexamethasone on motor development in young rats, Pediatr. Res. 44 (1998) 105–110. [35] S. Dalm, S. Grootendorst, E.R. de Kloet, M.S. Oitzl, Quantification of swim patterns in the Morris water maze, Behav. Res. Methods. Instrum. Comput. 32 (2000) 134–139. [36] M.S. Oitzl, E.R. de Kloet, M. Joëls, W. Schmid, T.J.,, Spatial learning deficits in mice with a targeted glucorticoid receptor gene disruption, Eur. J. Neurosci. 9 (1997) 2284–2296.

Conflicts of interest None. Acknowledgement Financial support: This work was made possible through a grant from Memorial Health Services Research Foundation, Long Beach, CA, 90806 USA (Grant #007-01). References [1] E.F. Fernandez, K.L. Watterberg, Relative adrenal insufficiency in the preterm and term infant, J. Perinatol. 29 (2009) S44–S49. [2] P.C. Ng, C.H. Lee, C.W. Lam, K.C. Ma, T.F. Fok, I.H. Chan, E. Wong, Transient adrenocortical insufficiency of prematurity and systemic hypotension in very low birthweight infants, Arch. Dis. Child Fetal Neonatal Ed. 89 (2) (2004) F119-26cc. [3] R.E. Grunau, D.W. Haley, M.F. Whitfield, J. Weinberg, W. Yu, P. Thiessen, Altered basal cortisol levels at 3, 6, 8 and 18 months in infants born at extremely low gestational age, J. Pediatr. 150 (2007) 151–156. [4] A.M. Watkins, C.R. West, R.W. Cooke, Blood pressure and cerebral hemorrhage and ischemia in very low birth weight infants, Early Hum. Dev. 19 (1989) 103–110. [5] R.F. Goldstein, R.J. Thompson, J.M. Oehler, J.E. Brazy, Influence of acidosis, hypoxemia, and hypotension on neurodevelopmental outcome in very low birth weight infants, Pediatrics 95 (1995) 238–243. [6] H.J. Helbock, R.M. Insoft, F.A. Conte, Glucocorticoid-responsive hypotension in extremely low birth weight newborns, Pediatrics 92 (1993) 715–717. [7] I. Seri, R. Tan, J. Evans, Cardiovascular effects of hydrocortisone in preterm infants with pressor-resistant hypotension, Pediatrics 107 (2001) 1070–1074. [8] P. Tantivit, N. Subramanian, M. Garg, R. Ramanathan, R.A. deLemos, Low serum cortisol in term newborns with refractory hypotension, J. Perinatol. 19 (1999) 352–357.

9

Steroids 140 (2018) 1–10

M.A. Abrantes et al.

synthesized by multiple tissues in the fetus, Dev. Biol. 75 (1980) 315–328. [55] K. Sjögren, J.L. Liu, K. Blad, S. Skrtic, O. Vidal, V. Wallenius, D. LeRoith, J. Törnell, O.G. Isaksson, J.O. Jansson, C. Ohlsson, Liver-derived insulin-like growth factor I (IGF-I) is the principal source of IGF-I in blood but is not required for postnatal body growth in mice, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 7088–7092. [56] D.R. Clemmons, Role of insulin-like growth factor binding proteins in controlling IGF actions, Mol. Cell. Endocrinol 140 (1998) 19–24. [57] S.M. Firth, R.C. Baxter, Cellular actions of the insulin-like growth factor binding proteins, Endocr. Rev. 23 (2002) 824–854. [58] S.M. Donovan, Y. Oh, H. Pham, R.G. Rosenfeld, Ontogeny of serum insulin-like growth factor binding proteins in the rat, Endocrinology 125 (1989) 2621–2627. [59] A.J. Forhead, J. Li, S.J.C. Gilmour, M.J. Dauncey, A.L. Fowden, Thyroid hormones and the mRNA for the growth hormone receptor and insulin-like growth factors in skeletal muscle of fetal sheep, Am. J. Physiol. Endocrinol. Metab. 282 (2002) E80–E86. [60] J. Li, J.C. Saunders, R.S. Gilmour, M. Silver, A.L. Fowden, Insulin-like growth factor-II messenger ribonucleic acid expression in fetal tissues of the sheep during late gestation: effects of cortisol, Endocrinology 132 (1993) 2083–2089. [61] V. Nobili, A. Alisi, N. Panera, C. Agostoni, Low birth weight and catch-up-growth associated with metabolic syndrome: a ten-year systematic review, Pediatr. Endocrinol. Rev. 6 (2008) 241–247. [62] J. Frystyk, Free insulin-like growth factors – measurements and relationships to growth hormone secretion and glucose homeostasis, Growth Horm. IGF. Res. 14 (2004) 337–375. [63] M. Desai, M. Beall, M.G. Ross, Developmental origins of obesity: programmed adipogenesis, Curr. Diab. Rep. 13 (2013) 27–33. [64] J.M.P. Holly, The physiological role of IGFBP-1, Acta Endocrinol. 124 (1991) 55–62. [65] L. Gathercole, I. Bujalska, P. Stewart, J.W. Tomlinson, Glucocorticoid modulation of insulin signaling in human subcutaneous adipose tissue, J. Clin. Endocrinol. Metab. 92 (2007) 4332–4339. [66] L.L. Gathercole, S.A. Morgan, I.J. Bujalska, P. Guest, L. Abrahams, D.M. Smith, P.M. Stewart, G.G. Lavery, J.W. Tomlinson, Regulation of lipogenesis by glucocorticoids and insulin in human adipose tissue, PLoS One 6 (2011) e26223. [67] M. Fichna, P. Fichna, Glucocorticoids and beta-cell function, Endokrynol. Pol. 68 (2017) 568–573. [68] K. Matsumoto, H. Yamasaki, S. Akazawa, H. Sakamaki, M. Ishibashi, N. Abiru, S. Uotani, H. Matsuo, Y. Yamaguchi, K. Tokuyama, S. Nagataki, High-dose but not low-dose dexamethasone impairs glucose tolerance by inducing compensatory failure of pancreatic beta-cells in normal men, J. Clin. Endcrinol. Metab. 81 (1996) 2621–2626. [69] J. Rodriguez-Arnao, J.P. Miell, R.J. Ross, Influence of thyroid hormones on the GHIGF-I axis, Trends. Endocrinol. Metab. 4 (1993) 169–173. [70] S.E. Scratch, R.W. Hunt, D.K. Thompson, Z.M. Ahmadzai, L.W. Doyle, T.E. Inder, P.J. Anderson, Free thyroxine levels after very preterm birth and neurodevelopmental outcomes at age 7 years, Pediatrics 133 (2014) e955–e963. [71] J. Dobbing, The later growth of the brain and its vulnerability, Pediatrics 53 (1974) 2–6. [72] A. Whitelaw, M. Thoresen, Antenatal steroids and the developing brain, Arch. Dis. Child. Fetal. Neonatal. Ed. 83 (2000) F154–F157.

[37] L.B. Tucker, A.G. Velosky, J.T. McCabe, Applications of the Morris water maze in translational traumatic brain injury research, Neurosci. Biobehav. Rev. 88 (2018) 187–200. [38] M. van der Heide-Jalving, P.J. Kamphuis, M.J. van der Laan, J.M. Bakker, V.M. Wiegant, C.J. Heijnen, S. Veen, F. van Bel, Short- and long-term effects of neonatal glucocorticoid therapy: is hydrocortisone an alternative to dexamethasone? Acta. Paediatr. 92 (2003) 827–835. [39] B.A. Yoder, M. Harrison, R.H. Clark, Time-related changes in steroid use and bronchopulmonary dysplasia in preterm infants, Pediatrics 124 (2009) 673–679. [40] S. Higgins, P. Friedlich, I. Seri, Hydrocortisone for hypotension and vasopressor dependence in preterm neonates: a meta-analysis, J. Perinatol. 30 (2010) 373–378. [41] A.L. Fowden, J. Li, A.J. Forhead, Glucocorticoids and the preparation for life after birth: are there long-term consequences of the life insurance? Proc. Nutr. Soc. 57 (1998) 113–122. [42] J. Li, A.J. Forhead, M.J. Dauncey, R.S. Gilmour, A.L. Fowden, Control of growth hormone receptor and insulin-like growth factor-I expression by cortisol in ovine fetal skeletal muscle, J. Physiol. 541 (2002) 581–589. [43] D.M. Vázquez, C.R. Neal Jr, P.D. Patel, N. Kaciroti, J.F. López, Regulation of corticoid and serotonin receptor brain system following early life exposure of glucocorticoids: long term implications for the neurobiology of mood, Psychoneuroendocrinology 37 (2012) 421–437. [44] S.M. Walker, M. Fitzgerald, G.J. Hathway, Surgical injury in the neonatal rat alters the adult pattern of descending modulation from the rostroventral medulla, Anesthesiology 122 (2015) 1391–1400. [45] M. Cotterell, R. Balazs, A.L. Johnson, Effects of corticosteroids on the biochemical maturation of rat brain: postnatal cell formation, J. Neurochem. 19 (1972) 2151–2161. [46] A. Gramsbergen, Consequences of cerebellar lesions at early and later ages: clinical relevance of animal experiments, Early Hum. Dev. 34 (1993) 79–87. [47] J.G. Eriksson, T. Forsén, J. Tuomilehto, P.D. Winter, C. Osmond, D.J. Barker, Catchup growth in childhood and death from coronary heart disease: longitudinal study, BMJ 318 (1999) 427–431. [48] M.B. Belfort, S.L. Rifas-Shiman, J. Rich-Edwards, K.P. Kleinman, M.W. Gillman, Size at birth, infant growth, and blood pressure at three years of age, J. Pediatr. 151 (2007) 670–674. [49] N. Soto, R.A. Bazaes, V. Pena, T. Salazar, A. Avila, G. Iniguez, K.K. Ong, D.B. Dunger, M.V. Mericq, Insulin sensitivity and secretion are related to catch-up growth in small-for-gestational-age infants at age 1 year: results from a prospective cohort, J. Clin. Endocrinol. Metab. 88 (2003) 3645–3650. [50] I. Tzoulaki, U. Sovio, D. Pillas, A.L. Hartikainen, A. Pouta, J. Laitinen, T.H. Tammelin, M.R. Jarvelin, P. Elliott, Relation of immediate postnatal growth with obesity and related metabolic risk factors in adulthood: the northern Finland birth cohort 1966 study, Am. J. Epidemiol. 171 (2010) 989–998. [51] Z. Hochberg, Mechanisms of steroid impairment of growth, Horm. Res. 58 (2002) 33–38. [52] T. Pischon, H. Boeing, K. Hoffmann, et al., General and abdominal adiposity and risk of death in Europe, N. Engl. J. Med. 359 (2008) 2105–2120. [53] G.A. Aguirre, J.R. De Ita, R.G. de la Garza, I. Castilla-Cortazar, Insulin-like growth factor-1 deficiency and metabolic syndrome, J. Transl. Med. 14 (2016) 3. [54] A.J. D’Ercole, G.T. Applewhite, L.E. Underwood, Evidence that somatomedin is

10