Maternal obesity during pregnancy and lactation programs the ...

Research Article

Maternal obesity during pregnancy and lactation programs the development of offspring non-alcoholic fatty liver disease in miceq Jude A. Oben1,2,*, Angelina Mouralidarane1,3, Anne-Maj Samuelsson3, Phillippa J. Matthews3, Maelle L. Morgan1, Chad Mckee1, Junpei Soeda1, Denise S. Fernandez-Twinn4, Malgorzata S. Martin-Gronert4, Susan E. Ozanne4, Barbara Sigala1, Marco Novelli5, Lucilla Poston3, Paul D. Taylor3 1

University College London, Centre for Hepatology, Royal Free Hospital, London, UK; 2Guy’s and St. Thomas’ Hospital, Department of Gastroenterology, London, UK; 3King’s College London, Division of Reproduction & Endocrinology, London, UK; 4Metabolic Research Laboratories, University of Cambridge, Institute of Metabolic Science, Cambridge, UK; 5University College London, Department of Pathology, London, UK See Editorial, pages 786–787

Background & Aims: Obesity induced, non-alcoholic fatty liver disease (NAFLD), is now the major cause in affluent countries, of the spectrum of steatosis-to-cirrhosis. Obesity and NAFLD rates in reproductive age women, and adolescents, are rising worldwide. Our hypothesis was that maternal obesity and lactation transmit to the offspring a pre-disposition to dysmetabolism, obesity and NAFLD. Methods: Female mice were fed standard or obesogenic chow, before, throughout pregnancy, and during lactation. The critical developmental period was studied by cross-fostering offspring of lean and obese dams. Offspring were then weaned onto standard chow and studied at 3 months. Read-outs included markers of metabolic dysfunction, biochemical and histological indicators of NAFLD, induction of liver fibrogenesis, and activation of profibrotic pathways. Mechanisms involved in programming a dysmetabolic and NAFLD phenotype were investigated by assaying breast milk components. Results: Offspring of obese dams had a dysmetabolic, insulin resistant and NAFLD phenotype compared to offspring of lean dams. Offspring of lean dams that were suckled by obese dams showed an exaggerated dysmetabolic and NAFLD phenotype, with increased body weight, as well as increased levels of insulin, leptin, aspartate transaminase, interleukin-6, tumour necrosis factor-a, liver triglycerides, steatosis, hepatic fibrogenesis, renal

Keywords: Obesity; NAFLD; Developmental programming; Sympathetic nervous system; Liver fibrosis. Received 8 September 2009; received in revised form 25 November 2009; accepted 14 December 2009; available online 1 April 2010 q Financial disclosures: There are no conflicts of interest to disclose for all authors. * Corresponding author at: University College London, Centre for Hepatology, Royal Free Hospital, Rowland Hill, London NW3 2PF, UK. Tel.: +44 207 433 2875; fax: +44 207 433 2871. E-mail address: [email protected] (J.A. Oben). Abbreviations: NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; TNF-a, tumour necrosis factor-a; IL-6, interleukin-6; SNS, sympathetic nervous system; NE, norepinephrine; HSC, hepatic stellate cells; AST, aspartate transaminase; ASMA, a-smooth muscle actin (a-SMA); Col 1-a2, collagen 1-a2; H&E, haematoxylin and eosin; PI3K, phosphatidyl inositol 3-kinase; STAT3, signal transducer and activator of transcription 3; ARC, arcuate nucleus; AGRP, agouti-related peptide.

norepinephrine, and liver a1-D plus b1-adrenoceptors, indicative of sympathetic nervous system activation. Obese dams also had raised breast milk leptin levels compared to lean dams. Conclusions: Maternal obesity programs development of a dysmetabolic and NAFLD phenotype, which is critically dependent on the early postnatal period and possibly involving alteration of hypothalamic appetite nuclei signalling by maternal breast milk and neonatal adipose tissue derived, leptin. Ó 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.

Introduction Rising rates of obesity and obesity-induced liver disease (nonalcoholic fatty liver disease, NAFLD) may be partially explained by increasing availability of cheap energy dense foods, perhaps compounded by a developmental programming effect of maternal obesity. The spectrum of NAFLD ranges from hepatic steatosis to cirrhosis [1]. NAFLD is now the commonest cause of chronic liver disease in affluent countries, with 23–34% of the United States population estimated to have NAFLD, and about 2.5% the more severe form of the disease, non-alcoholic steatohepatitis (NASH) [2]. Obesity amongst women of reproductive age is similarly rising with some 29% of women, aged 20–39 years, in the USA reported to be obese [3]. Earlier studies elicited relationships between in utero undernutrition, low birthweight and adult hypertension, type 2 diabetes and metabolic dysfunction [4–6], leading to the hypothesis that nutritional insults during early development may induce permanent alterations in plastic regulatory systems to induce adult disease [7]. The developmental over-nutrition hypothesis, conversely, suggests that maternal obesity pre-disposes offspring to obesity and metabolic dysfunction in adulthood [8–10], with implications for the induction of NAFLD. The initiating defect in the pathogenesis of NAFLD is thought to be insulin resistance [11] and factors implicated in propagating NAFLD, include the adipokines tumour necrosis factor-a (TNF-a)

Journal of Hepatology 2010 vol. 52 j 913–920

Research Article and interleukin-6 (IL-6) [1]. In addition, the sympathetic nervous system (SNS) through the actions of its neurotransmitters, predominantly norepinephrine (NE), on the liver’s primary fibrogenic cells – hepatic stellate cells (HSC) – is involved in fibrosis progression, such that experimental animals lacking a SNS are poorly fibrogenic, and adrenoceptor antagonists inhibit fibrosis progression [12,13]. In this study, our aims were: to test the hypothesis that maternal obesity in pregnancy and lactation transmits a pre-disposition to offspring obesity, dysmetabolism and NAFLD, to identify the critical period of development inducing these phenotypes, and to determine involved mechanisms. We show clearly that offspring of obese dams had a dysmetabolic and NAFLD phenotype compared to offspring of lean dams, a phenotype further accentuated in offspring of lean suckled by obese dams, and characterised by increased body weight, insulin, leptin, aspartate transaminase (AST), IL-6, TNF-a, liver triglycerides, steatosis, hepatic fibrogenesis, and sympathetic nervous system activation. The mechanism may involve alteration of hypothalamic appetite nuclei signalling by maternal breast milk and neonatal adipose tissue derived leptin. Therefore, developmental programming may be partly responsible for the rising rates of obesity and NAFLD. Our experimental paradigm moreover, potentially offers a new pathophysiologically relevant model of NAFLD.

recorded weekly. Dams were allowed to deliver spontaneously and left undisturbed with their litters for 48 h. All offspring (n = 20) were weaned onto standard chow at 3 weeks postpartum. A subgroup of offspring (n = 5) born to a lean dam were cross-fostered to be suckled by an obese dam. Conversely, a subgroup of offspring (n = 5) born to an obese dam were cross-fostered and suckled by a lean dam. Offspring food intake and body weight were measured weekly; biochemical analysis, markers of liver injury and fibrogenesis and liver histology were determined at 3 months. Radiotelemetry Remote radiotelemetric probes (TA11PA-C10, OD 0.4 mm, Data Science International) were implanted under general anaesthetic at 3 months. The indwelling catheter was then inserted into the left carotid artery with its tip positioned in the aortic arch. Following recovery, systolic blood pressure (SBP) was measured for 7 days in conscious, freely moving offspring. To measure cardiovascular reactivity to stress, telemetered animals were placed in a humane perspex restrainer for 30 min inside the home cage. Cardiovascular parameters were recorded for a further 120 min throughout the recovery period and expressed as a change from baseline. Plasma analysis Blood was collected via cardiac puncture and plasma assayed by ELISA for leptin (Biovender RD-1301), insulin (DRG Instruments GmbH), and AST (local clinical biochemistry department). Liver tissue triglyceride

Methods

Whole liver tissue triglyceride was determined by an adaptation of the Folch Method [14] and an enzymatic colorimetric assay (UNIMATE 5 TRIG, Roche BC1, Sussex, UK).

Experimental animals

Gene expression of liver injury and fibrogenic markers

Female C57BL/6J mice (n = 8 per group) (Charles River Laboratories, UK), of first order parity, approximately 100 days old were allowed 7 days to habituate and fed standard chow RM1 (7% simple sugars, 3% fat, 50% polysaccharide, 15% protein [w/w] RM1, Special Dietary Services, energy 3.5 kcal/g, n = 20) ad libitum. Subsequently, half the group was switched to a semi-synthetic energy-rich, highly palatable obesogenic diet (10% simple sugars, 20% animal lard, 28% polysaccharide, 23% protein [w/w], Special Dietary Services, energy 4.5 kcal/g, n = 30) supplemented ad libitum with sweetened condensed milk (approximately 55% simple sugar, 8% fat, 8% protein, w/w, Nestle) admixed with mineral mix (AIN93G, Special Dietary Services, 125 mg/pot). Macronutrient and calorific intake were calculated from measured daily intake of pellets and milk (approximately 16% fat, 33% simple sugars, 15% protein, energy 4.0 kcal/g). Mice entered the breeding protocol on achieving a 50% increase in body weight. All animals were treated in accordance with The Animals (Scientific Procedures) Act, UK, 1986 guidelines. After 6 weeks on their respective diets, mice were mated with breeding males selected from the same litter, to minimise genetic variability. Day 0 of pregnancy was determined by formation of a vaginal plug. Dams failing to become pregnant after development of a copulation plug were allowed to re-mate. Pregnant dams were maintained on their respective diets throughout pregnancy and lactation. During pregnancy, maternal weight and dietary intake were

Semi-quantitative reverse transcriptase polymerase chain reaction (RT-PCR) was performed using SuperScript™ III One-Step RT-PCR System with PlatinumÒ Taq High Fidelity (Invitrogen Life Technologies). Gene specific primers were designed for IL-6, TNF-a, adrenoceptor a1-D and b1, a-smooth muscle actin (a-SMA) and collagen type 1-a2 (Col 1-a2) (Table 1). Expression of target genes was normalised to GAPDH expression. Renal catecholamines Harvested kidneys were snap-frozen and stored at 80 °C. The left kidney from each animal was homogenised on ice with 0.1 M hydrochloric acid (Sigma– Aldrich, UK) containing 1 mM EDTA (BDH Chemicals Ltd., Poole, England) and 4 mM sodium metabisulfite. Renal NE content was assessed by ELISA (ALPCO Immunoassays, Salem, NH, USA) as per the manufacturer’s instructions. Western blotting – expression of insulin signalling proteins, ASMA, and collagen type 1 Total liver protein (n = 6 per group) was extracted in ice-cold lysis buffer and lysates protein concentrations determined using the copper/bicinchoninic assay

Table 1. Primer sequences, expected weights, and annealing temperatures. Gene

Primer sequence

IL-6

Sense 50 -TTCACAGAGGATACCACTCC-30 Antisense 50 -GTTTGGTAGCATCCATCATT-30 Sense 50 -TCCAGCTGACTAAACATCCT-30 Antisense 50 -CCCTTCATCTTCCTCCTTAT-30 Sense 50 -GAACGGTCCACGATTGCATG-30 Antisense 50 -GGCATGTTGCTAGGCACGAAG-30 Sense 50 -ATCTGGCACCACTCTTTCTA-30 Antisense 50 -GTACGTCCAGAGGCATAGAG-30 Sense 50 -TTGAATTCCTACAGAGACCCACGACCCAG-30 Antisense 50 -CGGAATTCTTAAATGTCAGTCTCCCGGAG-30 Sense 50 -ACGCTCACCAACCTCTTCAT-30 Antisense 50 -AGGGGCACGTAGAAGGGAGAC-30

TNF-a Col 1-a2 ASMA

a1-D b1

914

Expected weight (bp)

Annealing (°C)

203

55

220

55

167

55

191

59

229

55

440

55

Journal of Hepatology 2010 vol. 52 j 913–920

JOURNAL OF HEPATOLOGY (Sigma) and standardised to 2 mg/ml by dilution into Laemmli buffer and 20 lg total protein loaded for SDS–PAGE. Separated proteins were transferred onto PVDF Immobilon-P membrane (Millipore, Billerica, MA) and incubated in blocking buffer for 1 h (5% non-fat dehydrated milk, 1 TBS, 0.1% Tween 20), followed by overnight incubation with antibody against IRS-1, phospho-IRS1 (Ser 307), phospho-IRS1 (Tyr612) and PI3K p85a (Upstate Biotechnology, Lake Placid, USA); phospho-Akt (Ser 473) (Cell Signalling Technology, Beverly, USA); IR-b, PKCf, PI3K, p110b (Santa Cruz Biotechnology, Santa Cruz, USA); ASMA (Sigma) and collagen type 1 (Millipore) diluted in TBS–0.1% Tween 20 containing 5% dried milk or 5% BSA. Protein expression was assessed densitometrically using AlphaEase software (AlphaImager).

Liver histology Offspring liver sections at 3 months postpartum were formalin fixed and paraffin embedded prior to sectioning. All sections were then stained with haematoxylin and eosin (H&E) and assessed for steatosis and inflammation, by an expert liver pathologist blinded to the identity of the groups, using the Brunt–Kleiner NAFLD Activity Score as previously described in detail [15].

Results Offspring exposure to maternal obesity throughout pregnancy and lactation induced histological and biochemical evidence of hepatic steatosis plus biochemical evidence of liver injury and fibrogenesis with activation of the SNS Liver tissue triglyceride content (Fig. 1A) was elevated in offspring of obese compared to offspring of lean dams although the difference was not significant. However, mean steatosis score was significantly greater in offspring of obese compared to offspring of lean dams (Fig. 1B and C). In parallel, plasma aspartate transaminase (AST), as a marker of liver injury, was also significantly elevated in offspring of obese dams (Fig. 1D) along with collagen 1-a2 gene expression (Fig. 1E), indicating induced liver injury with fibrogenesis. Moreover, the blood pressure response to restraint stress was markedly elevated in offspring of obese dams, indicating increased SNS tone (Fig. 1F).

Breast milk leptin Breast milk samples were obtained from suckling dams after anaesthesia and analysed as described [16].

Statistical analysis Multiple comparisons on a single data set were performed using ANOVA and expressed as mean ± SEM unless otherwise stated. p < 0.05 was regarded as significant. Sample size per group; n = 5.

10

B

ns

C *

8 6 4

2

1

2 0

Offspring of lean dam

0 Offspring of obese dam


400 300 200 100 0


Offspring of obese dam


F 400

**

SBP, % from baseline

500



E **

Densitometry of collagen type 1 mRNA expression (arbitrary units)

D Plasma AST (IU/L) % control

Expression of IR-b in the livers of offspring exposed to maternal obesity throughout gestation and lactation was not altered compared to offspring of lean dams (Fig. 2). However, expression of IRS-1 protein was significantly decreased in the offspring of obese dams, while phosphorylation of IRS-1 at Ser 307, an inhibitory

3 Steatosis score

Liver tissue triglyceride (mmol/L)

A

Offspring exposure to maternal obesity throughout pregnancy and lactation induced altered expression/phosphorylation of insulin signalling proteins, indicative of insulin resistance

300 200

100 0

160 150

130

***

120 110 100 90


120 min recovery (in home cage)

***

140

30 min restraint stress


Fig. 1. Liver injury markers, liver histology, and stress response at 3 months postpartum. (A) Liver tissue triglyceride, (B) mean steatosis score, (C) H&E stain of representative liver section, (D) plasma AST, (E) relative expression of collagen mRNA, (F) blood pressure change from baseline after restraint (s, offspring born to and suckled by a lean dam; d, offspring born to and suckled by an obese dam). A similar pattern, as with systolic blood pressure was observed with heart rate variability after restraint (data not shown); *p