Free Radical Biology & Medicine 44 (2008) 1919–1925

Contents lists available at ScienceDirect

Free Radical Biology & Medicine

j ourna l homepage: www.e lsev ie r.com/ locate / f reeradb iomed

Original Contribution

Maternal nicotine exposure increases oxidative stress in the offspring

Jennifer E. Bruin a, Maria A. Petre a, Megan A. Lehman a, Sandeep Raha b, Hertzel C. Gerstein c,Katherine M. Morrison b, Alison C. Holloway a,⁎a Reproductive Biology Division, Department of Obstetrics and Gynecology, McMaster University, Hamilton, Ontario, Canada L8N 3Z5b Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada L8N 3Z5c Department of Medicine, McMaster University, Hamilton, Ontario, Canada L8N 3Z5

a r t i c l e i n f o

⁎ Corresponding author. Fax: +1 905 524 2911.E-mail address: hollow@mcmaster.ca (A.C. HollowayAbbreviations: PND, postnatal day; nAChR, nicotinic a

8-isoprostaglandin F2α; ROS, reactive oxygen species; GPMnSOD, manganese superoxide dismutase; CuZnSODmutase; HO-1, heme oxygenase 1; H2DCFDA, 2′,7′-dichtate; HBSS, Hank's balanced salt solution; FBS, fetal bosaline; TTBS, Tris-Tween-buffered saline; DTT, dithiothrtide triphosphate; SDS-PAGE, sodium dodecyl sulfate pesis; RT, reverse transcription; PCR, polymerase chain re

0891-5849/$ – see front matter © 2008 Elsevier Inc. Aldoi:10.1016/j.freeradbiomed.2008.02.010

a b s t r a c t

Article history:Received 11 July 2007Revised 25 January 2008Accepted 20 February 2008Available online 4 March 2008

Fetal and neonatal nicotine exposure causes β-cell apoptosis and loss of β-cell mass, but the underlyingmechanisms are unknown. The goal of this study was to determine whether maternally derived nicotine canact via the pancreatic nicotinic acetylcholine receptor (nAChR) during fetal and neonatal development toinduce oxidative stress in the pancreas. Female Wistar rats were given saline or nicotine (1 mg/kg/day) viasubcutaneous injection for 2 weeks prior tomating until weaning (postnatal day 21). In male offspring, nAChRsubunit mRNA expression was characterized in the developing pancreas and various oxidative stress markerswere measured at weaning following saline and nicotine exposure. The nAChR subunits α2-α4, α6, α7, andβ2–β4were present in the pancreas during development. Fetal and neonatal exposure to nicotine significantlyincreased pancreatic GPx-1 and MnSOD protein expression, as well as islet ROS production. Furthermore,protein carbonyl formation was higher in nicotine-exposed offspring relative to controls, particularly withinthe mitochondrial fraction. There was also a nonsignificant trend toward higher serum 8-isoPG levels. Thesedata suggest that β-cell apoptosis in the fetal and neonatal pancreas may be the result of a direct effect ofnicotine via its receptor and that this effect may be mediated through increased oxidative stress.

© 2008 Elsevier Inc. All rights reserved.

Keywords:NicotinePancreasOxidative stressNicotinic acetylcholine receptorReactive oxygen speciesAntioxidant enzymes

Introduction

Cigarette smoking during pregnancy is associated with a numberof adverse obstetrical outcomes including placenta previa, prematurerupture of the membranes, preterm birth, and low birth weight [1–6].Moreover, recent epidemiologic studies have shown a strong relation-ship between maternal smoking and subsequent obesity, hyperten-sion, and type 2 diabetes in the offspring [7,8]. We have previouslyshown in a rat model that maternal exposure to nicotine alone duringpregnancy and lactation results in permanent loss of β-cell mass andfunction in the offspring [9,10]. These results may partially explain theincreased risk of type 2 diabetes in children born to women whosmoked during pregnancy [11], but the mechanism(s) underlying thisβ-cell loss have not yet been identified.

It has been demonstrated that maternal smoking is associated withincreased levels of oxidative stress markers in mothers, newborns, and

).cetylcholine receptor; 8-isoPG,x-1, glutathione peroxidase 1;, copper/zinc superoxide dis-lorodihydrofluorescein diace-vine serum; TBS, Tris-bufferedeitol; dNTP, deoxyribonucleo-olyacrylamide gel electrophor-action.

l rights reserved.

infants [12,13]. Furthermore, there is considerable evidence in vivo andin vitro to suggest that exposure to nicotine results in increased oxi-dative stress in fetal, neonatal, and adult tissues [14–17]. Indeed, in adultrats, nicotine exposure has been shown to increase oxidative stress inpancreatic tissue in vitro [16] and to produce oxidative tissue injuries invivo [18,19]. Because the pancreatic β-cell has low expression of anti-oxidantenzymes [20,21], it is particularly susceptible to oxidative stress-mediated tissue damage including increased β-cell death [22–27].

The current study uses an animal model of fetal and neonatalexposure to nicotine, which has previously been shown to cause in-creased β-cell apoptosis at birth and weaning [9,10,28]. It was hy-pothesized that maternally administered nicotine would activate thefetal and neonatal pancreatic nicotinic acetylcholine receptors (nAChR),to cause an imbalance in the prooxidative/antioxidative status of theβ-cell (oxidative stress), thus signaling for β-cell apoptosis. The nAChRbelongs to a family of neurotransmitter-gated ion channels [29] that arehomo- orheteropentamers comprised of various combinationsofα- andβ-subunits (α2–α10 and β2–β4) [30]. The nAChR is best characterized inthe brain; however, these neuronal receptors also exist in variousnonneuronal cell types [31,32], including adult pancreatic β-cells [33]. Ithas been suggested that nicotinemay affect adult pancreatic function bydirect interaction with the pancreatic nAChR [33]. However, expressionof nAChR subunits has not been previously examined in the developingfetal or neonatal pancreas.

Therefore, the goals of this study were: (1) to determine the pat-tern of nAChR subunit expression in the developing pancreas, (2) to

1920 J.E. Bruin et al. / Free Radical Biology & Medicine 44 (2008) 1919–1925

examine whether fetal and neonatal nicotine exposure alters pro-and/or antioxidant markers in the pancreas, and (3) to assess whetherthe pancreatic and/or systemic oxidative balance has been disrupted.

Materials and methods

Maintenance and treatment of animals

All animal experiments were approved by the Animal ResearchEthics Board at McMaster University, in accordance with the guide-lines of the Canadian Council for Animal Care. Nulliparous 200–250 gfemale Wistar rats (Harlan, Indianapolis, IN) were maintained undercontrolled lighting (12:12 L:D) and temperature (22°C) with adlibitum access to food and water. Dams were randomly assigned(n=10 per group) to receive saline (vehicle) or nicotine bitartrate(1 mg kg−1 day−1, Sigma-Aldrich Inc., St. Louis, MO) via subcutaneousinjection daily for 2 weeks prior to mating until weaning. At postnatalday 1 (PND1) litters were culled to eight to assure uniformity of littersize between treated and control litters. To eliminate any confoundingeffects of the female reproductive cycle, onlymale offspringwere usedin this study. After weaning at postnatal day 21 (PND21), male off-spring (n=10 per group) were selected randomly for the experimentsdescribed below.

Reverse transcription (RT) and real-time PCR

Pancreatic tissue from pups born to saline- and nicotine-exposedmothers was removed at PND1 and PND21 and immediately placed inRNA later (Sigma-Aldrich Inc.) for analysis by either semiquantitativeRT-PCR or quantitative real-time PCR. The mRNA expression of nAChRsubunits (α2,α3,α4,α5,α6,α7,β2,β3, andβ4)was determined at birth(PND1) and weaning (PND21) by RT-PCR to assess the pattern of thereceptor subunit expression during fetal and neonatal development.The mRNA expression of the antioxidant enzymes heme oxygenase 1(HO-1), glutathione peroxidase (GPx), manganese superoxide dismu-tase (MnSOD), and copper/zinc superoxide dismutase (CuZnSOD) wasdetermined at PND21 by real-time PCR to compare the antioxidantresponse following nicotine or saline exposure. RNA was extractedusing a Qiagen RNA extraction kit (RNeasy Mini Kit, Mississauga, ON,Canada) and treated with TURBO DNA-free DNase-1 (Ambion, Austin,TX) according tomanufacturer's instructions; RNAwas stored at −80°Cprior to use.

RNA samples (n=6 per group) were reverse-transcribed to cDNA ina 20-µl reaction mixture containing 1 µg of extracted RNA, 1 µl ofrandom primers, 1 µl of dNTPs (10 mM), 4 µl of 5X First-Strand Buffer,1 µl of 0.1MDTT,1 µl of RNaseOUT,1 µl of Superscript III RT, andDNase/RNase-free water (Invitrogen, Carlsbad, CA), according to the standard

Table 1Primer sequences for reverse transcription and real-time PCR

Gene Forward primer

Glutathione peroxidase 5' - CGA CAT CGA ACC CGA TAT AGAMnSOD 5' - CCT TTC CCT GAC AAG GTA CACCuZnSOD 5' –TGG GTT CCA TGT CCA TCA ATAHO-1 5' - ACA CCA GCC ACA CAG CAC TAnAChR alpha 2 5' - CTC CTG CAG CAT CGA TGT GAC CnAChR alpha 3 5' - GGA GAA GTG ACT TGG ATC CnAChR alpha 4 5' - GCC ATC TAT AAG AGC TCC TGC AnAChR alpha 5 5' - CGA ACG TCT GGT TGA AGCnAChR alpha 6 5' - TCT TAA GTA CGA TGG GGT GAT AnAChR alpha 7 5' - TTG CCA GTA TCT CCC TCC AGnAChR beta 2 5' - GCT GAC GGC ATG TAC GAA GnAChR beta 3 5' - CTC ATT ATC CAC CTC CGT TTnAChR beta 4 5' - GGT TGC CTG ACA TCG TGT TGBeta actin 5' - GCT GTG CTA TGT TGC CCT AGA C

protocol suppliedwith each product. The reactionwas carried out in aniCycler thermal cycler (Bio-Rad, Hercules, CA) using the followingprogram: 5 min at 25°C, 60 min at 50°C, and 15 min at 70°C. The cDNAwas stored at −20°C until use.

RT-PCR to determine the pattern of nAChR subunit expression wasprepared in 50-µl reactions with 40.1 µl of RNase/DNase-free water,5.0 µl of 10XMg-free PCRbuffer,1.5 µl ofMgCl2 (50mM),1.0 µl of dNTPs(10 mM), 0.4 µl of Taq DNA Polymerase (Invitrogen), 1.0 µl templatecDNA, and 1.0 µl of the primer mix (10 µM, MOBIX, McMaster Uni-versity, Hamilton, ON, Canada). Primer sequences for nAChR subunitsare provided in Table 1. The cDNAwas amplified in the iCycler thermalcycler using the following program: 2 min at 94°C, followed by 35cycles of 30 s denaturing at 94°C, 30 s annealing at 55°C, 1 minelongation at 72°C, and then storage at 4°C. PCR products wereseparated on a 2% agarose gel and visualized using ethidium bromide(EMD, Gibbstown, NJ). PCR products were imaged with the UVPBioimaging Systems Epi Chemi II Darkroom and Labworks software(UVP Inc., Upland, CA).

Real-time PCR tomeasure antioxidant enzyme expression in saline-and nicotine- exposed pancreas was performed using SYBR Greenchemistry (n=5 per treatment group). Twenty-five-microliter reac-tions were prepared with 10.5 µl of RNase/DNase-free water, 12.5 µl ofiQ SYBR Green Supermix (Bio-Rad Laboratories), 1.0 µl of templatecDNA,1.0 µl of the primermix (25 µM for beta actin; 10 µM forMnSOD,CuZnSOD, and GPx-1; 2 µM for HO-1; MOBIX, McMaster University).Primer sequences for antioxidant enzymes are provided in Table 1. ThecDNA was amplified in an iCycler thermal cycler coupled with aniCycler IQ Multicolor real-time PCR detection system using iCycler IQsoftware v3.1.7050 (Bio-Rad Laboratories). Real-time PCRwas run for 1cycle (50°C for 2 min, 95°C for 10 min) followed immediately by 40cycles (95°C for 15 s, 60°C for 60 s), and fluorescence was measuredafter each of the repetitive cycles. Emission datawere quantified usingthe threshold cycle (Ct) value. A melting point dissociation curve ge-nerated by the instrument was used to confirm that only a singleproduct was present with each set of primers. Gene expression of eachantioxidant enzymewas quantified as the average Ct value normalizedto the beta actin Ct value for the same sample.

Western blotting

Protein expression was measured in whole pancreas homogenatesfromnicotine and saline-exposed offspring atweaning (PND21). Proteinwas extracted from the pancreas (n=4 per group) using RIPA lysis buf-fer (15 mM Tris–HCl, 1% (v/v) Triton X-100, 0.1% (w/v) SDS, 167 mMNaCl, 0.5% (w/v) sodiumdeoxycholatic acid), with CompleteMini EDTA-free protease inhibitors (Roche Applied Science, Laval QC, Canada).Thirty micrograms of protein was subjected to SDS-PAGE using a 10%

Reverse primer

5' - CCA TCA CCA AGC CCA GAT AC5' - CAA ATG CTG CAC AGG AAT ACA5' – CTG GAC CGC CAT GTT TCT TA5' - CCA GCA GCT CAG GAT GAG TA

TT CTT 5' - GAG ATG CAC AGC GTG ATC TTC TCT CCA C5' - CAA GTG GGC ATG GTG TGT G

GC ATC 5' - CTT CTC GCC AAA CTC TGA AGG CAG ATA G5' - CAC CAT AAT GGA ATA GGG

AC 5' - AAC ATG GTC TTC ACC CAC TTG5' - CTT CTC ATT CCT TTT GCC AG5' - GGA GGT GGG AGG CAC AAT C5' - CTG TAT CAC TCT CCT TTC CAT CC5' - GCC AAT GAG CGG TAT GTC5' - ACC GCT CAT TGC CGA TAG T

Fig. 1. mRNA expression for nAChR subunits in a representative saline control pancreasat: (A) PND1 and (B) PND21. Lane: (1) 100 bp DNA ladder; (2) α2; (3) α3; (4) α4; (5) α5;(6) α6; (7) α7; (8) β2; (9) β3; (10) β4; (11) 100 bp DNA ladder.

Table 2mRNA expression of antioxidant proteins in the pancreas at 3 weeks of age, asdetermined by quantitative real-time PCR

Gene Saline Nicotine P value

HO-1 1.43±0.018 1.42±0.016 0.628GPx 1.17±0.016 1.19±0.022 0.486MnSOD 1.36±0.089 1.35±0.020 0.640CuZnSOD 1.23±0.012 1.21±0.021 0.377

All gene expressionwas quantified relative to a β-actin loading control. Data are presentedas the mean±SE.

1921J.E. Bruin et al. / Free Radical Biology & Medicine 44 (2008) 1919–1925

separating gel and then electro-transferred to PVDF blottingmembrane(Bio-Rad Laboratories, CA). Membraneswere blocked overnightwith 5%(w/v) skim milk in TBST (TBS, 0.5% (v/v) Tween 20) at 4°C and thenincubated for 1 h at room temperature in primary antibody on a rockingplatform. The membrane was cut horizontally at approximately 40 kDaand the upper molecular weight portion was incubated in rabbit poly-clonal anti-beta actin for all blots (47 kDa running weight; 1:2000dilution, AbCam,MA). The lowermolecularweight portion of blotswereincubated with primary antibodies for MnSOD (rabbit polyclonal,1:5000, 25 kDa, Santa Cruz Biotechnology, CA), CuZnSOD (rabbit poly-clonal, 1:1000, 23 kDa, Santa Cruz Biotechnology), GPx-1 (rabbit polyc-lonal, 1:5000, 22 kDa, AbCam, MA), and HO-1 (mouse monoclonal,1:4000, 35 kDa, Stressgen Biotechnologies, BC, Canada). After washingwith TBST, blots were incubated with peroxidase-conjugated secondaryanti-rabbit (1:2000; Santa Cruz, CA) or anti-mouse (1:2000; AmershamBiosciences, NJ) antibodies for 1 h at room temperature on a rockingplatform. Blots were washed thoroughly in TBST followed by TBS afterimmunoblotting. Reactive protein was detected with ECL Plus chemi-luminescence (Amersham Biosciences) and Bioflex X-ray film (ClonexCorporation, ON). Densitometric analysis of immunoblots was per-formed using ImageJ 1.37v software (National Institutes of Health,Bethesda, MD); all proteins were quantified relative to beta actin.

Reactive oxidative species production by isolated pancreatic islets

Islet isolation from rats was performed as previously described[34]. Briefly, the pancreas was immediately excised following sacrificeand placed in 6 ml of Hank's balanced salt solution (HBSS) (HyClone,Logan, UT) containing 4 mg/ml collagenase type IA (Sigma-Aldrich),100 IU/ml penicillin G, and 0.25 μg/ml streptomycin (Gibco, GrandIsland, NY). The pancreas was minced finely and the resulting sus-pension was incubated at 37°C for 40 min. The reaction was thenquenched with 20 ml HBSS supplemented with 10% fetal bovineserum (FBS) (HyClone), 100 IU/ml penicillin G, and 0.25 μg/ml strep-tomycin. Islets were manually picked from the suspension using a

small glass pipette and a dissecting microscope. The islets were in-cubated at 37°C, 5% CO2/95% normal atmosphere in 5 ml RPMI 1640(Life Technologies, Burlington, ON) supplemented with 10% FBS,100 IU penicillin G, and 0.25 μg/ml streptomycin for 48 h.

ROS production by isolated islets at PND21 following saline andnicotine exposure was measured using 2′,7′-dichlorodihydrofluores-cein diacetate (H2DCFDA) (Molecular Probes Inc., Eugene, OR) fluo-rescence as previously described [35]. Briefly, 100 islets from saline-and nicotine-exposed offspring (n=3 per group) were washed twicewith PBS. Following centrifugation, the supernatant was removed andthe pelleted islets were then resuspended in 100 μl of PBS containing100 μM H2DCFDA and incubated for 3 h at 37°C. Because of therelatively low number of cells in this assay, a long incubation periodallows for the diffusion of the oxidized dye from inside the cell back outinto the culture medium [35]. This approach has been previouslyvalidated to determine ROS production by isolated rat islet cells [35]. Inaddition, since H2DCFDAmust bemade fresh immediately prior to use,islets isolated on different days were incubated in different batches ofreagent. To account for day-to-day variability within the experiment, a43 µM hydrogen peroxide reaction was prepared with each batch ofH2DCFDA to calibrate the performance of the dye. The hydrogenperoxidewas added to 100 µMH2DCFDA and incubated inparallelwiththe islet reactions. Following the incubation period, the islets werevigorously disrupted to release intracellular H2DCFDA. Both the isletsuspensions and the hydrogen peroxide control were centrifuged, andthe supernatants were transferred to black 96-well plates (BD Falcon,Mississauga, ON). Fluorescence of the 2′,7′-dichlorofluorescein pro-ductwas determinedusing a SpectaMaxGemini XS (Molecular DevicesCorp., Sunnyvale, CA) microplate spectrofluorometer at excitation andemission wavelengths of 505 and 540 nm, respectively. All measure-ments of islet ROS productionwere normalized to the 43 µM hydrogenperoxide control.

Protein carbonyl detection

To assess oxidative damage to pancreatic proteins, the presence ofprotein carbonyl groups was quantified using the OxyBlot proteinoxidation detection kit (Chemicon International, Temecula, CA). For-mation of protein carbonyl groups was measured at PND21 in bothwhole pancreas homogenates (n=5 per group) and in the mitochon-drial fraction of the pancreas (n=4 per group) from saline- andnicotine-exposed offspring. Whole pancreas homogenates wereprepared and quantified as described above. To separate the mito-chondrial fraction, the Compartmental Protein Extraction Kit(K3013010; Biochain Institute Inc., Hayward, CA) was used accordingto manufacturer's instructions. Protein samples were then preparedwith the Oxyblot kit, according to manufacturer's instructions.Briefly, equal amounts of protein (16 µg for whole pancreas, and10 µg for mitochondrial fractions) were derivatized with either 2,4-dinitrophenylahydrazine (DNPH) or a derivatization-control solu-tion. Gel electrophoresis, transfer to a PVDF membrane, immuno-blotting conditions, and detection of reactive protein were the sameas above. Blotswere incubated for 1 h in rabbit-DNP antibody (1:150),

Fig. 2. Quantification of protein expression for antioxidant enzymes: (A) HO-1, (B) GPx-1, (C) MnSOD, and (D) CuZnSOD in saline-exposed (black bar) and nicotine-exposed (whitebar) pancreas at 3 weeks of age (n=4 per group). A representative Western blot for each protein is presented in panel E. All protein expression was quantified relative to a β-actinloading control. Data are presented as the mean±SE. Values with an asterisk are significantly different from the saline control (Pb0.05).

Fig. 3. ROS production in PND21 isolated islets of saline-exposed (black bar) andnicotine-exposed (white bar) offspring (n=3 per group). Data are expressed as relativefluorescent units (normalized to a hydrogen peroxide control in H2DCFDA)±SE. Valueswith an asterisk are significantly different (Pb0.05).

1922 J.E. Bruin et al. / Free Radical Biology & Medicine 44 (2008) 1919–1925

followed by 1 h in secondary goat anti-rabbit IgG (HRP-conjugated;1:300). No reactive proteinwas detected in the derivatization-controlimmunoblots. Densitometric analysis of immunoblots was per-formed using ImageJ 1.37v software (National Institutes of Health,Bethesda, MD).

8-iso prostaglandin F2α

Blood samples (n=10 per group) for the analysis of 8-iso-prostaglandin F2α (8-isoPG), a marker of lipid peroxidation, werecollected immediately after sacrifice from saline- and nicotine-exposed animals at weaning (PND21). Blood samples were allowedto clot at 4°C and then centrifuged and the serumwas stored at −80°Cuntil assayed. 8-isoPG concentrations were determined using theCorrelate EIA Direct 8-iso-prostaglandin F2α enzyme immunoassaykit (Assay Designs, MI) according to manufacturer's instructions.

Statistical analysis

All statistical analyses were performed using SigmaStat (v.2.03,SPSS, Chicago, IL). Data from nicotine-exposed offspring were com-pared to the control group using Student's t test (α=0.05).

Results

nAChR subunit mRNA expression

At PND1 and PND21 all of the nAChR subunits except α5 werepresent in the pancreas of both saline- and nicotine-exposed offspring(Fig. 1).

Antioxidant enzyme expression

Fetal and neonatal exposure to nicotine did not alter mRNA ex-pression of HO-1, GPx, MnSOD, or CuZnSOD in the pancreas at PND21(Table 2). Nicotine exposure significantly increased the proteinexpression of both GPx-1 and MnSOD in the pancreas at PND21(Figs. 2B and C, respectively; Pb0.05), but did not alter the proteinexpression of HO-1 or CuZnSOD (Figs. 2A and D, respectively).

Reactive oxygen species production by isolated pancreatic islets

ROS production by isolated pancreatic islets was significantly ele-vated in nicotine-exposed offspring relative to saline control animals atPND21 (Fig. 3; Pb0.05).

Protein carbonyl formation

To assess whether the balance of ROS production and antioxidantenzyme expression has been disrupted by nicotine exposure, oxidativedamage to proteinwasmeasured in the pancreas at PND21. Total proteincarbonyl levels were unchanged in the whole pancreas (Fig. 4), but asignificant increase in oxidative damage to a 25-kDa protein wasobserved (Fig. 4; Pb0.05). Furthermore, total protein carbonyl levelswere approximately 3 times higher in the mitochondrial fraction fromnicotine-exposed relative to saline-exposed pancreas (Fig. 4; Pb0.001).

8-iso prostaglandin F2α

Mean serum levels of 8-isoPG in nicotine-exposed animals were 3times higher than in control offspring, but due to variability in the

Fig. 4. (A) Quantification of protein carbonyl groups in whole tissue homogenates(n=5 per group) and mitochondrial fractions (n=4 per group) from the pancreas ofsaline-exposed (black bar) and nicotine-exposed (white bar) offspring at 3 weeks of age.(B) A representative Western blot of protein carbonyl groups is shown for the wholepancreas and mitochondrial fraction from saline-exposed (S) and nicotine-exposed (N)offspring. The black arrow indicates a 25-kDa protein that is significantly altered by fetaland neonatal nicotine exposure in thewhole pancreas homogenates. Data are presentedas the mean±SE. Values with an asterisk are significantly different from the salinecontrol (Pb0.05); a double asterisk indicates a P valueb0.001.

1923J.E. Bruin et al. / Free Radical Biology & Medicine 44 (2008) 1919–1925

nicotine-exposed offspring, this difference did not reach significance(saline 256±48.6 pg/ml, nicotine 767±375.4 pg/ml; P=0.17).

Discussion

Recent epidemiological studies have shown that the offspringof women who smoke during pregnancy have an increased risk ofdeveloping obesity, hypertension, and type 2 diabetes [7,8,11,36–40],although the mechanism(s) underlying these associations are un-known. We have previously demonstrated in a rat model that mater-nal nicotine exposure during pregnancy and lactation causes β-cellapoptosis in fetal and neonatal offspring, followed by development ofimpaired glucose metabolism in early adult life [9,10,28]. We hy-pothesized that nicotine acts directly on the nicotinic acetylcholinereceptor in the developing pancreas to induce oxidative stress, andultimately, β-cell apoptosis. Indeed, results from this study demon-strate that subunits of the nAChR are present during early postnatallife, and that in the pancreas, perinatal nicotine exposure results inboth increased oxidative stress and increased β-cell apoptosis.

In this study, the α2 – α4, α6 – α7, and β2 – β4 subunits of nAChRwere all present in the pancreas at birth and weaning. Nicotine-exposed offspring have a mean serum cotinine (the major metaboliteof nicotine) concentration of 26.2±1.78 ng/ml at birth [41], whichprovides evidence that maternally administered nicotine reaches thepups. The presence of the nAChR subunits in the fetal and neonatalpancreas suggests that this maternally derived nicotine may act vianonneuronal nicotinic acetylcholine receptors to cause the observedincrease in β-cell death in the offspring [9,10,28]. However, furtherbinding studies and coadministration of an nAChR antagonist wouldbe required to determine conclusively that nicotine is exerting itseffects by direct activation of the nAChR.

Nicotine exposure has been shown to increase oxidative stressin pancreatic tissue in vitro [16] and it is well established that pancreaticβ-cells are particularly susceptible to oxidative stress-mediated tissuedamage due to their low level of antioxidant enzyme expression [20,21].Therefore, we hypothesized that the observed increase in β-cell apop-tosis in neonatal nicotine-exposed offspring [9,10,28] might be due, inpart, to nicotine-induced oxidative stress in the developing pancreas.Indeed, nicotine-exposed offspring had elevated expression of antiox-idant proteins, MnSOD and GPx-1, in the pancreas and a concomitantrise in islet ROS production. Cellular oxidative status is a balance be-tween oxidative stress and antioxidant capacity. Therefore if the level ofROS exceeds the antioxidant capacity of the cell, oxidative stress willensue [25,42]. Data from this study suggest that fetal and neonatalnicotine exposure triggers both the prooxidant and the antioxidantresponse in the pancreas.

The consequences of oxidative stress include damage tomitochon-dria, cellular proteins, lipids, and nucleic acids [42,43], which can leadto cell death through a variety of mechanisms [22]. Indeed, nicotine-exposed animals had evidence of oxidative protein damage, as therewas increased protein carbonyl formation in the nicotine-exposedoffspring relative to saline controls. Interestingly, individual proteins inboth the whole pancreas and the isolated mitochondrial fraction ap-peared to be particularly susceptible to oxidative damage. This trendhas also been observed by other groups who report increased proteincarbonyl modification to susceptible proteins [44–46]. In a futurestudy, we plan to identify specific proteins that are targeted by oxi-dative stress in this animal model using two-dimensional gelelectrophoresis andmass spectrometry. Furthermore, our data suggestthat the oxidative stress is not limited to the pancreas, as there was atrend toward higher levels of 8-isoPG (a marker of whole bodyoxidative stress) in the nicotine-exposed offspring at weaning. Takentogether, these results suggest that developmental nicotine exposureinduces an antioxidant response that is associated with loss of redoxbalance in the pancreas.

Results from this study also indicate that the increased oxidativestress in the pancreas may differentially affect the mitochondria. Astriking increase in protein carbonyl levels was observed in the iso-lated mitochondrial fraction from the pancreas following nicotineexposure whereas there was no significant difference in protein car-bonyl formation in the whole pancreas homogenate. Furthermore, theincrease in MnSOD protein expression, but not CuZnSOD proteinexpression, also indicates that the oxidative stress may be localizedwithin the mitochondria. The superoxide dismutase antioxidantresponse is essential for protecting the cell from oxidative stress;SOD enzymes catalyze the conversion of the highly reactive super-oxide anion to hydrogen peroxide, which can then be safely convertedtowater andmolecular oxygen by catalase and glutathione peroxidase[47]. Mitochondrial DNA is more vulnerable to ROS damage thannuclear DNA [48], whichmay explainwhy elevated levels of ROS in theislets induced the mitochondrial SOD response (MnSOD) and not thecytosolic response (CuZnSOD). In addition, the mitochondria not onlyare targeted by ROS but also are the major source of ROS production inthe cell. The iron–sulfur centers of the electron transport chain (ETC)enzyme complexes within the mitochondrial inner membrane are

1924 J.E. Bruin et al. / Free Radical Biology & Medicine 44 (2008) 1919–1925

extremely sensitive to ROS inactivation [43,49]. When the function ofelectron carrier complexes is impaired, electrons build up at the initialstages of the ETC, leading to further production of ROS [43,43].Oxidative damage within the mitochondria can lead to mitochondrialswelling and ultimately trigger programmed cell death [22,49,50]. Theexact role of the mitochondria in the observed β-cell apoptosis in thisanimal model has been examined in a separate study [51].

Results from this study have shown thatmaternal nicotine, deliveredto the fetus and neonate through either cigarette smoking or nicotinereplacement therapy, during pregnancy and lactation, may act directlyon the nAChR in the developing pancreas to induce oxidative stressand subsequent β-cell loss in the pancreas. Furthermore, this oxidativestress may target the mitochondria, suggesting a potential mechanismthrough which fetal and neonatal nicotine exposure leads to β-cellapoptosis. An early reduction in β-cell mass is associated with an in-creased riskof developing type2diabetes later in life [52,53],whichmayexplain, in part, the increased risk of type 2 diabetes in children born towomenwho smoked during pregnancy. This study also provides furthersupport to the recent concerns about the safety of nicotine replacementtherapy during pregnancy and lactation [54].

Acknowledgments

This work was supported by an operating grant from the CanadianInstitutes of Health Research in Nutrition, Metabolism, and Diabetes toA.C.H. and K.M.M. (MOP 69025). J.E.B. was funded by a CIHR StrategicTraining Program in Tobacco Research Fellowship, a CIHR OntarioWomen's Health Council Doctoral award, and an Ontario GraduateScholarship. J.E.B. andM.A.P. were both provided funding by an AshleyStudentship for Research in Tobacco Control. S.R. was provided salarysupport by Mr. Warren Lammert, Ms Kathy Corkins. and the HamiltonHealth Sciences Foundation. H.C.G. holds the Population HealthInstitute Chair in Diabetes Research (sponsored by Aventis). Wethank the staff of the McMaster University Central Animal Facility forassistance with the animal work.

References

[1] Andres, R. L.; Day, M. C. Perinatal complications associated with maternal tobaccouse. Semin. Neonatol. 5:231–241; 2000.

[2] Cnattingius, S.; Lambe, M. Trends in smoking and overweight during pregnancy:prevalence, risks of pregnancy complications, and adverse pregnancy outcomes.Semin. Perinatol. 26:286–295; 2002.

[3] Faiz, A. S.; Ananth, C. V. Etiology and risk factors for placenta previa: an overviewand meta-analysis of observational studies. J. Matern. Fetal Neonatal Med.13:175–190; 2003.

[4] Nordentoft, M.; Lou, H. C.; Hansen, D.; Nim, J.; Pryds, O.; Rubin, P.; Hemmingsen, R.Intrauterine growth retardation and premature delivery: the influence of maternalsmoking and psychosocial factors. Am. J. Public Health 86:347–354; 1996.

[5] Robinson, J. S.; Moore, V. M.; Owens, J. A.; McMillen, I. C. Origins of fetal growthrestriction. Eur. J. Obstet. Gynecol. Reprod. Biol. 92:13–19; 2000.

[6] Shiverick, K. T.; Salafia, C. Cigarette smoking and pregnancy I: ovarian, uterine andplacental effects. Placenta 20:265–272; 1999.

[7] Bergmann, K. E.; Bergmann, R. L.; Von,K. R.; Bohm,O.; Richter, R.; Dudenhausen, J.W.;Wahn, U. Early determinants of childhood overweight and adiposity in a birth cohortstudy: role of breast-feeding. Int. J. Obes. Relat. Metab. Disord. 27:162–172; 2003.

[8] Wideroe, M.; Vik, T.; Jacobsen, G.; Bakketeig, L. S. Does maternal smoking duringpregnancy cause childhood overweight? Paediatr. Perinat. Epidemiol. 17:171–179;2003.

[9] Holloway, A. C.; Lim, G. E.; Petrik, J. J.; Foster, W. G.; Morrison, K. M.; Gerstein, H. C.Fetal and neonatal exposure to nicotine inWistar rats results in increased beta cellapoptosis at birth and postnatal endocrine and metabolic changes associated withtype 2 diabetes. Diabetologia 48:2661–2666; 2005.

[10] Bruin, J. E.; Kellenberger, L. D.; Gerstein, H. C.; Morrison, K. M.; Holloway, A. C. Fetaland neonatal nicotine exposure and postnatal glucose homeostasis: identifyingcritical windows of exposure. J. Endocrinol. 194:1–9; 2007.

[11] Montgomery, S. M.; Ekbom, A. Smoking during pregnancy and diabetes mellitus ina British longitudinal birth cohort. BMJ 324:26–27; 2002.

[12] Noakes, P. S.; Thomas, R.; Lane, C.; Mori, T. A.; Barden, A. E.; Devadason, S. G.;Prescott, S. L. Maternal smoking is associated with increased infant oxidative stressat 3 months of age. Thorax 62:714–717; 2007.

[13] Ermis, B.; Ors, R.; Yildirim, A.; Tastekin, A.; Kardas, F.; Akcay, F. Influence ofsmoking on maternal and neonatal serum malondialdehyde, superoxide dis-mutase, and glutathione peroxidase levels. Ann. Clin. Lab. Sci. 34:405–409; 2004.

[14] Zhao, Z.; Reece, E. A. Nicotine-induced embryonic malformations mediated byapoptosis from increasing intracellular calcium and oxidative stress. Birth DefectsRes. B: Dev. Reprod. Toxicol. 74:383–391; 2005.

[15] Crowley-Weber, C. L.; Dvorakova, K.; Crowley, C.; Bernstein, H.; Bernstein, C.;Garewal, H.; Payne, C. M. Nicotine increases oxidative stress, activates NF-kappaBand GRP78, induces apoptosis and sensitizes cells to genotoxic/xenobiotic stressesby a multiple stress inducer, deoxycholate: relevance to colon carcinogenesis.Chem. Biol. Interact. 145:53–66; 2003.

[16] Wetscher, G. J.; Bagchi, M.; Bagchi, D.; Perdikis, G.; Hinder, P. R.; Glaser, K.; Hinder,R. A. Free radical production in nicotine treated pancreatic tissue. Free Radic. Biol.Med. 18:877–882; 1995.

[17] Ozokutan, B. H.; Ozkan, K. U.; Sari, I.; Inanc, F.; Guldur, M. E.; Kilinc, M. Effects ofmaternal nicotine exposure during lactation on breast-fed rat pups. Biol. Neonate88:113–117; 2005.

[18] Husain, K.; Scott, B. R.; Reddy, S. K.; Somani, S. M. Chronic ethanol and nicotineinteraction on rat tissue antioxidant defense system. Alcohol 25:89–97; 2001.

[19] Yildiz, D. Nicotine, its metabolism and an overview of its biological effects. Toxicon43:619–632; 2004.

[20] Lenzen, S.; Drinkgern, J.; Tiedge, M. Low antioxidant enzyme gene expression inpancreatic islets compared with various other mouse tissues. Free Radic. Biol. Med.20:463–466; 1996.

[21] Tiedge,M.; Lortz, S.; Drinkgern, J.; Lenzen, S. Relation between antioxidant enzymegene expression and antioxidative defense status of insulin-producing cells. Dia-betes 46:1733–1742; 1997.

[22] Ryter, S. W.; Kim, H. P.; Hoetzel, A.; Park, J. W.; Nakahira, K.; Wang, X.; Choi, A. M.Mechanisms of cell death in oxidative stress. Antioxid. Redox Signal. 9:49–89; 2007.

[23] Donath, M. Y.; Ehses, J. A.; Maedler, K.; Schumann, D. M.; Ellingsgaard, H.; Eppler,E.; Reinecke, M. Mechanisms of beta-cell death in type 2 diabetes. Diabetes 54Suppl. 2:S108–S113; 2005.

[24] Kajimoto, Y.; Kaneto, H. Role of oxidative stress in pancreatic beta-cell dysfunction.Ann. N.Y. Acad. Sci. 1011:168–176; 2004.

[25] Kaneto, H.; Kawamori, D.; Matsuoka, T. A.; Kajimoto, Y.; Yamasaki, Y. Oxidativestress and pancreatic beta-cell dysfunction. Am. J. Ther. 12:529–533; 2005.

[26] Robertson, R. P. Oxidative stress and impaired insulin secretion in type 2 diabetes.Curr. Opin. Pharmacol. 6:615–619; 2006.

[27] Robertson, R. P.; Harmon, J. S. Diabetes, glucose toxicity, and oxidative stress: acase of double jeopardy for the pancreatic islet beta cell. Free Radic. Biol. Med.41:177–184; 2006.

[28] Holloway, A. C.; Petrik, J. J.; Bruin, J. E.; Gerstein, H. C. Rosiglitazone preventsdiabetes by increasing beta-cell mass in an animal model of type 2 diabetescharacterized by reduced beta-cell mass at birth. Diabetes Obes. Metab. (Oct. 29[Electronic publication ahead of print]); in press.

[29] Karlin, A.; Akabas, M. H. Toward a structural basis for the function of nicotinicacetylcholine receptors and their cousins. Neuron 15:1231–1244; 1995.

[30] Dajas-Bailador, F.; Wonnacott, S. Nicotinic acetylcholine receptors and theregulation of neuronal signalling. Trends Pharmacol. Sci. 25:317–324; 2004.

[31] Maus, A. D.; Pereira, E. F.; Karachunski, P. I.; Horton, R. M.; Navaneetham, D.;Macklin, K.; Cortes, W. S.; Albuquerque, E. X.; Conti-Fine, B. M. Human and rodentbronchial epithelial cells express functional nicotinic acetylcholine receptors. Mol.Pharmacol. 54:779–788; 1998.

[32] Conti-Fine, B. M.; Navaneetham, D.; Lei, S.; Maus, A. D. Neuronal nicotinic receptorsin non-neuronal cells: new mediators of tobacco toxicity? Eur. J. Pharmacol.393:279–294; 2000.

[33] Yoshikawa, H.; Hellstrom-Lindahl, E.; Grill, V. Evidence for functional nicotinicreceptors on pancreatic beta cells. Metabolism 54:247–254; 2005.

[34] Ge, Q.; Dong, Y. M.; Hu, Z. T.; Wu, Z. X.; Xu, T. Characteristics of Ca2+-exocytosiscoupling in isolated mouse pancreatic beta cells. Acta Pharmacol. Sin. 27:933–938;2006.

[35] Simmons, R. A.; Suponitsky-Kroyter, I.; Selak, M. A. Progressive accumulation ofmitochondrial DNA mutations and decline in mitochondrial function lead to beta-cell failure. J. Biol. Chem. 280:28785–28791; 2005.

[36] Blake, K. V.; Gurrin, L. C.; Evans, S. F.; Beilin, L. J.; Landau, L. I.; Stanley, F. J.;Newnham, J. P. Maternal cigarette smoking during pregnancy, low birth weightand subsequent blood pressure in early childhood. Early Hum. Dev. 57:137–147;2000.

[37] Morley, R.; Leeson, P. C.; Lister, G.; Lucas, A. Maternal smoking and blood pressurein 7.5 to 8 year old offspring. Arch. Dis. Child 72:120–124; 1995.

[38] Toschke, A.M.; Koletzko, B.; Slikker Jr.,W.;Hermann,M.; Von, K. R. Childhood obesityis associatedwithmaternal smoking inpregnancy. Eur. J. Pediatr.161:445–448; 2002.

[39] Vik, T.; Jacobsen, G.; Vatten, L.; Bakketeig, L. S. Pre- and post-natal growth inchildren of womenwho smoked in pregnancy. Early Hum. Dev. 45:245–255; 1996.

[40] Von, K. R.; Toschke, A. M.; Koletzko, B.; Slikker Jr., W. Maternal smoking duringpregnancy and childhood obesity. Am. J. Epidemiol. 156:954–961; 2002.

[41] Holloway, A. C.; Kellenberger, L. D.; Petrik, J. J. Fetal and neonatal exposure tonicotine disrupts ovarian function and fertility in adult female rats. Endocrine30:213–216; 2006.

[42] Luo, Z. C.; Fraser, W. D.; Julien, P.; Deal, C. L.; Audibert, F.; Smith, G. N.; Xiong, X.;Walker, M. Tracing the origins of “fetal origins” of adult diseases: programming byoxidative stress? Med. Hypotheses 66:38–44; 2006.

[43] Simmons, R. A. Developmental origins of diabetes: the role of oxidative stress. FreeRadic. Biol. Med. 40:917–922; 2006.

[44] Castegna, A.; Aksenov,M.; Aksenova,M.; Thongboonkerd,V.; Klein, J. B.; Pierce,W.M.;Booze, R.; Markesbery,W. R.; Butterfield, D. A. Proteomic identification of oxidativelymodified proteins in Alzheimer's disease brain. Part I: creatine kinase BB, gluta-mine synthase, and ubiquitin carboxy-terminal hydrolase L-1. Free Radic. Biol. Med.33:562–571; 2002.

1925J.E. Bruin et al. / Free Radical Biology & Medicine 44 (2008) 1919–1925

[45] Reinheckel, T.; Korn, S.; Mohring, S.; Augustin, W.; Halangk, W.; Schild, L.Adaptation of protein carbonyl detection to the requirements of proteome analysisdemonstrated for hypoxia/reoxygenation in isolated rat liver mitochondria. Arch.Biochem. Biophys. 376:59–65; 2000.

[46] Dalle-Donne, I.; Rossi, R.; Giustarini, D.; Milzani, A.; Colombo, R. Protein carbonylgroups as biomarkers of oxidative stress. Clin. Chim. Acta 329:23–38; 2003.

[47] Tomohiro, T.; Kumai, T.; Sato, T.; Takeba, Y.; Kobayashi, S.; Kimura, K. Hypertensionaggravates glomerular dysfunction with oxidative stress in a rat model of diabeticnephropathy. Life Sci. 80:1364–1372; 2007.

[48] Droge, W. Free radicals in the physiological control of cell function. Physiol. Rev.82:47–95; 2002.

[49] Wallace,D. C. Amitochondrial paradigmofmetabolic anddegenerativediseases, aging,and cancer: a dawn for evolutionary medicine. Annu. Rev. Genet. 39:359–407; 2005.

[50] Gulbins, E.; Dreschers, S.; Bock, J. Role of mitochondria in apoptosis. Exp. Physiol.88:85–90; 2003.

[51] Bruin, J. E.; Gerstein, H. C.; Morrison, K. M.; Holloway, A. C. Increased pancreaticbeta cell apoptosis following fetal and neonatal exposure to nicotine is mediatedvia the mitochondria. Toxicol. Sci. (Jan. 17 [Electronic publication ahead of print]);in press.

[52] Leahy, J. L. Pathogenesis of type 2 diabetes mellitus. Arch. Med. Res. 36:197–209;2005.

[53] Rhodes, C. J. Type 2 diabetes—a matter of beta-cell life and death? Science307:380–384; 2005.

[54] Ginzel, K. H.; Maritz, G. S.; Marks, D. F.; Neuberger, M.; Pauly, J. R.; Polito, J. R.;Schulte-Hermann, R.; Slotkin, T. A. Critical review: nicotine for the fetus, the infantand the adolescent? J. Health Psychol. 12:215–224; 2007.