Maternal obesity impairs brain glucose metabolism and neural response to hyperglycemia in male rat offspring

Download Maternal obesity impairs brain glucose metabolism and neural response to hyperglycemia in male rat offspring

Post on 24-Mar-2017

212 views

Category:

Documents

0 download

TRANSCRIPT

<ul><li><p>,</p><p>*School of Medical and Molecular Biosciences, Faculty of Science, Centre for Health Technology,</p><p>University of Technology, Sydney, NSW, Australia</p><p>Department of Pharmacology, School of Medical Sciences, University of New South Wales, Sydney,</p><p>NSW, Australia</p><p>Inflammation and Infection Research, School of Medical Sciences, University of New South Wales,</p><p>Sydney, NSW, Australia</p><p>AbstractHypothalamic appetite regulators neuropeptide Y (NPY) andpro-opiomelanocortin (POMC) are modulated by glucose.This study investigated how maternal obesity disturbs glu-cose regulation of NPY and POMC, and whether thisderegulation is linked to abnormal hypothalamic glucoseuptake-lactate conversion. As post-natal high-fat diet (HFD)can exaggerate the effects of maternal obesity, its additionalimpact was also investigated. Female Sprague Dawley ratswere fed a HFD (20 kJ/g) to model maternal obesity. Atweaning, male pups were fed chow or HFD. At 9 weeks, invivo hypothalamic NPY and POMC mRNA responses toacute hyperglycemia were measured; while hypothalami wereglucose challenged in vitro to assess glucose uptake-lactaterelease and related gene expression. Maternal obesitydampened in vivo hypothalamic NPY response to acute</p><p>hyperglycemia, and lowered in vitro hypothalamic glucoseuptake and lactate release. When challenged with 20 mMglucose, hypothalamic glucose transporter 1, monocarboxy-late transporters, lactate dehydrogenase-b, NPY and POMCmRNA expression were down-regulated in offspring exposedto maternal obesity. Post-natal HFD consumption reducedin vitro lactate release and monocarboxylate transporter 2mRNA, but increased POMC mRNA levels when challengedwith 20 mM glucose. Overall, maternal obesity producedstronger effects than post-natal HFD consumption to impairhypothalamic glucose metabolism. However, they both dis-turbed NPY response to hyperglycemia, potentially leading tohyperphagia.Keywords: GLUT1, lactate production, MCTs, mTOR, NPY,POMC.J. Neurochem. (2014) 129, 297303.</p><p>Traditionally, glucose-responsive neurons are defined as thosethat increase their action potential frequency when ambientglucose is increased above resting levels (up to 1020 mmol/L), whereas glucose-sensitive neurons decrease their actionpotential frequency under the same conditions (Song et al.2001). In this regard, in the hypothalamic arcuate nucleus(Arc), neurons expressing orexigenic neuropeptide Y (NPY)are inhibited by rising glucose levels, while neurons express-ing anorexigenic pro-opiomelanocortin (POMC) are excitedby glucose abundance (Muroya et al. 1999; Ibrahim et al.2003; Stefater and Seeley 2010). Mammalian target ofrapamycin (mTOR) can sense changes in ambient glucoseand amino acid levels; a function that is critical for regulationof cellular activity. mTOR is colocalized in NPY- and POMC-expressing cells (Cota et al. 2006), suggesting mTOR may</p><p>play an important role in NPY and POMC neural glucosesensing and changes in their mRNA expression. Indeed, theshort-term activation of mTOR can inhibit NPY expression to</p><p>Received April 3, 2013; revised manuscript received November 6, 2013;accepted November 21, 2013.Address correspondence and reprint requests to Hui Chen, School of</p><p>Medical and Molecular Biosciences, Faculty of Science, University ofTechnology, Sydney, NSW 2007, Australia. E-mail: hui.chen-1@uts.edu.au; Margaret J. Morris, Department of Pharmacology, School ofMedical Sciences, University of New South Wales, Sydney, NSW 2052,Australia. E-mail: m.morris@unsw.edu.auAbbreviations used: Arc, arcuate nucleus; GLUT, glucose transporter;</p><p>HFD, high-fat diet; IPGTT, intraperitoneal glucose tolerance test; LDH,lactate dehydrogenase; MCT, monocarboxylate transporter; mTOR,mammalian target of rapamycin; NPY, neuropeptide Y; POMC, pro-opiomelanocortin; PVN, paraventricular nucleus; TG, triglyceride.</p><p> 2013 International Society for Neurochemistry, J. Neurochem. (2014) 129, 297--303 297</p><p>JOURNAL OF NEUROCHEMISTRY | 2014 | 129 | 297303 doi: 10.1111/jnc.12623</p></li><li><p>reduce energy intake and body weight (Cota et al. 2006).High-fat diet (HFD) consumption can impair brain mTORactivation, leading to overeating in rodents (Um et al. 2006;Cota et al. 2006).We previously showed that maternal obesityis also linked to deregulation of both hypothalamic mTOR andNPY expression, contributing to hyperphagia and obesity inoffspring (Chen and Morris 2009; Chen et al. 2012).It has been suggested that brain glucose sensing activity by</p><p>mTOR relies on the ATP generated from glucoselactatepyruvate conversion (Stefater and Seeley 2010). In astro-cytes, glucose is taken up and can be further metabolizedanaerobically into lactate, which is transferred to neurons tosupply fuel (Magistretti 2006), in addition to the directglucose uptake by neurons from the extracellular space. Inneurons, lactate is further metabolized into pyruvate throughlactate dehydrogenase (LDH)b (Lam et al. 2007). Thisprocess can not only induce appetite suppression, but canalso lower blood lipid and glucose levels through amechanism involving the liver (Lam et al. 2008; Lam2007; Lam et al. 2005). Therefore, altered central glucosemetabolism may potentially lead to abnormal glucoseregulation of NPY and POMC-expressing neurons, and thusappetite deregulation. We have previously shown thathypothalamic mTOR expression and activity are significantlydown-regulated by maternal obesity (Chen et al. 2008,2012). It is not known whether this is linked to abnormalhypothalamic glucose sensing and metabolic ability. Thisformed one of the rationales of this study.We have observed that a change in blood glucose level</p><p>(hypoglycemia) led to hyper-activation of hypothalamicNPY expression in offspring from obese dams compared tooffspring from lean mothers (Chen and Morris 2009). Studieshave also reported impaired glucose sensing under lowglucose conditions in both genetic and dietary obese ratmodels, as well as in rat models of maternal overnutrition(Fuente-Martin et al. 2012; Song et al. 2001; Parton et al.2007; Canabal et al. 2007). However, it is unknown whethermaternal obesity leads to impaired hyperglycemia-inducedregulation of NPY and POMC (Sarbassov et al. 2005). Wehypothesized that hypothalamic glucose metabolism wasimpaired by maternal obesity, which further leads toderegulated hypothalamic NPY and POMC expression inresponse to hyperglycemia. In addition, these alterations canbe further exacerbated by post-natal HFD consumption.Therefore, we aimed to determine the effect of maternalobesity and post-natal HFD on the mRNA expression ofhypothalamic NPY and POMC during hyperglycemia in vivoand under increased ambient glucose level in vitro.</p><p>Materials and method</p><p>Animals</p><p>All procedures were approved by the Animal Care and EthicsCommittee of the University of New South Wales. Female</p><p>SpragueDawley rats (8 weeks, Animal Resource Centre, Yanderra,WA, Australia) were assigned to two groups, one fed a pelleted HFD(n = 15, 20 kJ/g, 43% fat; SF03-020, Specialty Feeds, Glen Forrest,WA, Australia) to induce obesity, and a control group fed standardrodent chow (n = 10, 11 kJ/g, 14% fat, Gordons Specialty Stock-feeds, NSW, Australia) for 6 weeks prior to mating, throughoutgestation and lactation (Chen et al. 2012). At post-natal day 1, littersizes were adjusted to 1012 pups/litter (sex 1 : 1). At 20 days,within each litter half of the male pups were fed chow and the otherhalf were fed HFD. This yielded four groups, male offspring fromchow-fed dams consuming chow (CC) or HFD (CH), and maleoffspring from HFD-fed dams consuming chow (HC) or HFD (HH).There were 23 pups from the same litter within each group.Intraperitoneal glucose tolerance test (IPGTT) was performed at8 weeks (n = 68 in each group) (Chen et al. 2009). After 5 hfasting, baseline glucose (T0) was measured using tail tip blood(Accu-Chek glucose meter, Roche Diagnostics, Nutley, NJ, USA).Blood glucose levels were measured at 15, 30, 60, and 90-min post-glucose injection (2 g/kg, i.p.). Area under the curve (AUC) wascalculated for each rat using the equation 0.5 9 (GT0 + GT15)9 15 min + 0.5 9 (GT15 + GT30) 9 15 min + 0.5 9 (GT30 +GT60) 9 30 min + 0.5 9 (GT60 + GT90) 9 30 min.</p><p>Sample collection and analysis</p><p>At 9 weeks, offspring were fasted for 5 h then anesthetized withPentothal (0.1 mg/g, i.p., Abbott Australasia Pty. Ltd., Lane Cove,NSW, Australia). Half of the rats within each group were injectedwith glucose (0.5 g/kg, i.v.), while the other half received saline ascontrol (n = 56 under each condition). Cardiac blood was collected10 min after injection and glucose levels were measured immedi-ately. The hypothalamic areas containing Arc and paraventricularnucleus (PVN) were isolated (Chen et al. 2009) and stored at80C. Abdominal fat pads (retroperitoneal, mesenteric, andepididymal fat)) and liver were weighed.</p><p>Another cohort (9-week old, n = 1112) was anesthetized as aboveand killed in a non-fasting state. Thewhole hypothalamuswas quicklysliced into 440 lm prisms (McIlwain tissue chopper, Mickle Labo-ratory Engineering Co. Ltd. Guildford, Surrey, UK.), then carefullyplaced into individual wells containing 1 mL of modified Krebssolution (containing 5 mM glucose and no lactate as previouslydescribed (Parton et al. 2007; Bertrand et al. 2011). The culture platewas incubated at 37C/5% CO2 in an orbital incubator at 55 rpm(oscillation amplitude 20 mm) and the Krebs solution was replacedevery 20 min. After 1 h equilibration, a basal 20 min samplecollection was performed. Then, the tissues were challenged with20 mM glucose for 20 min. Superfusates were collected, centrifugedand stored at 20C. Brain tissue was kept for later determination ofmRNA expression of NPY, POMC, mTOR, Glucose transporter(GLUT)1, monocarboxylate transporter (MCT)2 and 4, and LDHb.</p><p>Plasma triglyceride (TG) was measured using glycerol standard(Sigma, St. Louis, MO, USA) and TG reagent (Roche Diagnostics)(Chen et al. 2008); plasma insulin was measured using radioim-munoassay (Linco, St. Charles, MI, USA). Glucose and lactateconcentrations in the hypothalamus superfusate were measuredusing an electrode based glucose analyzer (EML105-analyzer,Radiometer Medical A/S, Copenhagen, Denmark) (Simar et al.2012). The difference in glucose between Krebs solution and thesuperfusate after 20 min incubation was calculated as hypothalamic</p><p> 2013 International Society for Neurochemistry, J. Neurochem. (2014) 129, 297--303</p><p>298 H. Chen et al.</p></li><li><p>glucose consumption in vitro. As the Krebs solution was lactate free,the lactate in the superfusate represented hypothalamic lactaterelease.</p><p>Tissue total RNA was extracted using TRIZOL reagent (Invitro-gen Australia Pty Ltd, VIC, Australia). First-strand cDNA wassynthesized from total RNA using M-MLV Reverse Transcriptase,RNase H-, Point Mutant Kit (Promega, Madison, WI, USA).Manufacturer optimized and validated TaqMan probe/primers(Applied Biosystem, Foster City, CA, USA) were used forquantitative real-time PCR (Eppendorf Realplex2, Hamburg, Ger-many). The target gene and housekeeping 18s rRNA probes werelabeled with FAM and VIC, respectively. Gene expression wasquantified in a single multiplexing reaction, by standardizing thegene of interest to 18s rRNA. The average value of the CC groupwas assigned as the calibrator, against which all other samples areexpressed as fold difference. In both Arc and whole hypothalamusmRNA expression of NPY, POMC was measured. In addition,mRNA expression of mTOR, GLUT1, MCT2, MCT4, LDHB wasmeasured in the whole hypothalamus. In the PVN, mRNAexpression of NPY, POMC, and GLUT1 was measured.</p><p>The results were expressed as mean SE. The differencebetween groups was analyzed using two-way analysis of variance(Statistica 10, StatSoft Inc., Tulsa, OK, USA). If there wassignificant interaction between maternal obesity and post-natalHFD consumption, the data were further analyzed by post hocFishers Least Significance Difference tests. Pearson correlationcoefficient was used to assess the correlations between lactateconcentration and mRNA expression of hypothalamic markers(Statistica 10). Differences between saline and glucose-injected ratswithin the same group were analyzed using Students t-test(Statistica 10). p &lt; 0.05 was considered as significantly differentbetween groups.</p><p>Results</p><p>Growth and adiposity</p><p>At 9 weeks of age, pups from obese mothers had significantlygreater body weight, liver weight and fat mass (p &lt; 0.05,maternal effect, n = 1115, Table 1). Post-weaning HFDconsumption also significantly increased body and liverweights, with fat mass more than doubled in rats comparedwith their chow-fed litter mates (p &lt; 0.05, post-natal HFD</p><p>effect, Table 1). There were significant interactions betweenmaternal obesity and post-weaning HFD consumption onretroperitoneal, mesenteric, and epididymal fat mass (post hoctest p &lt; 0.05, HH vs. CH, Table 1).</p><p>In vivo response to glucose injectionAt 8 weeks, baseline (T0) glucose levels during IPGTT weresimilar between groups (Fig. 1a). Maternal obesity exerted amore pronounced effect than post-weaning HFD consump-tion to increase blood glucose levels during IPGTT. Maternalobesity led to increased glucose levels from 15 to 90-minpost-glucose injection, whereas post-natal HFD consumptionincreased blood glucose level only at 15 min and 30-minpost-glucose injection (p &lt; 0.05, n = 68, Fig. 1a). Signif-icant maternal and post-natal HFD effects, as well as asignificant interaction between these two factors were shownon AUC (p &lt; 0.05, Fig. 1a), with the AUC of HH nearlydoubled compared to HC (p &lt; 0.05 post hoc test); however,it was not different between the two groups from the leanmothers (Fig. 1a).At 9 weeks, saline-injected rats showed similar glucose</p><p>levels between groups; while both plasma insulin and TGlevels were significantly higher in post-natal HFD-fed rats(p &lt; 0.05, post-natal HFD effect, n = 56, Fig. 1c, d).Glucose injection significantly increased blood glucoseconcentrations in all groups (p &lt; 0.05, n = 56, Fig. 1b),but only significantly increased plasma insulin levels in theHC rats (p &lt; 0.05, glucose effect, Fig. 1c). In saline-injectedrats, Arc NPY mRNA was significantly down-regulated byboth maternal obesity and post-weaning HFD consumption(p &lt; 0.05, maternal and post-weaning HFD effects, Fig. 1e).There was a significant interaction between maternal obesityand post-natal HFD on Arc NPY mRNA expression(p &lt; 0.05), with the level in CH and HC significantly lowerthan CC (p &lt; 0.05, post hoc tests, n = 56, Fig. 1e). POMCmRNA was only up-regulated by post-natal HFD consump-tion (Fig. 1f). There was no difference in Arc GLUT1mRNA expression (data not shown), or in PVN NPY, POMCand GLUT1 mRNA expression (data not shown).</p><p>Table 1 Parameters of male offspring from chow and HFD-fed dams at 9 weeks</p><p>CCn = 11</p><p>CHn = 11</p><p>HCn = 15</p><p>HHn = 12</p><p>Body Weight (g)*# 303 12 408 11 349 10 493 12Liver (g)*# 12.6 0.8 19.3 0.7 14.7 0.6 25.0 1.4Retroperitoneal fat (g)*#a 1.53 0.18 6.28 0.47# 2.19 0.16 9.40 0.43*#Mesenteric fat (g)*#a 2.14 0.17 5.62 0.37# 2.7...</p></li></ul>

Recommended

View more >