Investigation of the Use of Antioxidants to Diminish the Adverse Effects of Postnatal Glucocorticoid...

Fax +41 61 306 12 34E-Mail karger@karger.chwww.karger.com

Original Paper

Neonatology 2010;98:73–83 DOI: 10.1159/000275561

Investigation of the Use of Antioxidants to Diminish the Adverse Effects of Postnatal Glucocorticoid Treatment on Mortality and Cardiac Development

Alexandra Adler Emily J. Camm Jeremy A. Hansell Hans G. Richter Dino A. Giussani

Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge , UK

towards control values survival, growth symmetry the aortic lumen:total vessel area, and increased the cardiac expres-sion of Hsp90 relative to Dex. In addition, relative to controls, the decrease in the cardiac expression of eNOS was no lon-ger significant in DexCE animals (–20.3 8 14.4%, p 1 0.05). However, DexCE did not prevent growth retardation, cardiac 4-HNE upregulation (DexCE: +29%) or LV thinning (DexCE: 40.1 8 1.1 mm 3 ). Treatment of neonates with vitamins alone affected somatic growth and promoted thinner LV walls (CtrlCE: 39.9 8 0.5 mm 3 , p ! 0.05). Conclusions: Combined glucocorticoid and antioxidant therapy in premature infants may be safer than glucocorticoids alone in the treatmentof chronic lung disease. However, antioxidant therapy in healthy offspring is not recommended.

Copyright © 2010 S. Karger AG, Basel

Introduction

Chronic lung disease (CLD) is a major cause of mor-bidity and mortality in premature infants [1, 2] , affecting 22% of surviving extremely low birth weight infants ( ! 1,500 g). For more than 10 years, there have been no significant increases in survival without neonatal and long-term morbidity in this population of infants, high-

Key Words

Glucocorticoids � Oxidative stress � Prematurity

Abstract

Background: In premature infants, glucocorticoids amelio-rate chronic lung disease, but have adverse effects on growth and the cardiovascular system. Glucocorticoid excess pro-motes free radical overproduction and vascular dysfunction. Objectives: We hypothesized that the adverse effects of postnatal glucocorticoid therapy are secondary to oxidative stress and that antioxidant treatment would diminish un-wanted effects. Methods: Male rat pups received a clinically relevant course of dexamethasone (Dex), or Dex with vita-mins C and E (DexCE), on postnatal days 1–6 (P1–6). Controls received saline (Ctrl) or saline with vitamins (CtrlCE). Results: At P21, Dex reduced survival (Ctrl: 96 vs. Dex: 70%) and pro-moted asymmetric growth restriction (ponderal index, Ctrl: 6.3 8 0.1 g � mm –3 ! 10 –5 vs. Dex: 7.4 8 0.2 g � mm –3 ! 10 –5 ), both p ! 0.05. Dex increased cardiac oxidative stress (protein expression: 4-HNE +20%, Hsp90 –42% and eNOS–54%), induced left ventricle (LV) wall thinning (LV wall vol-ume: Ctrl: 47.2 8 1.2 mm 3 vs. Dex: 38.9 8 1.7 mm 3 ) and de-creased the ratio of the aortic lumen:total vessel area (Ctrl: 0.74 8 0.01 vs. Dex: 0.66 8 0.02), all p ! 0.05. DexCE restored

Received: July 28, 2009 Accepted after revision: August 18, 2009 Published online: January 13, 2010

formerly Biology of the Neonate

Dino A. Giussani, PhD Department of Physiology, Development and Neuroscience, University of Cambridge Downing Street, Cambridge CB2 3EG (UK) Tel. +44 1223 333 894, Fax +44 1223 333 840 E-Mail dag26 @ cam.ac.uk

© 2010 S. Karger AG, Basel1661–7800/10/0981–0073$26.00/0

Accessible online at:www.karger.com/neo

Adler/Camm/Hansell/Richter/Giussani Neonatology 2010;98:73–8374

lighting the need for improved therapies [3] . A major con-tributor to the development of CLD is excessive inflam-mation and thus glucocorticoids, such as dexamethasone (Dex), are used to prevent or reduce the severity of this complication due to their anti-inflammatory properties [4–6] . Glucocorticoids are also effective in treating CLD because they enhance lung maturation and surfactant production, thereby improving respiratory function, fa-cilitating weaning from mechanical ventilation and re-ducing mortality in premature infants [5–8] .

Despite the well-established beneficial effects of post-natal glucocorticoid therapy on lung maturation, there has been serious growing concern regarding their clinical use because of unwanted side effects on somatic growth and the development of important organ systems, such as the cardiovascular system. Treatment with Dex reduces somatic growth and weight gain in both premature hu-man infants [9–12] and newborn rats [13–15] . In prema-ture infants, Dex induces transient hypertension and hy-pertrophic cardiomyopathy, changes that resolve follow-ing the cessation of therapy [16–18] . In the rat neonate, Dex administered during the first few days of life also causes pronounced structural and functional changes to the cardiovascular system [19] . Dex inhibits the prolif-eration of cardiomyocytes [20] , and results in cardiomyo-cyte hypertrophy and reduced heart weight [21] , as well as systolic dysfunction and compensatory ventricular dilatation [22] . Long-term studies of cardiovascular func-tion in adult humans who received postnatal Dex as in-fants have not been conducted, however in rats, systolic dysfunction and reduced heart weight persist into old age, suggesting the need for cardiac screening in humans [23] . Dex has been the most frequently used glucocorti-coid, particularly in extremely premature ( ! 32 weeks) and low birth weight infants unlikely to survive without therapy [24] , although other glucocorticoids, including hydrocortisone, have also been used and studied [25] . While hydrocortisone appears to cause fewer unwanted adverse effects than Dex, it is also less potent in promot-ing the beneficial effects of glucocorticoids [25–27] . Dos-ages for postnatal Dex, or its stereoisomer betametha-sone, vary and the American Academy of Pediatrics has most recently recommended a 12-day tapering course be-ginning at 0.5 mg/kg therapy [24] .

Accumulating evidence suggests that glucocorticoids promote oxidative stress, both by enhancing the produc-tion of reactive oxygen species, including the superoxide anion, and by reducing the levels of endogenous antioxi-dants [28–31] . Oxidative stress plays a major role in the development of cardiovascular pathology including cor-

onary heart disease and atherosclerosis, type 2 diabetes, and hypertension [32–34] . Both glucocorticoids and re-active oxygen species are known to decrease the bioavail-ability of nitric oxide (NO), a potent vasodilator [32] , and studies have reported that oxidative stress and reductions in NO bioavailability may underlie the relationship be-tween glucocorticoids and cardiovascular dysfunction [28, 30] . In rats, Zhang et al. [31] demonstrated the im-portance of superoxide in glucocorticoid-induced hyper-tension by using tempol, a superoxide scavenger to pre-vent and partially reverse the Dex-induced hypertension. In humans, Iuchi et al. [28] found that glucocorticoid ex-cess is associated with impaired forearm reactive hyper-aemia, indicating vascular dysfunction, and that this ef-fect was largely resolved by administration of the anti-oxidant vitamin C.

We propose that the unwanted side effects of postnatal glucocorticoids on the cardiovascular system may in part be due to oxidative stress-induced depression of NO bio-availability. If true, combined treatment of premature in-fants with glucocorticoid and antioxidants may amelio-rate the unwanted side effects while maintaining the ben-eficial effects of glucocorticoid therapy in the postnatal period. Using a well-established model of glucocorticoid-induced cardiovascular dysfunction in the rat [21, 22] , we tested the interrelated hypotheses that postnatal treat-ment of newborn pups with Dex promotes oxidative stress, thereby leading to alterations in somatic growth and in the development of the cardiovascular system, and that co-administration of Dex with antioxidant vitamins ameliorates these adverse effects.

Methods

Ethical Approval The study was approved by the Cambridge University Ethical

Review Committee. All procedures were carried out under the UK Animals (Scientific Procedures) 1986 Act.

Animals and Experimental Design Twenty-five pregnant Wistar rats (Charles River, UK) with

timed gestations were individually housed under standard condi-tions (23 8 1 ° C, light:dark, 12: 12 h) with access to food (Special Diet Services, UK) and water. All dams delivered on day 22 of gestation (assigned postnatal day 0, P0). Litters were then divided into four treatment groups: control (Ctrl, n = 6), dexamethasone (Dex, n = 6), Dex with vitamins C and E (DexCE, n = 7), and con-trol with vitamins C and E (CtrlCE, n = 6).

Within 3–5 h of birth, pups were sexed and weighed, and lit-ters reduced to 8 pups per dam (4 males and 4 females) in order to standardize postnatal nutrition and maternal care. To account for sex differences, only male pups within each litter received

Antioxidants and Glucocorticoid Treatment

Neonatology 2010;98:73–83 75

treatment (Ctrl, n = 24; Dex, n = 24; DexCE, n = 28, and CtrlCE, n = 24). Male pups received two intraperitoneal (i.p.) injections daily (10 � l � g –1 for each) of some or all of the following solutions from Sigma (Sigma-Aldrich, UK): Dex (dexamethasone-21-phos-phate, disodium salt), vitamin C ( L -ascorbic acid), and vitamin E ( dl - � -tocopherol acetate). Two injections were used due to the different solubility of vitamin E (dissolved in groundnut oil) and vitamin C and Dex (both dissolved in 0.9% NaCl). Ctrl pups re-ceived injections of saline and groundnut oil for the duration of the treatment period, postnatal days 1–6 (P1–6; fig. 1 ). Dex pups received a 3-day, tapering course of Dex (0.5, 0.3, and 0.1 � g � g –1 � day –1 ) plus separate injections of oil on P1–3 and then only saline and oil from P4 to P6. DexCE pups received the same treat-ment as Dex pups, except that vitamins C (200 mg � kg –1 � day –1 ) and E (100 mg � kg –1 � day –1 ) were administered over the entire treatment period in addition to the 3-day tapering course of Dex. CtrlCE pups received injections of vitamin C and E from P1 to P6. Body weight from P0 to P7 and every other day thereafter was recorded; fractional growth rates (FGRs) were calculated from birth to weaning.

The dose and duration of Dex used in this study was derived from and is proportional to the 21-day tapering course of Dex used in human preterm infants to prevent or lessen CLD, starting at 0.5 mg � kg –1 and lasting up to 6 weeks [21] . The doses of vita-mins C and E used in this study were adopted from studies indi-cating successful antioxidant effects in adult Wistar rats at these concentrations [35] . Ascorbic acid and � -tocopherol were used in the present study as these vitamins are often combined to pro-mote antioxidant activity and they are reported to act synergisti-cally to provide optimal conditions for NO production [36–38] .

Tissue Collection During the treatment period, 10 male Dex pups, 2 male DexCE

pups and 1 male Ctrl pup died, leaving the following remaining male pups for study (Ctrl, n = 23; Dex, n = 14; DexCE, n = 26, and CtrlCE, n = 24). On P21, approximately half of these surviving male pups from each litter (Ctrl: n = 11; Dex: n = 7; DexCE: n = 12; CtrlCE: n = 12) were deeply anaesthetized (0.2 ml total volume,i.p., 100 mg � ml –1 ketamine, Fort Dodge Animal Health, UK and 20 mg � ml –1 xylazine, Millpledge Veterinary, UK). To assess the symmetry of growth, body weight (BW) and crown-rump length (CRL) were determined. Tissues were weighed and the heart was divided into the left ventricle + interventricular septum (LV + IVS), and the right ventricle (RV). Cardiac tissue was then snap frozen in liquid nitrogen and stored at –80 ° C for subsequent West-ern blot analysis. On P22, the remaining male pups (Ctrl:n = 12; Dex: n = 7; DexCE: n = 14; CtrlCE: n = 12) were deeply an-aesthetized as described above and perfused intracardially with a NaCl solution (10 m M PIPES, 139 m M NaCl, 2.7 m M KCl, 19.4 m M D -glucose, 7.5 � M PVP, pH 7.2) followed by 4% paraformaldehyde. Hearts were collected, and weighed. A segment of the thoracic aorta was also collected. Tissues were stored overnight at 4 ° C in 4% paraformaldehyde for subsequent stereological analysis.

Western Blotting Western blots were performed using 10- to 20- � g aliquots of

protein, resolved on 10–12% SDS-PAGE gels. Proteins were trans-ferred to polyvinylidene fluoride Immobilon-P membranes (Mil-lipore, UK) by electroblotting. Membranes were blocked at room temperature for 1 h with 5% dry skim milk in Tris-buffered-saline containing 1% Tween-20 (TBS-T; Sigma-Aldrich, UK). Purified

// P1 P3 P6 P21 P22

Ctrl

Frozentissue

collection

I I I I ISaline

Groundnut oil

CtrlCEVitamin C

Vitamin E

DexGroundnut oil

Dexamethasone Saline

DexCEVitamin E

Dexamethasone+ Vitamin C Vitamin C

Fixedtissue

collection

Fig. 1. Treatment protocol. Beginning on postnatal day 1 (P1), male pups received two daily injections of saline and ground-nut oil (Ctrl, n = 24), dexamethasone (Dex, n = 24), Dex in combination with vitamins C and E (DexCE, n = 28), or vitamins C and E (CtrlCE, n = 24). Dex was adminis-tered as a 3-day tapering course beginning on P1 (0.5, 0.3, and 0.1 � g � g –1 � day –1 ). On P21, tissues were snap frozen for subse-quent protein analysis. On P22, tissue was perfusion fixed for histological analysis.

Adler/Camm/Hansell/Richter/Giussani Neonatology 2010;98:73–8376

antibodies to � -actin (1: 40,000; Sigma-Aldrich, UK), endothelial nitric oxide synthase (eNOS; 1: 2,000; BD Transduction Labora-tories, UK), 4-hydroxynonenal (4-HNE; 1: 1,000; Alexis Bio-chemicals, USA), or heat-shock protein 90 (Hsp90; 1: 1,000; Stress-gen Bioreagents, UK) in 5% milk in TBS-T were added, and incu-bated at 4 ° C overnight. Membranes were washed in TBS-T, incubated for 1 h in a secondary antibody conjugated to horse-radish peroxidase (donkey anti-rabbit IgG or sheep anti-mouse IgG; 1: 10,000; GE Healthcare, UK) and washed in TBS-T. Proteins were visualized using enhanced chemiluminescence (Amersham, UK), exposed to X-ray film and films were developed (Fuji FPM100A Processor). Bands densities were quantified and ex-pressed relative to � -actin (ImageJ software, NIH).

Cardiovascular Stereology Analysis was performed on coded slides, with the observer

(A.A.) blind to treatment (BX-50 microscope; Computer Assisted Stereology Toolbox CAST; Olympus, Denmark). To account for shrinkage due to paraffin processing, the diameter of erythrocytes in heart sections at P22 was measured and compared to that ob-tained by measuring fresh erythrocytes from rat pups of the same age [39] . All measurements were corrected using this factor.

Fixed hearts were exhaustively sectioned at 5 � m using a mi-crotome (Leica RM 2235, Germany). To assess the total volume of the cardiac ventricles, 15 sections were selected an equal distance apart (approx. 200 � m), with the starting section being random-ly selected. Slides were stained using Masson’s trichrome. A point grid was superimposed on the sections and viewed using a ! 1.25 objective. The following compartment volumes were then ana-lysed by point counting: LV + IVS, left lumen (LL), RV wall, and right lumen (RL). Points falling on each compartment were count-ed and the Cavalieri principle [40] was applied in order to calcu-late estimated volumes:

V ( obj ) = t ! � a = t ! a ( p ) ! � P

where V ( obj ) is the estimated volume of the heart compartment, t is the total thickness of the heart [ t = number of sections ! sec-tion thickness], a ( p ) is the area associated with each point, and � P is the sum of points for that compartment. The thickness for the LV and RV free walls was also estimated. A line grid was super-imposed on sections to establish random start points for measur-ing distances between the inner and outer membrane using the method of orthogonal intercepts [41] .

Aortae were sectioned at 5 � m. Ten sections per animal were stained with Masson’s trichrome. The area of the wall and lumen were determined using the point grid system. Points falling on either the wall or lumen were counted and the areas were calcu-lated as:

A ( obj ) = a ( p ) ! � P

where A ( obj ) is the estimated area, a ( p ) is the area associated with each point, and � P is the sum of points falling on the relevant area, averaged over the 10 sections.

Calculations and Statistical Analysis FGR was calculated as the change in BW over the period of

interest divided by the number of days within that period. This value was then divided by the starting weight in grams for a final unit of g � day –1 � (g of starting weight) –1 . Ponderal index (PI) was

calculated as BW:CRL 3 . Data for BW and FGR were obtained from all surviving pups. To account for within-litter variation, no more than 2 male pups from anyone litter were used for biometry and organ weights and the molecular biology analysis; 1 male pup per litter was used for the histological analysis. Data are presented as mean 8 SEM. Ratios and percentages were arcsine trans-formed for statistical analysis. Survival data were analysed using Fisher’s exact test. Other data were analysed by one- or two-way ANOVA followed by the Tukey test, as appropriate. Significance was defined as p ! 0.05 (SigmaStat 2.0; SPSS, Inc., Chicago, Ill., USA).

Results

Survival Pups were all born spontaneously on day 22 of gesta-

tion. There was no significant difference in BW at birth be-tween the groups (Ctrl: 6.58 8 0.11 g, Dex: 6.55 8 0.14 g, DexCE: 6.48 8 0.07 g, CtrlCE: 6.43 8 0.08 g, p 1 0.05). Relative to controls, treatment with Dex from P1 to P3 markedly reduced survival (p ! 0.05; fig. 2 ). Co-admin-istration of vitamins C and E with Dex significantly in-creased survival compared to Dex-treated pups (p ! 0.05). Survival was not affected by administration of vitamins alone (p 1 0.05).

Weight Gain and Growth The BWs of control pups increased steadily during the

postnatal period ( fig. 3 a). Dex decreased BW gain, an ef-

0

Ctrl DexDexCE

CtrlCE

20

40

60

80

100

Surv

ival

(%)

*

Fig. 2. Survival rates. The percentage of pups that survived the treatment period in the control (Ctrl, n = 24), dexamethasone (Dex, n = 24), Dex with vitamins C and E (DexCE, n = 28) and control with vitamins C and E (CtrlCE, n = 24) groups. * p ! 0.05 vs. control (Fisher’s exact test).

Antioxidants and Glucocorticoid Treatment

Neonatology 2010;98:73–83 77

fect that was already evident 24 h following the beginning of treatment and continued until P21. This effect of Dex on BW gain was not altered by the addition of vitamins C and E. Vitamins alone decreased BW gain only on P13 and from P17 to P21. At P21, the BWs relative to Ctrl pups were reduced by 19.96 8 1.87% in Dex pups, 21.36 8 2.80% in DexCE pups, and 6.26 8 1.67% in CtrlCE pups (all p ! 0.05).

FGR was calculated for individual animals during the treatment period (P1–6), the second 2 weeks of life (P6–19), and the last 3 days before weaning (P19–21). In con-trol pups, FGR was highest during P1–6 and subsequent-ly decreased during P6–19 and P19–21 (p ! 0.05; fig. 3 b). Conversely, in Dex-treated pups, with or without vita-mins, FGR was the highest during P6–19 (p ! 0.05). In pups that received only vitamins, the FGR pattern was similar to that seen in control pups. Relative to controls, FGR in the Dex-treated pups was markedly reduced from P1 to P6, increased from P6 to P19, and reduced from P19 to P21 (all p ! 0.05). A similar trend was observed with the co-administration of vitamins C and E (p ! 0.05). Compared to controls, administration of vitamins alone significantly reduced FGR at P1–6 and P19–21 (p ! 0.05), but not between P6 and P19. In DexCE relative to Dex pups, FGR was decreased during P1–6 and increased dur-ing P6–19 (p ! 0.05).

Organ Weights and Biometry Absolute and relative organ weights at P21 are shown

in table 1 . Compared to controls, in Dex-treated pups, relative liver and kidney weights were decreased, relative lung and brain weights were increased (all p ! 0.05), and relative heart weight was unaffected. Treatment with Dex also increased two measures of asymmetric growth: the ratio of brain-to-liver weight and brain-to-kidney weight (all p ! 0.05). Administration of Dex with vita-mins C and E did not affect these changes, however, rel-ative heart weights were reduced (p ! 0.05), an effect due to a reduction in the weight of LV (p ! 0.05). Compared to controls, organ weights were not altered in the CtrlCE pups.

Relative to controls, Dex increased PI (Ctrl: 6.3 8 0.1 g � mm –3 ! 10 –5 vs. Dex: 7.4 8 0.2 g � mm –3 ! 10 –5 ,p ! 0.05) and decreased CRL (Ctrl: 100.3 8 0.7 mm vs. Dex: 88.4 8 1.4 mm, p ! 0.05). The ratio of the absolute brain weight-to-CRL (Brain:CRL) was also increased in Dex-treated pups (Ctrl: 0.0150 8 0.0002 g � mm –1 vs. Dex: 0.0159 8 0.0003 g � mm –1 , p ! 0.05). Additions of vita-mins C and E restored PI and Brain:CRL to control levels, but it did not significantly restore CRL (DexCE: 91.3 8 1.0 mm, p ! 0.05). Administration of vitamins C and E alone did not affect PI, CRL, or Brain:CRL.

0 2 4 6 8 10 12 14 16 18 20

CtrlDexDexCECtrlCE

a

0

10

20

30

40

50

60

70

Body

wei

ght (

g)

Days

Treatment

*

‡

‡

‡

‡

0

b P1–6 P6–19 P19–21

0.1

0.2

0.3

Frac

tiona

l gro

wth

rate

(g•d

ay–1

•g s

tart

ing

wei

ght–1

)

†

†

*

*

*

*

*

*#

#

*#

#

Ctrl DexDexCE

CtrlCE

Ctrl DexDexCE

CtrlCE

Ctrl DexDexCE

CtrlCE

Fig. 3. Postnatal body weights and growth. a Body weight (in g) was assessed daily from postnatal day 1–6 (P1–6, treatment pe-riod) and subsequently on every other day in the control (Ctrl,n = 23), dexamethasone (Dex, n = 14), Dex with vitamins C and E (DexCE, n = 26), and control with vitamins C and E (CtrlCE, n =

24) pups. Black bar signifies treatment period. b FGR was calcu-lated during three periods: P1–6, P6–19, and P19–21. * p ! 0.05 vs. Ctrl; ‡ p ! 0.05 CtrlCE vs. Ctrl; # p ! 0.05 vs. P1–6; † p ! 0.05 vs. Dex (two-way ANOVA + Tukey test).

Adler/Camm/Hansell/Richter/Giussani Neonatology 2010;98:73–8378

Western Blotting Compared to controls, administration of Dex in-

creased the cardiac expression of 4-HNE (Dex: +19.8 8 7.6%, p ! 0.05; fig. 4 a) and decreased the expression of Hsp90 (Dex: –42.4 8 12.7%, p ! 0.05; fig. 4 b) and eNOS (Dex: –54.4 8 12.0%, p ! 0.05; fig. 4 c). Relative to con-trols, in Dex-treated pups that received vitamins, 4-HNE remained upregulated (DexCE: +28.8 8 5.1%, p ! 0.05). The cardiac expression of Hsp90 remained reduced (DexCE: –21.6 8 3.0%, p ! 0.05), however, expression was significantly increased relative to pups that received Dex alone (p ! 0.05). Relative to controls, the decrease in the cardiac expression of eNOS was no longer signif-icant following combined treatment of Dex with vita-mins (DexCE: –20.3 8 14.4%, p 1 0.05). Treatment of control pups with vitamins alone did not significantly affect the cardiac expression of 4-HNE, Hsp90 or eNOS (p ! 0.05).

Stereology Heart. Images of representative mid-cardiac sections

are shown in figures 5 a–d. Compared to controls, total

heart volume was not significantly affected in any of the treatment groups (Ctrl: 246.2 8 17.3 mm 3 , Dex: 240.5 8 20.0 mm 3 , DexCE: 213.9 8 18.2 mm 3 , CtrlCE: 293.1 8 12.4 mm 3 , p 1 0.05). However, treatment with Dex sig-nificantly decreased the volume of the LV + IVS, and sig-nificantly increased the lumen volume of the LV and RV (all p ! 0.05; fig. 5 e). Administration of Dex in combina-tion with vitamins C and E did not alter the lumen volume of the LV. However, the volume of the LV + IVS and the RV lumen remained decreased and increased, respective-ly (both p ! 0.05; fig. 5 e). Treatment with vitamins alone had similar effects to Dex: the volume of the LV + IVS was decreased and the lumen volume of the LV was increased (p ! 0.05; fig. 5 e). Relative to controls, in all three treat-ment groups, the thickness of the LV and RV walls was reduced to a similar extent (all p ! 0.05; fig. 5 f, g).

Aortae. Compared to controls, treatment with Dex did not significantly alter the area of the aortic lumen (Ctrl: 0.87 8 0.08 mm 2 vs. Dex: 0.71 8 0.03 mm 2 , p 1 0.05), wall (Ctrl: 0.35 8 0.01 mm 2 vs . Dex: 0.37 8 0.03 mm 2 ,p 1 0.05), or whole vessel (Ctrl: 1.18 8 0.11 mm 2 vs. Dex: 1.08 8 0.05 mm 2 , p 1 0.05). However, when the areas of

Table 1. Weanling somatic and organ weights at postnatal day 21

Variable Ctrl Dex DexCE CtrlCE

Body weight, g 65.182.7 50.981.1* 51.181.7* 61.781.1*Heart

Whole abs. 0.26780.011 0.19880.003* 0.19580.008* 0.25780.002rel. 4.1180.09 3.9080.07 3.8280.05* 4.0780.04

LV+IVS abs. 0.19780.009 0.14680.003* 0.14480.006* 0.19280.003rel. 3.0280.04 2.8780.05 2.8180.05* 3.0480.02

RV abs. 0.05780.002 0.04580.001* 0.04680.002* 0.05780.001rel. 0.8880.02 0.8980.01 0.8980.02 0.9080.02

Lungs abs. 0.46980.016 0.42680.014* 0.41380.012* 0.48680.017rel. 7.2480.14 8.3880.24* 8.1080.15* 7.6980.19

Brain abs. 1.5180.03 1.4180.01* 1.4180.01* 1.4880.02rel. 23.4780.85 27.7080.71* 27.9081.00* 23.4880.30

Liver abs. 2.580.1 1.780.1* 1.780.1* 2.480.1rel. 38.1780.50 34.2980.87* 33.7580.51* 37.8580.78

Kidney abs. 0.70080.033 0.50480.019* 0.49880.021* 0.66680.009rel. 10.6380.14 9.9180.31* 9.7280.22* 10.5780.13

Brain:liver 0.61680.022 0.81180.029* 0.82980.031* 0.62580.021Brain:kidney 2.2080.09 2.8180.10* 2.8980.14* 2.2280.035

Organ weights are expressed as absolute values (g) and relative to body weight (mg � g–1) in the control (Ctrl, n = 11), dexamethasone (Dex, n = 7), Dex with vitamins C and E (DexCE, n = 12) and control with vitamins C and E (CtrlCE, n = 12) pups. Kidney weight is the sum of left and right kidney weights. Brain:liver and brain:kidney are ratios of absolute weights. abs. = Absolute weight, rel. = relative weight. * p < 0.05 vs. Ctrl (one-way ANOVA + Tukey test).

Antioxidants and Glucocorticoid Treatment

Neonatology 2010;98:73–83 79

the aortic lumen and wall were expressed relative to whole vessel area, Dex significantly decreased the ratio of the lumen:total vessel area (Ctrl: 0.74 8 0.01 vs. Dex: 0.66 8 0.02, p ! 0.05), but did not affect the ratio of the wall:total vessel area (Ctrl: 0.30 8 0.02 vs. Dex: 0.34 8 0.02, p 1 0.05). Addition of vitamins C and E to Dex-treated pups restored the ratio of lumen:total vessel area to control lev-els (DexCE: 0.69 8 0.02, p ! 0.05), but it did not affect the area of the lumen (DexCE: 0.95 8 0.09 mm 2 , p 1 0.05), wall (DexCE: 0.44 8 0.07 mm 2 , p 1 0.05), whole vessel (DexCE: 1.39 8 0.15 mm 2 , p 1 0.05) or the ratio of the wall:total vessel area (DexCE: 0.31 8 0.02, p 1 0.05). Administration of vitamins alone did not affect aortic area measurements (CtrlCE: lumen = 0.93 8 0.09 mm 2 ; wall = 0.37 8 0.03 mm 2 ; total = 1.31 8 0.12 mm 2 ; lumen:total vessel area ratio = 0.71 8 0.01; wall:total vessel area ratio = 0.29 8 0.01, all p 1 0.05).

Discussion

The data show that combined treatment of Dex in hu-man clinically relevant doses with vitamins C and E ame-liorates the adverse effects of Dex on survival, growth symmetry, cardiac protein expression and aortic mor-phology, but it does not improve decreased weight gain or altered cardiac structure. Surprisingly, vitamins C and E alone also affect somatic growth and cardiac structure.

In this study, Dex treatment reduced survival to 70%, a finding supported by two other studies in neonatal rat pups [42, 43] . Plausible explanations for this reduction in survival include a glucocorticoid-induced suppression in immunity and/or somatic growth, promoting failure to thrive in the immediate period following postnatal treat-ment. The newborn rat is protected from infection dur-ing the first 18 days following birth by maternal antibod-ies, which it absorbs through the small intestine [44] . Al-

0

0.5

1.0 eNOS protein

eNOS

Ctrl CtrlCEDex DexCE

*

Ratio

to �

-act

in

Ratio

to �

-act

in

Ratio

to �

-act

in

4-Hydroxynonenal protein

0

0.5

1.0

* *

Ctrl CtrlCEDex DexCE

4-HNE

Hsp90

0

0.4

0.8

1.2Hsp90 protein

**†

Ctrl CtrlCEDex DexCEDex

DexCECtrl

CtrlCE

Dex

DexCECtrl

CtrlCE

Dex

DexCECtrl

CtrlCE

�-Actin

�-Actin

�-Actin

a

b

c

Fig. 4. Expression of 4-HNE, Hsp90 and eNOS protein in the left ventricle. Repre-sentative Western blots are shown from the control (Ctrl, n = 11), dexamethasone (Dex, n = 7), Dex with vitamins C and E (DexCE, n = 12), and control with vita-mins C and E (CtrlCE, n = 12) groups. Ex-pression of 4-HNE ( a , middle band at 29 kDa), Hsp90 ( b , single band at 90 kDa), and eNOS ( c , single band at 140 kDa) was quantified and expressed as a ratio to � -actin. * p ! 0.05 vs. Ctrl; † p ! 0.05 vs. Dex (one-way ANOVA + Tukey test).

Adler/Camm/Hansell/Richter/Giussani Neonatology 2010;98:73–8380

though this transfer usually ceases between 18 and 21 days following birth, synthetic glucocorticoids promote premature closure of the gut, an effect that is complete within 6 days following administration. Given that most of the deaths in the Dex-treated group occurred between P3 and P6, 2–5 days after initiating treatment, accelerated gut closure and susceptibility to infection may explain the increase in mortality. Another plausible explanation for glucocorticoid-induced increases in neonatal mortal-ity is provided by the well-described effects of synthetic steroids on somatic growth, due to the direct effects of glucocorticoids on tissue accretion and catabolism [45] or reduced food intake due to poor suckling [15] . When ad-ministered in combination with Dex, vitamins restored survival to control levels, but did not prevent the suppres-sion of growth or alterations in FGR. This dissociation thus favours decreased immunity rather than growth suppression and failure to thrive as the underlying mech-anism for the deleterious effects on survival following

glucocorticoid treatment. Vitamin E has been shown to enhance immune responses in many animal models and in male weanling rats, a dose of vitamin E 1 50 mg � kg –1 is necessary for optimum immune function [46] . Thus the dose used in this study (100 mg � kg –1 ) may have pro-tected pups from infection during the treatment period.

Analysis of cardiac protein expression revealed that clinically relevant postnatal doses of Dex increased the stable marker of free radical-induced lipid peroxidation, 4-HNE [47] , and decreased levels of two proteins impor-tant to NO bioavailability, eNOS and Hsp90. In tissue derived from adults, Dex has been shown to decrease eNOS activity by limiting the availability of the essential substrates L -arginine [48] and tetrahydrobiopterin (BH 4 ; [30, 49] ), leading to uncoupling, superoxide production and reduced NO bioavailability [30, 49, 50] . A recent study demonstrated that 4-HNE inhibits both eNOS ac-tivity and Hsp90 expression and promotes further oxida-tive stress in vivo [51] . These changes were ascribed to

* * *0

20

40

60

80

100

**

* *

Ctrl Dex DexCE CtrlCE

Perc

ent o

f tot

alve

ntric

ular

vol

ume

Right ventricle lumen

Right ventricle wall

Left ventricle lumen

Left ventricle wall

00.5

1.5

2.5

Left

ven

tric

le w

all

thic

knes

s (m

m)

* * *

LLRL

RV

IVSDexCtrl

LLRL

RVIVS

LLRL

RVIVSDexCE

RV

LLRL

IVSCtrlCE

0

0.50

1.00

Righ

t ven

tric

le w

all

thic

knes

s (m

m)

* * *

0.25

0.75

LV

LV LV LV

Dex

DexCECtrl

CtrlCE

Dex

DexCECtrl

CtrlCE

a b c d

e

f g

Fig. 5. Measurement of cardiac volumes and wall thickness. Measurements were performed in the control (Ctrl, n = 5), dexamethasone (Dex, n = 5), Dex with vi-tamins C and E (DexCE, n = 5) and control with vitamins C and E (CtrlCE, n = 5) pups. a–d Coronal sections of hearts taken at the mid-cardiac level. e Wall and lumen volumes of the left and right cardiac ven-tricles expressed as a percentage of total ventricular volume. f , g Wall thickness of the left and right ventricles. * p ! 0.05 vs. control (one-way ANOVA + Tukey test). Scale bar = 2 mm. IVS = Interventricular septum, LL = left lumen, LV = left ventri-cle, RL = right lumen, RV = right ventri-cle.

Antioxidants and Glucocorticoid Treatment

Neonatology 2010;98:73–83 81

depletion of BH 4 and were prevented by the BH 4 donor, sepiapterin. Thus, by reducing the expression of eNOS and Hsp90 and by increasing the expression of 4-HNE, Dex may promote eNOS uncoupling and loss of NO bio-availability. The addition of the antioxidant vitamins to Dex-treated pups significantly increased Hsp90 expres-sion relative to Dex-treated pups. In addition, relative to control pups, the cardiac protein expression of eNOS was no longer significantly altered. The additions of vitamins, however, did not prevent the increase in 4-HNE expres-sion, suggesting that their beneficial effect is not due to prevention of excessive glucocorticoid-induced free radi-cal generation. Rather, antioxidant vitamins appear to counteract the effects of free radicals once generated. The partial restoration of antioxidant vitamins of cardiac eNOS protein may be largely due to the ability of vitamin C to increase intracellular BH 4 content [52] . By doing so, vitamin C not only reduces eNOS uncoupling and en-hances NO production, but also stabilizes this protein and thereby increases steady-state levels [49] .

In the present study, Dex treatment markedly altered cardiac structure, resulting in a reduction in ventricular wall volume. Despite a reduction in the aortic lumen:total vessel area ratio following Dex treatment, other aortic measurements were not significantly altered. Our results support the findings of a previous study by Bal et al. [22] , which reported an increase in end-diastolic volume and systolic dysfunction using the same animal model. The authors proposed that these changes promoted a state re-sembling dilated cardiomyopathy in later life [22] . Al-though increased cardiac afterload initially induces car-diac hypertrophy in order to maintain cardiac output [53] , failure of the heart to adapt to sustained increases in afterload could, over time, lead to cardiac decompensa-tion, dilatation, and heart failure. Therefore, at first sight, it appears that the effects of Dex on cardiac morphology in the present study may result from a sustained increase in cardiac afterload secondary to sustained Dex-induced peripheral vasoconstriction [54, 55] . Consequently, co-administration of glucocorticoids with antioxidant vita-mins, by maintaining the bioavailability of NO, may al-leviate peripheral vascular resistance, cardiac afterload and its consequences. However, while Dex in combina-tion with antioxidant vitamins partially restored cardiac NO bioavailability, growth asymmetry and an index of aortic thickening, supporting the statement above, this combined therapy did not improve cardiac morphology. Thus, in the present study, glucocorticoid-induced car-diac remodelling may have been independent of altera-tions in cardiac afterload. A likely explanation is that Dex

had direct effects on cardiac cellular development inde-pendent of an effect on NO bioavailability. For instance, in the rat, the proliferative capacity of cardiomyocytes declines shortly after birth, and glucocorticoids are known to promote a premature switch from proliferation to differentiation in these cells [20] . Dex administered during the first few days of life thus markedly reduces the number of cardiomyocytes and causes an increase in car-diomyocyte volume, length and width, indicating hyper-trophy [21] . Reduced cardiomyocyte number may explain not only the decreased cardiac weight measured in the present study, but also the wall thinning, decreased ven-tricle wall volume and increased lumen volume.

An unexpected finding of this study was that admin-istration of vitamins C and E alone reduced weight gain and FGR during the postnatal period, and had marked effects on cardiac structure, similar to the changes seen in Dex-treated pups. While an overwhelming number of studies have been conducted on antioxidant treatment in disease states, or conditions that may involve oxidative stress, comparatively few studies have investigated the ef-fects of antioxidant vitamin supplementation in healthy humans and animals, especially in infants and develop-ing children. Studies that have investigated vitamin sup-plementation in healthy humans and animals have coun-ter-intuitively shown adverse effects of antioxidant treat-ment on the physiology of the individual. For instance, a recent Cochrane analysis suggested serious caution re-garding the use of antioxidants as a primary or secondary method of disease prevention in human beings [56] . That study concluded that vitamin E significantly increased mortality in randomized trials of antioxidant supple-mentation, whereas vitamin C did not have a significant effect on longevity. In healthy pigs, doses of vitamins C and E, which had been shown to reduce oxidative stress under pathological conditions, were found to increase oxidative stress in the kidney and heart and to impair en-dothelial function [57, 58] . Therefore, combined, past and present evidence suggest that although antioxidant vita-min supplementation may improve disease states or con-ditions associated with enhanced free radical generation, antioxidant treatment in healthy conditions where the physiology of the individual is replenished with appropri-ate concentrations of antioxidants may, in fact, lead to excess NO bioavailability and peroxynitrite generation, thereby triggering unwanted side effects resembling oxi-dative stress [57, 58] .

In conclusion, treatment of newborn rats with Dex us-ing a human clinically relevant dosing regimen has det-rimental effects on survival, somatic growth and cardio-

Adler/Camm/Hansell/Richter/Giussani Neonatology 2010;98:73–8382

vascular structure during early postnatal development. Dex promotes oxidative stress in the developing heart by decreasing proteins essential in NO biosynthesis and by upregulating lipid peroxidation. Co-administration of Dex with antioxidant vitamins improves survival and partially restores the cardiac levels of proteins important to NO biosynthesis, but does not ameliorate the effects of Dex on somatic growth or cardiac structure. Administra-tion of vitamins alone to healthy pups decreases weight gain, induces thinning of the cardiac walls, and increases the lumen volume of the LV. The data suggest that com-bined glucocorticoid and antioxidant treatment may di-minish some of the adverse consequences of postnatal glucocorticoid therapy on the cardiovascular system. It is now important to assess the value of this combined ther-

apy on survival and development in a model that closely mimics premature birth in humans (i.e. one involving the presence of inflammation). Antioxidant vitamin supple-mentation alone is not recommended in healthy off-spring.

Acknowledgments

The work was supported by The British Heart Foundation. The authors would like to thank Prof. Abigail Fowden for assis-tance with the study, and Mrs. Anita Shelley for histological ad-vice. Alexandra Adler was supported by a Gates-Cambridge Scholarship. Dr. Dino A. Giussani is a Wolfson Research Merit Award Holder from the Royal Society.

References

1 Kinsella JP, Greenough A, Abman SH: Bron-chopulmonary dysplasia. Lancet 2006; 367: 1421–1431.

2 Northway WH Jr, Rosan RC, Porter DY: Pul-monary disease following respirator therapy of hyaline-membrane disease. Bronchopul-monary dysplasia. N Engl J Med 1967; 276: 357–368.

3 Fanaroff AA, Stoll BJ, Wright LL, Carlo WA, Ehrenkranz RA, Stark AR, Bauer CR, Dono-van EF, Korones SB, Laptook AR, Lemons JA, Oh W, Papile LA, Shankaran S, Steven-son DK, Tyson JE, Poole WK: Trends in neo-natal morbidity and mortality for very low birthweight infants. Am J Obstet Gynecol 2007; 196: 147.e1–147.e8.

4 Cummings JJ, D’Eugenio DB, Gross SJ: A controlled trial of dexamethasone in pre-term infants at high risk for bronchopulmo-nary dysplasia. N Engl J Med 1989; 320: 1505–1510.

5 Halliday HL, Ehrenkranz RA, Doyle LW: Early ( ! 8 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev 2009:CD001146.

6 Halliday HL, Ehrenkranz RA, Doyle LW: Late ( 1 7 days) postnatal corticosteroids for chronic lung disease in preterm infants. Co-chrane Database Syst Rev 2009:CD001145.

7 Ballard PL, Ballard RA: Scientific basis and therapeutic regimens for use of antenatal glucocorticoids. Am J Obstet Gynecol 1995; 173: 254–262.

8 Liggins GC: Fetal lung maturation. Aust NZ J Obstet Gynaecol 1994; 34: 247–250.

9 Stark AR, Carlo WA, Tyson JE, Papile LA, Wright LL, Shankaran S, Donovan EF, Oh W, Bauer CR, Saha S, Poole WK, Stoll BJ: Ad-verse effects of early dexamethasone in ex-tremely-low-birth-weight infants. National Institute of Child Health and Human Devel-opment Neonatal Research Network. N Engl J Med 2001; 344: 95–101.

10 Thorp JA, Jones PG, Peabody JL, Knox E, Clark RH: Effect of antenatal and postnatal corticosteroid therapy on weight gain and head circumference growth in the nursery. Obstet Gynecol 2002; 99: 109–115.

11 Yeh TF, Lin YJ, Huang CC, Chen YJ, Lin CH, Lin HC, Hsieh WS, Lien YJ: Early dexameth-asone therapy in preterm infants: a follow-up study. Pediatrics 1998; 101:E7.

12 Yeh TF, Lin YJ, Lin HC, Huang CC, Hsieh WS, Lin CH, Tsai CH: Outcomes at school age after postnatal dexamethasone therapy for lung disease of prematurity. N Engl J Med 2004; 350: 1304–1313.

13 Flagel SB, Vazquez DM, Watson SJ Jr, Neal CR Jr: Effects of tapering neonatal dexa-methasone on rat growth, neurodevelop-ment, and stress response. Am J Physiol 2002; 282:R55–R63.

14 He J, Varma A, Weissfeld LA, Devaskar SU: Postnatal glucocorticoid exposure alters the adult phenotype. Am J Physiol 2004; 287:R198–208.

15 Neal CR Jr, Weidemann G, Kabbaj M, Vazquez DM: Effect of neonatal dexametha-sone exposure on growth and neurological development in the adult rat. Am J Physiol l 2004; 287:R375–R385.

16 Bensky AS, Kothadia JM, Covitz W: Cardiac effects of dexamethasone in very low birth weight infants. Pediatrics 1996; 97: 818–821.

17 Ohning BL, Fyfe DA, Riedel PA: Reversible obstructive hypertrophic cardiomyopathy after dexamethasone therapy for broncho-pulmonary dysplasia. Am Heart J 1993; 125: 253–256.

18 Werner JC, Sicard RE, Hansen TW, Solomon E, Cowett RM, Oh W: Hypertrophic cardio-myopathy associated with dexamethasone therapy for bronchopulmonary dysplasia. J Pediatr 1992; 120: 286–291.

19 De Vries WB, Bal MP, Homoet-van der Kraak P, Kamphuis PJ, van der Leij FR, Baan J, Steendijk P, de Weger RA, van Bel F, van Oosterhout MF: Suppression of physiologi-cal cardiomyocyte proliferation in the rat pup after neonatal glucocorticosteroid treat-ment. Basic Res Cardiol 2006; 101: 36–42.

20 Li F, Wang X, Capasso JM, Gerdes AM: Rap-id transition of cardiac myocytes from hy-perplasia to hypertrophy during postnatal development. J Mol Cell Cardiol 1996; 28: 1737–1746.

21 De Vries WB, van der Leij FR, Bakker JM, Kamphuis PJ, van Oosterhout MF, Schipper ME, Smid GB, Bartelds B, van Bel F: Altera-tions in adult rat heart after neonatal dexa-methasone therapy. Pediatr Res 2002; 52: 900–906.

22 Bal MP, de Vries WB, van der Leij FR, van Oosterhout MF, Berger RM, Baan J, van der Wall EE, van Bel F, Steendijk P: Neonatal glucocorticosteroid treatment causes systol-ic dysfunction and compensatory dilation in early life: studies in 4-week-old prepubertal rats. Pediatr Res 2005; 58: 46–52.

Antioxidants and Glucocorticoid Treatment

Neonatology 2010;98:73–83 83

23 Bal MP, de Vries WB, van Oosterhout MF, Baan J, van der Wall EE, van Bel F, Steendijk P: Long-term cardiovascular effects of neo-natal dexamethasone treatment: hemody-namic follow-up by left ventricular pressure-volume loops in rats. J Appl Physiol 2008; 104: 446–450.

24 American Academy of Pediatrics: Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Pediatrics 2002; 109: 330–338.

25 Grier DG, Halliday HL: Management of bronchopulmonary dysplasia in infants: guidelines for corticosteroid use. Drugs 2005; 65: 15–29.

26 Baden M, Bauer CR, Colle E, Klein G, Taeusch HW, Stern L: A controlled trial of hydrocortisone therapy in infants with re-spiratory distress syndrome. Pediatrics 1972; 50: 526–534.

27 Rademaker KJ, de Vries LS, Uiterwaal CS, Groenendaal F, Grobbee DE, van Bel F: Post-natal hydrocortisone treatment for chronic lung disease in the preterm newborn and long-term neurodevelopmental follow-up. Arch Dis Child Fetal Neonatal 2008; 93:F58–F63.

28 Iuchi T, Akaike M, Mitsui T, Ohshima Y, Shintani Y, Azuma H, Matsumoto T: Gluco-corticoid excess induces superoxide produc-tion in vascular endothelial cells and elicits vascular endothelial dysfunction. Circ Res 2003; 92: 81–87.

29 Rajashree S, Puvanakrishnan R: Dexameth-asone induced alterations in enzymatic and nonenzymatic antioxidant status in heart and kidney of rats. Mol Cell Biochem 1998; 181: 77–85.

30 Whitworth JA, Schyvens CG, Zhang Y, An-drews MC, Mangos GJ, Kelly JJ: The nitric oxide system in glucocorticoid-induced hy-pertension. J Hypertens 2002; 20: 1035–1043.

31 Zhang Y, Croft KD, Mori TA, Schyvens CG, McKenzie KU, Whitworth JA: The antioxi-dant tempol prevents and partially reverses dexamethasone-induced hypertension in the rat. Am J Hypertens 2004; 17: 260–265.

32 Cai H, Harrison DG: Endothelial dysfunc-tion in cardiovascular diseases: the role of oxidant stress. Circ Res 2000; 87: 840–844.

33 Hamilton CA, Miller WH, Al-Benna S, Bros-nan MJ, Drummond RD, McBride MW, Dominiczak AF: Strategies to reduce oxida-tive stress in cardiovascular disease. Clin Sci (Lond) 2004; 106: 219–234.

34 Li JM, Shah AM: Endothelial cell superoxide generation: regulation and relevance for car-diovascular pathophysiology. Am J Physiol 2004; 287:R1014–R1030.

35 Oncu M, Gultekin F, Karaöz E, Altuntas I, Delibas N: Nephrotoxicity in rats induced by chlorpryfos-ethyl and ameliorating effects of antioxidants. Hum Exp Toxicol 2007; 21: 223–230.

36 Burton GW: Vitamin E: molecular and bio-logical function. Proc Nutr Soc 1994; 53: 251–262.

37 Carr AC, Zhu BZ, Frei B: Potential antiath-erogenic mechanisms of ascorbate (vitamin C) and � -tocopherol (vitamin E). Circ Res 2000; 87: 349–354.

38 Jackson TS, Xu A, Vita JA, Keaney JF Jr: Ascorbate prevents the interaction of super-oxide and nitric oxide only at very high phys-iological concentrations. Circ Res 1998; 83: 916–922.

39 Burton GJ, Palmer ME: Eradicating fetoma-ternal f luid shift during perfusion fixation of the human placenta. Placenta 1988; 9: 327–332.

40 Gundersen HJG, Bendtsen TF, Korbo L, Marcussen N, Moller A, Nielsen K, Nyen-gaard JR, Pakkenberg B, Sorensen FB, Vesterby A, West MJ: Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. AP-MIS 1988; 96: 379–394.

41 Jensen EB, Gundersen HJG, Osterby R: De-termination of membrane thickness distri-bution from orthogonal intercepts. J Microsc 1979; 115: 19–33.

42 Sicard RE, Werner JC: Dexamethasone in-duces a transient relative cardiomegaly in neonatal rats. Pediatr Res 1992; 31: 359–363.

43 Thibeault DW, Heimes B, Rezaiekhaligh M, Mabry S: Chronic modifications of lung and heart development in glucocorticoid-treated newborn rats exposed to hyperoxia or room air. Pediatr Pulmonol 1993; 16: 81–88.

44 Daniels VG, Hardy RN, Malinowska KW, Nathanielsz PW: The influence of exogenous steroids on macromolecule uptake by the small intestine of the newborn rat. J Physiol 1973; 229: 681–695.

45 Weiler HA, Wang Z, Atkinson SA: Whole body lean mass is altered by dexamethasone treatment through reductions in protein and energy utilization in piglets. Biol Neonate 1997; 71: 53–59.

46 Bendich A, Gabriel E, Machlin LJ: Dietary vitamin E requirement for optimum im-mune responses in the rat. J Nutr 1986; 116: 675–681.

47 Poli G, Biasi F, Leonarduzzi G: 4-Hy-droxynonenal-protein adducts: a reliable biomarker of lipid oxidation in liver diseases. Mol Aspects Med 2008; 29: 67–71.

48 Schafer SC, Wallerath T, Closs EI, Schmidt C, Schwarz PM, Forstermann U, Lehr HA: Dexamethasone suppresses eNOS and CAT-1 and induces oxidative stress in mouse re-sistance arterioles. Am J Physiol 2005; 288:H436–H444.

49 Cai S, Alp NJ, McDonald D, Smith I, Kay J, Canevari L, Heales S, Channon KM: GTP cy-clohydrolase I gene transfer augments intra-cellular tetrahydrobiopterin in human en-dothelial cells: effects on nitric oxide synthase activity, protein levels and dimeri-sation. Cardiovasc Res 2002; 55: 838–849.

50 Wallerath T, Witte K, Schafer SC, Schwarz PM, Prellwitz W, Wohlfart P, Kleinert H, Lehr HA, Lemmer B, Forstermann U: Down-regulation of the expression of endothelial NO synthase is likely to contribute to gluco-corticoid-mediated hypertension. Proc Natl Acad Sci USA 1999; 96: 13357–13362.

51 Whitsett J, Picklo MJ Sr, Vasquez-Vivar J: 4-Hydroxy-2-nonenal increases superoxide anion radical in endothelial cells via stimu-lated GTP cyclohydrolase proteasomal deg-radation. Arterioscler Thromb Vasc Biol 2007; 27: 2340–2347.

52 Huang A, Vita JA, Venema RC, Keaney JF Jr: Ascorbic acid enhances endothelial nitric-oxide synthase activity by increasing intra-cellular tetrahydrobiopterin. J Biol Chem 2000; 275: 17399–17406.

53 Munzel T, Harrison DG: Increased superox-ide in heart failure: a biochemical baroreflex gone awry. Circulation 1999; 100: 216–218.

54 Derks JB, Giussani DA, Jenkins SL, Went-worth RA, Visser GH, Padbury JF, Natha-nielsz PW: A comparative study of cardio-vascular, endocrine and behavioural effects of betamethasone and dexamethasone ad-ministration to fetal sheep. J Physiol 1997; 499: 217–226.

55 Fletcher AJ, McGarrigle HH, Edwards CM, Fowden AL, Giussani DA: Effects of low dose dexamethasone treatment on basal cardio-vascular and endocrine function in fetal sheep during late gestation. J Physiol 2002; 545: 649–660.

56 Bjelakovic G, Nikolova D, Gluud L, Simonet-ti R, Gluud C: Antioxidant supplements for prevention of mortality in healthy partici-pants and patients with various diseases. Co-chrane Database Syst Rev 2008:CD007176.

57 Daghini E, Chade AR, Krier JD, Versari D, Lerman A, Lerman LO: Acute inhibition of the endogenous xanthine oxidase improves renal hemodynamics in hypercholesterol-emic pigs. Am J Physiol 2006; 290:R609–R615.

58 Versari D, Daghini E, Rodriguez-Porcel M, Sattler K, Galili O, Pilarczyk K, Napoli C, Lerman LO, Lerman A: Chronic antioxidant supplementation impairs coronary endothe-lial function and myocardial perfusion in normal pigs. Hypertension 2006; 47: 475–481.