impact of hyperlipidemia on plasma protein binding...

Impact of Hyperlipidemia on Plasma Protein Binding andHepatic Drug Transporter and Metabolic Enzyme Regulation ina Rat Model of Gestational Diabetes□S

Gregory J. Anger and Micheline Piquette-MillerDepartment of Pharmaceutical Sciences, Faculty of Pharmacy, University of Toronto, Toronto, Ontario, Canada

Received January 6, 2010; accepted March 25, 2010

ABSTRACTIt is currently unknown whether gestational diabetes mellitus(GDM), a prevalent obstetrical complication, compounds thechanges in drug disposition that occur naturally in pregnancy.Hyperlipidemia occurs in GDM. Using a rat model of GDM, wedetermined whether excess lipids compete with drugs forplasma protein binding. Because lipids activate nuclear recep-tors that regulate drug transporters and metabolic enzymes, weused proteome analysis to determine whether hyperlipidemiaindirectly leads to the dysregulation of these proteins in theliver. GDM was induced on gestational day 6 (GD6) via strep-tozotocin injection. Controls received either vehicle alone orstreptozotocin with subsequent insulin treatment. Liver andplasma were collected on GD20. Glyburide and saquinavir pro-tein binding was determined by ultrafiltration, and an estab-lished solvent method was used for plasma delipidation. Pro-teomics analysis was performed by using isobaric tags for

relative and absolute quantitation methodology with mem-brane-enriched hepatic protein samples. Relative to controls,GDM rat plasma contained more cholesterol and triglycerides.Plasma protein binding of glyburide and saquinavir was de-creased in GDM. Delipidation normalized protein binding inGDM plasma. Proteins linked to lipid metabolism were stronglyaffected in the GDM proteomics data set, with prohyperlipidemicand antihyperlipidemic changes observed, and formed networksthat implicated several nuclear receptors. Up-regulation of drugtransporters and metabolic enzymes was observed (e.g., multi-drug resistance 1/2, CYP2A1, CYP2B9, and CYP2D3). In thisstudy, GDM-induced hyperlipidemia decreased protein bindingand was associated with drug transporter and metabolic enzymeup-regulation in the liver. Both of these findings could change drugdisposition in affected pregnancies, compounding changes asso-ciated with pregnancy itself.

Gestational diabetes mellitus (GDM) is an obstetrical com-plication that affects up to 8% of all pregnancies in NorthAmerica (Lawrence et al., 2008). At present, little is knownabout how the mechanisms that determine drug dispositiondiffer in women with this condition relative to women with

uncomplicated pregnancies. Although it has been establishedthat pregnancy itself decreases the plasma protein binding ofdrugs (Perucca et al., 1981; Notarianni, 1990; Tsen et al.,1999), there have been few attempts to determine whethermaternal diseases such as GDM alter this parameter further.Given the difficulty associated with dosage determination inpregnancy, the potential compounding effects of maternaldisease on protein binding could represent a largely over-looked source of variability.

Protein binding strongly influences a drug’s distributionand/or clearance. With distribution, placental transfer is lim-ited to unbound drugs, so altered protein binding in preg-nancy can significantly affect the materno-fetal transfer of

This work was supported by the Canadian Institutes of Health Research[Grant 57688]. G.J.A. received a Frederick Banting and Charles Best CanadaGraduate Scholarship Doctoral Award from the Canadian Institutes of HealthResearch.

Article, publication date, and citation information can be found athttp://jpet.aspetjournals.org.

doi:10.1124/jpet.110.165639.□S The online version of this article (available at http://jpet.aspetjournals.org)

contains supplemental material.

ABBREVIATIONS: GDM, gestational diabetes mellitus; AAG, �1-acid glycoprotein; ABC, ATP-binding cassette; CAR, constitutive androstanereceptor; CRP, C-reactive protein; CYP, cytochrome P450; FXR, farnesoid X receptor; FFA, free fatty acid; GDn, gestational day n; HNF1, hepaticnuclear factor 1; HDL, high-density lipoprotein; IKB, Ingenuity Knowledge Base; LDL, low-density lipoprotein; MS, mass spectrometry; MS/MS,tandem MS; PPAR, peroxisome proliferator-activated receptor; PXR, pregnane X receptor; PCR, polymerase chain reaction; STZ, streptozotocin;SCX, strong cation exchange; UGT, UDP glucuronosyltransferase; iTRAQ, isobaric tag for relative and absolute quantitation; LC, liquid chroma-tography; IPA, Ingenuity Pathway Analysis; FDFT1, farnesyl-diphosphate farnesyltransferase 1; ACACA, acetyl-CoA carboxylase �; ACAT1,acetyl-CoA acetyltransferase 1; FASN, fatty acid synthase; ACADL, acyl-CoA dehydrogenase, long chain; CD36, CD36 molecule (thrombospondinreceptor); ABCB4, ATP-binding cassette, subfamily B, member 4; ABCD3, ATP-binding cassette, subfamily D, member 3; ABCG2, ATP-bindingcassette, subfamily G, member 2; SOAT2, sterol O-acyltransferase 2; OATP, organic anion transport protein; ALDH1, aldehyde dehydrogenasefamily 1.

0022-3565/10/3341-21–32$20.00THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 334, No. 1Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics 165639/3593595JPET 334:21–32, 2010 Printed in U.S.A.

21

http://jpet.aspetjournals.org/content/suppl/2010/03/26/jpet.110.165639.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on M

ay 18, 2018jpet.aspetjournals.org

Dow

nloaded from

http://jpet.aspetjournals.org/content/suppl/2010/03/26/jpet.110.165639.DC1

http://jpet.aspetjournals.org/

drugs (Nanovskaya et al., 2006). With clearance, the removalof drugs by the liver and kidneys, a process that depends ondrug transporters and metabolic enzymes, often varies withplasma protein binding, because uptake by these organs isalso limited to unbound drugs. Atypical clearance has fetalimplications in the form of changing the absolute amount ofdrug that is available for placental transfer and maternalimplications in the form of diminished therapeutic effect(when increased) or toxicity (when decreased).

Protein binding in GDM is of interest not just becausethere is a paucity of information concerned with it but be-cause there are reports of changes to this parameter in otherforms of diabetes (Ruiz-Cabello and Erill, 1984; Trovik et al.,1992; Arredondo et al., 1999). For example, it has been dem-onstrated that the plasma protein binding of itraconazole inpatients with type 1 and type 2 diabetes is significantly lessthan in healthy volunteers (Arredondo et al., 1999). Somestudies have implicated increased glycosylation of abundantplasma proteins such as albumin and �1-acid glycoprotein(AAG) (Worner et al., 1992; Koyama et al., 1997), but reportson this topic are conflicting (Olsen et al., 1995).

Hyperlipidemia is a common comorbidity among patientswith type 1 or 2 diabetes (Ginsberg, 1996; Betteridge, 2000;Krentz, 2003; Harris et al., 2005) and represents an alterna-tive explanation for altered protein binding in diabetes (Ruiz-Cabello and Erill, 1984; de la Fuente et al., 2002). In a studyusing patients with type 1 diabetes, a significant positivecorrelation between elevated plasma lipids, such as choles-terol and triglycerides, and the urinary excretion of albuminwas observed (Mattock et al., 2001). In addition to increasingthe urinary excretion of this important plasma protein, it isknown that lipids can interact directly with proteins to altertheir capacity for drug binding via competition or allostericmodulation. Even small increases in free fatty acids (FFAs)displace chlorophenoxyisobutyrate from human albumin(Spector et al., 1973). Along these same lines, the binding ofpropranolol to human AAG can be significantly increased ifAAG-bound lipids are removed first (Chauvelot-Moachon etal., 1988). In pregnancy, diabetes exacerbates the naturalrise in plasma lipids that occurs across pregnancy (Couch etal., 1998; Basaran, 2009).

In this study, plasma lipids and protein binding were ex-amined in the streptozotocin (STZ) rat model of GDM. Sim-ilar to the clinical scenario, STZ-induced diabetes is associ-ated with hyperlipidemia (Wasan et al., 1998), and our initialexperiments with nonpregnant rats have demonstrated cor-relations between plasma lipids and the unbound percentageof both acidic and basic drugs after STZ injection (unpub-lished data). We hypothesized that excess plasma lipidswould compete with drugs for protein binding in rats withSTZ-induced GDM.

In addition to assessing plasma lipids and protein binding,insight into the effects of hyperlipidemia on hepatic proteinslinked to drug transport and metabolism was sought by usinga global proteomics approach (8Plex iTRAQ). The regulationof drug transporters and metabolic enzymes in the liver iscontrolled largely by nuclear receptors, such as the farnesoidX receptor (FXR) and pregnane X receptor (PXR), which arethemselves activated by various plasma lipids (Handschinand Meyer, 2005). Dysregulation of hepatic drug transport-ers and metabolic enzymes would be expected to exacerbateany variability in drug disposition introduced by altered

plasma protein binding. We hypothesized that hyperlipid-emia in STZ-induced GDM would indirectly result in hepaticdrug transporter and metabolic enzyme dysregulation.

Our findings advance our understanding of the impact thatGDM-induced hyperlipidemia can have on plasma proteinbinding and the expression of drug transporters and meta-bolic enzymes in the liver.

Materials and MethodsAnimals. Pregnant Sprague-Dawley rats (207–260 g) were pur-

chased from Charles River Laboratories, Inc. (Senneville, QC, Can-ada) and housed individually in a temperature-controlled facility ona 12:12-h light/dark cycle. Rats were given free access to water andstandard chow [Harlan Teklad (Madison, WI) Global Diet 2018].Experiments were approved by the Office of Research Ethics at theUniversity of Toronto and performed in accordance with CanadianCouncil on Animal Care Guidelines.

Animal Treatments/Monitoring and Tissue Collection. Ratswere randomly assigned to one of three groups (n � 5/group): non-treated diabetics (the GDM group), vehicle controls (the vehiclegroup), and insulin-treated diabetics (the insulin-treated group). An-imals in the vehicle and insulin-treated groups are collectively re-ferred to as controls. On gestational day 6 (GD6), diabetes wasinduced by way of a single subcutaneous injection of STZ (PubChemCompound ID 5300; Sigma-Aldrich, Oakville, ON, Canada) in 0.1 Mcitrate buffer (pH 4.5) at a dosage of 45 mg/kg. Vehicle controlanimals received a single subcutaneous injection of citrate buffer onGD6. For all animals, body weight was measured before injections(GD6) and then daily until sacrifice. For animals in the GDM andvehicle groups, blood glucose was measured before injections (GD6)and on GDs 7, 8, 9, 11, 13, 15, 18, and 20. For animals in theinsulin-treated group, blood glucose was measured before STZ injec-tion (GD6) and then daily until sacrifice to determine daily insulindosage requirements. Blood glucose was measured at noon, in anonfasted state, by using a commercially available glucometer (Free-style Freedom Meter, Abbott Laboratories, Alameda, CA) with bloodobtained via tail prick. When values exceeded the range of thisglucometer, a commercially available enzymatic glucose assay wasused according to the manufacturer’s instructions (Autokit Glucose;Wako Chemicals USA, Inc., Richmond, VA). Mild hyperglycemia wasdefined as blood glucose concentration at 10 to 14 mM, and diabeteswas defined as blood glucose concentration at �14 mM. For diabeticrats receiving insulin treatment, treatment was initiated once dia-betes was confirmed and consisted of an initial subcutaneous injec-tion of 8 to 10 U/kg Humulin-L (human biosynthetic intermediate-acting insulin; Eli Lilly Canada Inc., Toronto, ON, Canada) andsubsequent daily injections as required. For all animals, water in-take, urinary output, feed intake, and fecal output were monitored inmetabolic cages on GD18. After measurement of blood glucose andbody weight on GD20, animals were sacrificed to collect maternalblood (cardiac puncture) and liver. Aliquots of whole blood weretransferred to heparin- and thrombin-coated BD Vacutainer tubes(BD Canada, Mississauga, ON, Canada) and spun to collect plasmaand serum, respectively. Samples were snap-frozen in liquid nitro-gen and stored at �80°C until use.

Blood Chemistry. Determination of total cholesterol, high-den-sity lipoprotein (HDL)/low-density lipoprotein (LDL) cholesterol, tri-glyceride, FFA, and albumin concentrations in serum was out-sourced to the Banting and Best Diabetes Centre, University ofToronto. Total cholesterol, HDL/LDL cholesterol, triglyceride, andalbumin concentrations were analyzed by using an AutoAnalyzerand commercially available reagents (Roche Diagnostics, Laval, QC,Canada), whereas FFA was analyzed by using a commercially avail-able enzyme-linked immunosorbent assay kit (Wako ChemicalsUSA, Inc.). Determination of alanine transaminase concentrations inplasma was outsourced to Vita-Tech Laboratories (Markham, ON,

22 Anger and Piquette-Miller

at ASPE

T Journals on M


Dow

nloaded from


Canada), where an AutoAnalyzer and commercially available re-agents (Roche Diagnostics) were used. C-reactive protein (CRP) andAAG concentrations were determined in-house by using commer-cially available, rat-specific enzyme-linked immunosorbent assaykits (BD Biosciences, San Jose, CA and GenWay, San Diego, CA,respectively), according to the manufacturers’ instructions.

Protein Binding Assays. Glyburide (PubChem compound ID3488) and saquinavir (PubChem compound ID 60787) protein bind-ing in rat plasma was determined in vitro by ultrafiltration using theCentrifree micropartition system (Millipore Canada, Ltd., Etobicoke,ON, Canada). Glyburide is an acidic drug that binds to albumin inplasma, and saquinavir is a basic drug that binds to AAG in plasma.Aliquots of [3H]glyburide (PerkinElmer Life and Analytical Sciences,Waltham, MA) or [3H]saquinavir (American Radiolabeled Chemi-cals, St. Louis, MO), dissolved in ethanol, were placed in microcen-trifuge tubes and dried under nitrogen gas. Dried [3H]glyburide or[3H]saquinavir was then reconstituted with drug-free plasma fromeach of this study’s three groups (n � 5/group) to obtain a finalglyburide or saquinavir concentration of 30 ng/ml or 0.5 �g/ml,respectively, in a final volume of 200 �l. Samples were then incu-bated at 36°C for 1 h with gentle agitation every 20 min. At 1 h,150-�l aliquots were transferred to ultrafiltration tubes and centri-fuged at 1000g for 30 min. After centrifugation, aliquots (25 �l) offiltered and unfiltered sample were supplemented with 4 ml of Pico-Fluor 40 (PerkinElmer Life and Analytical Sciences) and subjected toscintillation counting. Percentage unbound was determined by usingthe following formula: percentage unbound � (filtered count/unfil-tered count) � 100. Preliminary experiments demonstrated constantglyburide and saquinavir protein binding across a wide range ofconcentrations. The final concentrations selected for plasma proteinbinding experiments reflect values within therapeutic ranges. Todirectly determine the capacity for plasma lipids to alter proteinbinding, total lipid content was removed from drug-free nontreateddiabetic, insulin-treated diabetic, and vehicle control plasma sam-ples (n � 5/group) using the butanol/diisopropyl ether extractionmethod of Cham and Knowles (1976). This delipidation procedureremoves 100% of plasma lipids without affecting the concentrationsof other plasma constituents (Cham and Knowles, 1976). Delipidatedsamples were then subjected to the glyburide protein binding proce-dure as described. Reported values are the result of triplicate exper-iments and are reported as mean � S.D.

Proteomics: Sample Preparation and Isobaric Mass Tag-ging. A proteomics approach (8Plex iTRAQ) was used to characterizethe expression of hepatic proteins involved in lipid metabolismand/or drug (Supplemental Fig. 1). Our experimental design con-sisted of two experiments: a first experiment comparing four non-treated diabetic samples to three vehicle control samples, and asecond experiment comparing four insulin-treated diabetic sampleswith three vehicle control samples. To eliminate the contribution ofblood proteins in these experiments, a proteomics-specific cohort ofrats was randomly assigned to the treatments described above (n �3–4/group) but subjected, on GD20, to hepatic saline perfusion beforeliver collection (Mehta et al., 2009). Protein homogenates, obtainedfrom perfused livers, underwent a membrane protein enrichmentprocedure as described previously (Molloy, 2008). For isobaric masstagging, an 8Plex iTRAQ kit from Applied Biosystems (Scoresby,Victoria, Australia) was used according to the manufacturer’s in-structions. iTRAQ-labeled samples were combined (one combinationper experiment) and then dried by SpeedVac (Thermo Fisher Scien-tific, Scoresby, Victoria, Australia) before storage at �20°C to awaitclean-up and fractionation.

Proteomics: High-Performance Liquid Chromatography-Tandem Mass Spectrometry. Sample clean-up and fractionationwas performed by strong cation exchange (SCX) chromatography onan Agilent Technologies (Forest Hill, Victoria, Australia) 1100 qua-ternary high-performance liquid chromatography (LC) pump with aPolySulfoethyl A column (2.1 mm � 200 mm, 5-�m particle size,300-Å pore) from PolyLC, Inc. (Columbia, MD). Two buffers (A and

B) were used for SCX. Buffer A was 5 mM phosphate and 25%acetonitrile (pH 2.7), and buffer B was 5 mM phosphate, 350 mMpotassium chloride and 25% acetonitrile (pH 2.7). Combined, iTRAQ-labeled samples were resuspended in 14 ml of buffer A. This solutionwas filtered, and 7 ml was then loaded onto the SCX column. Aftersample loading and washing with buffer A, the concentration ofbuffer B was increased from 10 to 45% in 70 min and then increasedto 100% for 10 min at 300 �l/min. For each experiment, a total of 23SCX fractions were collected and dried by SpeedVac before storage at�20°C to await mass spectrometry (MS).

Reverse-phase nanoLC electrospray ionization tandem mass spec-trometry (MS/MS) was conducted on a QStar Elite MS/MS system(Applied Biosystems, Foster City, CA). SCX fractions were resus-pended in 60 �l of a loading/desalting solution consisting of 0.1%trifluoroacetic acid and 2% acetonitrile. Forty microliters of thissolution was loaded on a reverse-phase peptide Cap Trap (MichromBioresources, Inc., Auburn, CA) desalted with 0.1% formic acid and2% acetonitrile at 8 �l/min for 13 min. The trap was then switchedonline with a ProteoCol C18 trap cartridge (150 �m � 100 mm, 3-�mparticle size, 300-Å pore) from SGE Analytical Science (Ringwood,Victoria, Australia). Channel 1 contained desalting solution, channel2A contained 0.1% formic acid, and channel 2B contained 0.1%formic acid and 90% acetonitrile. The concentration of solution inchannel 2 was increased from 5 to 95% in 95 min in three lineargradient steps to elute peptides. After elution, the column wascleaned by increasing the concentration of channel 2B from 0 to 100%for 20 min followed by equilibration with 95% channel 1 and 5%channel 2B for 7 min before injection of the next SCX fraction.

Reverse-phase nanoLC eluent was subjected to positive ion nano-flow electrospray analysis in an information-dependent acquisitionmode (Applied Biosystems). In this mode, a time-of-flight MS surveyscan was acquired (m/z 370–1600, 0.5 s) with the three most intensemultiply charged ions (counts �50) in the survey scan sequentiallysubjected to MS/MS analysis. Spectra were accumulated for 2 s inthe range of m/z 100 to 1600 with a modified Q2 transition settingfavoring low mass ions so that iTRAQ ion (113, 114, 115, 116, 117,118, 119, and 121) intensities were enhanced.

Proteomics: Database Processing and Analysis. Experimen-tal nanoLC electrospray ionization MS/MS data were submitted toProteinPilot (Applied Biosystems, version 2.0) for database process-ing. The Paragon Algorithm was used in a thorough ID search effortwith biological modifications selected in the ID focus box. Softwarecorrection factors, provided in the iTRAQ kit, were entered in theiTRAQ isotope correction factors table. The detected protein thresh-old (unused ProtScore) was set as larger than 1.3, which correspondsto an identification confidence of 95%. The unused ProtScore esti-mates the confidence associated with identifications and uses onlypeptide evidence that is not better explained by a higher-rankingprotein. Significantly altered proteins (see Statistics), up-regulatedor down-regulated more than 10%, were imported into IngenuityPathway Analysis (IPA) software (Ingenuity Systems, Inc., http://www.ingenuity.com). IPA software was used to identify protein net-works and generate lists of proteins associated with specific molec-ular and cellular functions. IPA software was also used tographically illustrate the molecular relationships between proteinson our lists and proteins stored in the Ingenuity Knowledge Base(IKB). In those figures, molecules are represented as nodes, and thebiological relationships between nodes are represented as connectinglines; lines are supported by at least one reference from IKB, whichis based on the literature or canonical information stored in the IKB.IPA software was then used to compare all altered proteins in ourexperiments to tox lists in the IKB. Tox lists are lists of moleculesthat are known to be involved in particular aspects of toxicity, andour focus was on lists that addressed nuclear receptor activation.

Characterization of Hepatic Mdr1 Expression. In rodents,two genes encode the Mdr1 protein: Mdr1a (Abcb1a) and Mdr1b(Abcb1b). In this study, Mdr1a and Mdr1b mRNA levels and Mdr1protein levels were determined in liver. For mRNA levels, total RNA

Hyperlipidemia in a Rat Model of Gestational Diabetes 23

at ASPE

T Journals on M


Dow

nloaded from


was extracted from tissue by using the TRIzol method (Invitrogen,Carlsbad, CA) and then reverse-transcribed to cDNA via the FirstStrand cDNA synthesis kit (Fermentas, Burlington, ON, Canada),according to the manufacturer’s instructions. mRNA levels weredetermined by quantitative reverse transcriptase-polymerase chainreaction (PCR) using LightCycler technology with SYBR Green Ifluorescence detection (Roche Diagnostics). PCR oligonucleotideswere synthesized at The Hospital for Sick Children, DNA SynthesisCentre, Toronto, ON, Canada. Their sequences have been publishedpreviously (Petrovic et al., 2008). mRNA levels were normalized to18S rRNA and expressed as a gene/18S ratio.

For Mdr1 protein levels, tissue samples were first homogenized inRIPA buffer (Cell Signaling Technologies, Pickering, ON, Canada),containing freshly added phenylmethylsulfonyl fluoride and 1� pro-tease inhibitor cocktail (Sigma-Aldrich), using a motorized pestle.Homogenates were then incubated at 4°C for 2 h on a rocker andthen centrifuged at 14,000 rpm for 20 min. For each sample, thesupernatant was isolated and subjected to a Bradford assay to de-termine total protein concentration. Samples containing 60 �g ofprotein in Laemmli sample buffer were heated at 36°C for 20 minand then separated via 10% SDS-polyacrylamide gel electrophoresisand transferred to polyvinylidene fluoride membranes (Bio-Rad Lab-oratories Canada Ltd., Mississauga, ONT, Canada). Membraneswere blocked in 5% fat-free milk powder at 4°C overnight on a rockerand incubated with a mixture of monoclonal anti-Mdr1 (1:500, 1mg/ml; Abcam Inc., Cambridge, MA) and anti-�-actin (1:20,000, 2mg/ml; Sigma-Aldrich) antibodies at room temperature for 3 h. Aftera series of washes, membranes were incubated with a peroxidase-labeled secondary antibody for 1 h at room temperature (1:3000; goatanti-mouse from Jackson ImmunoResearch Laboratories, Inc., WestGrove, PA). Immunoreactive Mdr1 and �-actin proteins were simul-taneously detected in membranes by using ECL Plus (GE Health-care, Little Chalfont, Buckinghamshire, UK) and scanned by usingan Alpha Innotech (San Leandro, CA) FluorChem imaging system.Using AlphaEaseFC (version 6.0) software, also from Alpha Inno-tech, the optical density of each band was determined. Mdr1 opticaldensities were normalized to �-actin.

Statistics. Data were analyzed with Prism 4 for Macintosh(GraphPad Software, Inc., San Diego, CA). For comparisons of bloodglucose and body weight, two-way analysis of variance followed by

Bonferroni post-tests were performed. These same tests were used toexamine any potential differences between this study’s two cohorts,with respect to blood glucose and body weight. Protein binding ofdrugs (percentage unbound), plasma concentrations (e.g., individuallipids, albumin, etc.), hepatic Mdr1 expression data, and metaboliccage variables were analyzed by one-way analysis of variance fol-lowed by Bonferroni post-tests. Linear regression was used to deter-mine the relationship between the protein binding of glyburide andsaquinavir. Paired t tests were used to statistically evaluate theeffects of plasma delipidation on the protein binding of glyburide. Forproteomics data, ProteinPilot was used to generate control-standard-ized “fold changes” for each sample’s identified proteins. By stan-dardizing to controls, the mean of control values was made to rep-resent 1 � S.D., and noncontrol values were made, consequently, todistribute around this value. Statistically significant deviations fromcontrol values were identified by using t tests. Levels of significancefor all statistical analyses were set at or below � � 0.05, indicated asfollows: �/#, p 0.05; ��/##, p 0.01; ��/###, p 0.001. All resultsare presented as mean � S.D.

ResultsSTZ-Injected Pregnant Rats Exhibited a Variety of

Characteristics That Were Consistent with Unman-aged or Poorly Managed GDM. Induction of GDM in STZ-injected rats was determined on the basis of their bloodglucose concentrations as measured on GDs 7–9, 11, 13, 15,18, and 20 (Supplemental Fig. 2A). All of the STZ-injectedrats developed mild hyperglycemia within 24 h and thendeveloped GDM (� 14 mM blood glucose concentration) byGD9. All animals in the GDM group were, therefore, diabeticfor 12 to 13 days because sacrifice occurred on GD20. For theinsulin-treated group, insulin treatment was initiated whenblood glucose concentrations first exceeded 14 mM and wasefficacious within hours. Blood glucose concentrations in theGDM group, relative to controls, were higher at every timepoint after GD9 (p 0.001; Table 1). Blood glucose concen-trations in the insulin-treated group fluctuated but were

TABLE 1The effect of STZ-induced GDM, with and without insulin treatment, on various physiological and biochemical characteristicsAll results are presented as mean � S.D. of five rats/group. �, Significantly different from vehicle controls. #, Significantly different from GDM. �/#, P 0.05; ��/##, P 0.01;��/###, P 0.001.

Characteristic Vehicle GDM Insulin-Treated

Blood glucose and body weightBlood glucose (mM, GD6) 6.2 � 0.8 6.6 � 0.9 6.5 � 1.8Blood glucose (mM, GD11) 5.8 � 0.7 24.7 � 3*** 8.4 � 3.2###

Blood glucose (mM, GD18) 5.1 � 0.5 26 � 2*** 5.5 � 2.3###

Blood glucose (mM, GD20) 5.4 � 1 26 � 2.6*** 5.7 � 1.9###

Body weight (g, GD6) 241 � 23 242 � 28 246 � 22Body weight (% baseline, GD11) 114 � 2 104 � 2*** 112 � 2###

Body weight (% baseline, GD18) 138 � 7 124 � 5*** 136 � 8###

Body weight (% baseline, GD20) 150 � 8 136 � 7** 150 � 8##

Blood chemistry (GD20)Total cholesterol (mM) 1.9 � 0.4 5 � 1.4** 2.2 � 0.6#

HDL cholesterol (mM) 0.8 � 0.2 0.7 � 0.2 1.2 � 0.1#

Triglycerides (mM) 3.65 � 1.54 19.15 � 9.08* 2.12 � 0.74#

Free fatty acids (mEq/l) 0.9 � 0.3 2.5 � 1.3 0.8 � 0.1AAG (�g/ml) 25.2 � 4.4 39.5 � 5.8* 33.8 � 8.5Albumin (g/l) 35.7 � 3.4 21.8 � 5.3** 39 � 0.1##

ALT (I.U./l) 54.8 � 6.6 104.8 � 20.4** 74.8 � 22.7CRP (�g/ml) 318.2 � 33.5 305.2 � 45.3 296.2 � 58.1

Metabolic cage (GD18)Water intake (ml/kg/day) 152 � 5 800 � 201** 189 � 133##

Urinary output (ml/kg/day) 78 � 23 671 � 214** 148 � 85#

Feed intake (g/kg/day) 81.1 � 28.2 177.6 � 61.9 77.4 � 21.5Fecal output (g/kg/day) 20 � 7.5 95.6 � 33.2** 20.9 � 0.2#

Urinalysis (GD18)Total protein output (mg/kg/day) 11.3 � 6.5 35.2 � 43.4 8.2 � 1.1


at ASPE

T Journals on M


Dow

nloaded from


different from values obtained in the vehicle group only onGD13.

As expected during gestation, all animals gained weightacross the study duration; however, animals with GDM werelighter from GD8 until sacrifice (p 0.01–0.001; Table 1 andSupplemental Fig. 2B). On GD20, the body weight of animalsin the GDM group was 136 � 7% of baseline values versusthe 150 � 8% increase observed in the vehicle and insulin-treated groups (p 0.01).

Animals in the GDM group exhibited other physiologicaland biochemical characteristics consistent with unmanaged,or poorly managed, GDM (Table 1). Hyperlipidemia was ev-ident in the GDM group as plasma cholesterol (total) andtriglyceride concentrations were profoundly elevated. FFAconcentrations were also elevated, but this finding was notstatistically significant (p � 0.06). In addition, animals in theGDM group exhibited obvious polydipsia and polyuria onGD18. While in metabolic cages, relative to the vehicle andinsulin-treated groups, animals in the GDM group exhibiteda 5.2- and 4.2-fold increase in water consumption, respec-tively, and a 8.6- and 4.6-fold increase in urination, respec-tively. Animals in the GDM group also exhibited hyperpha-gia with an associated increase in fecal output. Relative tocontrols, animals in the GDM group consumed over twice asmuch food and produced nearly five times as many feces.Plasma from animals in the GDM group contained higherlevels of AAG and lower levels of albumin, compared with thevehicle group. Albumin and AAG concentrations in the insu-lin-treated group were slightly elevated, but this finding wasnot statistically significant. Compared with the vehicle andinsulin-treated groups, alanine transaminase was found tobe elevated in the GDM group, whereas CRP values werefound to be similar (p � 0.05).

Decreased Plasma Protein Binding of Acidic and BasicDrugs Was Observed in GDM. The plasma protein bindingof glyburide, an acidic drug that binds to albumin, and sa-

quinavir, a basic drug that binds to AAG, was examined.Unbound glyburide was found to be 13.4 � 2.6% in plasmafrom animals with GDM, which was higher than controls:5.2 � 2.6% in the vehicle group and 3.8 � 0.9% in theinsulin-treated group (p 0.0001; Fig. 1A). For saquinavir,12.2 � 1.1% was unbound in plasma from animals withGDM, which was again higher than controls: 9.7 � 1.2% inthe vehicle group and 8.9 � 1.1% in the insulin-treated group(p 0.01; Fig. 1B). Although the unbound percentages in theinsulin-treated group were lower than those in the vehiclegroup, the magnitude of these changes was quite modest.Linear regression revealed a positive relationship betweenthe plasma protein binding of glyburide and saquinavir (r2 �0.85; p 0.0001; Fig. 1C).

Evidence for Hyperlipidemia as a Mechanism forDecreased Plasma Protein Binding in GDM. To deter-mine the contribution of plasma lipids to the altered proteinbinding observed in the GDM group, plasma samples weredelipidated before analysis of plasma protein binding capac-ity. As illustrated in Fig. 1D, delipidation of plasma from theGDM group dramatically reduced the percentage of unboundglyburide from 13.4 � 2.6 to 6.1 � 0.3% (p 0.01). Postde-lipidation protein binding in the GDM group was not statis-tically different from postdelipidation protein binding in thevehicle group (p � 0.05). Delipidation had no effect on theprotein binding of glyburide in the vehicle or insulin-treatedgroups (p � 0.05).

Alterations in the Expression of Hepatic MembraneProteins in GDM Were Largely Linked to Lipid Metab-olism and Were Both Prohyperlipidemic and Antihy-perlipidemic. Proteomic analysis consisted of two iTRAQexperiments: experiment 1 compared the GDM and vehiclegroups, and experiment 2 compared the insulin-treated andvehicle groups. Experiment 1 detected 625 hepatic proteins,142 of which were significantly altered (Fig. 2A), and exper-iment 2 detected 607, 125 of which were significantly altered

Fig. 1. Plasma protein binding character-istics. A and B, the plasma protein bind-ing of glyburide (GLIB) (A) and saquina-vir (SQV) (B) was determined in the GDMand control groups and expressed as per-centage unbound. C, linear regressionwas used to determine the relationshipbetween the binding of glyburide and sa-quinavir. D, plasma samples were delipi-dated and then subjected to the glyburideprotein binding procedure. All results arepresented as mean � S.D. of five rats/group. �, Significantly different from ve-hicle controls. #, Significantly differentfrom GDM. �/#, p 0.05; ��/##, p 0.01;��/###, p 0.001.


at ASPE

T Journals on M


Dow

nloaded from


(Fig. 2B). Comparison of the two iTRAQ datasets revealed atotal of 476 proteins that were common: meaning that 149proteins were detected only in the GDM versus vehicle compar-ison, and 131 proteins were detected only in the insulin-treatedversus vehicle comparison. The raw data associated with theproteomics component of this study may be downloaded fromthe ProteomeCommons.org Tranche network by using the fol-lowing hash: T4stJnbow8wF2ifkOiHs3zEw5QOaxzU6-fUUL98P2mZ8603m4dj2YU/EXV/kB4kGcvw3DJ3TRmp-V0xSp/bfzggyNlgoAAAAAAAApRwA.

Members of the ATP-binding cassette (ABC) and solutecarrier organic anion transporters, cytochrome P450 (CYP)enzymes and UDP glucuronosyltransferases (UGTs) werepresent in both data sets; however, owing probably to theextensive sharing of peptide sequences, we positively identi-fied only a modest fraction of drug transporters and meta-bolic enzymes known to be present in rat liver.

Altered lipid metabolism appeared as a function in two ofthe top five molecular networks that were identified in GDMby IPA software. Of the 142 proteins altered in GDM, 22.5%(32) were identified as having a molecular/cellular functionin lipid metabolism (Table 2). Of these proteins, 72% (23)were normalized by insulin treatment, 22% (7) lacked com-

parison data in the insulin treatment group, and 6% (2) weresimilarly altered.

A review of the literature concerning the proteins listedin Table 2 indicated that alterations in the GDM groupwere both prohyperlipidemic and antihyperlipidemic. Forthe purposes of presenting and discussing results, alter-ations were considered prohyperlipidemic if they would beexpected to increase the hepatic synthesis of lipids orpromote their retention in the blood compartment. Con-versely, alterations were considered antihyperlipidemic ifthey would be expected to decrease the hepatic synthesis oflipids or promote their retention in nonblood compart-ments (e.g., within hepatocytes). The up-regulation ofFDFT1 and CYP51A1 in GDM are examples of prohyper-lipidemic alterations. Examples of antihyperlipidemic al-terations in GDM include the down-regulation of ACACA,ACAT1, and FASN and the up-regulation of ACADL. Theexpression of lipid transporting proteins was also alteredin GDM. CD36 was down-regulated, whereas ABCB4(Mdr2) and solute carrier organic anions 27A2 and 27A5were up-regulated.

It is noteworthy that lipid metabolism did not appear as afunction in any of the major molecular networks that wereidentified in the insulin-treated group. Despite this, alter-ations in 14 proteins involved in lipid metabolism were foundin the insulin-treated group’s proteomics data set (Table 3).As with GDM, alterations in the insulin-treated group wereboth prohyperlipidemic and antihyperlipidemic. The down-regulation of ABCD3, ABCG2, and CD36 in this group areexamples of prohyperlipidemic alterations. Examples ofantihyperlipidemic alterations after insulin treatment in-clude the finding that emopamil binding protein andSOAT2 were down-regulated and the finding that LDLRwas up-regulated.

Several proteins involved in drug transport and metabolismwere up-regulated in GDM, including: ABCB4, CYP4A11,CYP4A14, CYP51A1, and UGT1A1. In the insulin-treated group,ABCD3, ABCG2, and CYP2B6 were down-regulated, whereasCYP4F8 and xanthine dehydrogenase were up-regulated.OATP1A1 was down-regulated in both groups.

To investigate the relationships specifically between pro-teins involved in lipid metabolism, all altered proteins asso-ciated with this function in experiments 1 and 2 were isolatedand individually subjected to IPA analysis. Figures 3 and 4depict molecular networks for these proteins, illustrating theconnections between proteins quantified in our experimentsand a myriad of proteins implicated on the basis of IKB data.In Figs. 3 and 4, one feature that stands out is the presenceof network “hubs” that are comprised of nuclear receptors(these hubs appear in blue).

Evidence for Nuclear Receptor Activation Contrib-uting to Drug Transporter and Metabolic EnzymeUp-Regulation. The networks depicted in Figs. 3 and 4highlight the involvement of nuclear receptors with rolesin lipid metabolism and the regulation of drug transport-ers and metabolic enzymes. These networks implicatedinvolvement of FXR, liver X receptor (LXR), nuclear factor�B, peroxisome proliferator-activated receptors (PPARs),and thyroid hormone receptor in GDM-mediated alter-ations. On the other hand, network analysis implicatedhepatic nuclear factor 1 � and � (HNF1A/HNF1B), nuclearfactor-�B, and PPAR� in alterations that were observed in

Fig. 2. Volcano plots depicting the magnitude and statistical significanceof hepatic protein alterations in iTRAQ experiments. Negative log pvalues for all 625 proteins identified in the GDM versus vehicle groupcomparison (A) and all 607 proteins identified in the insulin-treatedversus vehicle group comparison (B) were plotted against log fold change.Black horizontal lines correspond with � � 0.05, meaning all proteinsfound above the lines are statistically different from controls.


at ASPE

T Journals on M


Dow

nloaded from


the insulin-treated group. Given the apparent activity ofthese nuclear receptors, at least within the subset of ourdata concerned with lipid metabolism, we decided to com-pare all altered proteins observed in both iTRAQ experi-ments to nuclear receptor activation tox lists in the IKB

(Table 4). At present, tox lists are available for the activa-tion of FXR, LXR, PPARs, and PXR as well as the consti-tutive androstane receptor (CAR) and nuclear factor ery-throid 2-related factor 2. With the exception of CAR,relative to animals in the insulin-treated group, animals in

TABLE 2IPA-identified proteins related to lipid metabolism in STZ-induced GDMControl-standardized fold changes were generated for proteins linked to lipid metabolism. Differences of less than 10% were disregarded, and statistically significantdeviations from control values were identified by using t tests. �, P 0.05; ��, P 0.01; ��, P 0.001. Comparison values in the insulin-treated group were designated asno change (NC) when they were either not statistically different from those of controls or differed from controls by less than 10%.

UniProt Symbol Entrez Gene NameSTZ STZ Insulin

Coverage Fold Change Coverage Fold Change

% %

Q08201 ABCB4 ATP-binding cassette, subfamily B, member 4 (Mdr2) 52.4 1.12** — —P11497 ACACA Acetyl-CoA carboxylase � 38.5 �1.15*** 33.8 NCP15650 ACADL Acyl-CoA dehydrogenase, long chain 61.9 1.21*** 37.9 NCP17764 ACAT1 Acetyl-CoA acetyltransferase 1 19.3 �1.18** — —P13601 ALDH1A7 Aldehyde dehydrogenase family 1, subfamily A7 45.1 1.18* — —P04639 APOA1 Apolipoprotein A-I 51 1.17* 41.3 NCP10719 ATP5B ATP synthase, H transporting, mitochondria F1 complex, �

polypeptide45.9 �1.15* 41.4 NC

P21571 ATP5J ATP synthase, H transporting, mitochondria F0 complex,subunit F6

94.4 �1.35* 61.1 NC

P01026 C3 Complement component 3 47.1 �1.12* 45.2 NCQ07969 CD36 CD36 molecule (thrombospondin receptor) 41.1 �1.25** 31.8 �1.19*P20070 CYB5R3 Cytochrome B5 reductase 3 81.1 �1.18* 96 NCP08516 CYP4A11 Cytochrome P450, family 4, subfamily A, polypeptide 11 45.8 1.28* — —P20817 CYP4A14 Cytochrome P450, family 4, subfamily A, polypeptide14 75.7 1.38* 58.2 NCQ64654 CYP51A1 Cytochrome P450, family 51, subfamily A, polypeptide 1 65.2 1.2* 52.7 NCP23965 DCI Dodecenoyl-CoA delta isomerase 52.6 1.34** — —Q64591 DECR1 2,4-dienoyl CoA reductase 1, mitochondrial 50.7 1.12* 32.8 NCP12785 FASN Fatty acid synthase 45.5 �1.22*** 45.7 NCQ02769 FDFT1 Farnesyl-diphosphate farnesyltransferase 1 46.2 1.12** 44.7 NCP05369 FDPS Farnesyl diphosphate synthase 35.1 �1.13* 39.4 NCP43428 G6PC Glucose-6-phosphatase, catalytic subunit 26.9 1.68* 17.4 NCQ64232 GPSN2 Glycoprotein, synaptic 2 54.2 �1.17** 60.4 NCP06762 HMOX1 Heme oxygenase (decycling) 1 49.5 �1.19* — —P46720 OATP1A1 Solute carrier organic anion transporter family, member 1A1 27.2 �1.14* 28.4 �1.17*P02696 RBP1 Retinol binding protein 1, cellular 29.6 �1.16** 48.1 NCQ8VHE9 RETSAT Retinol saturase (all-trans-retinol 12,14-reductase) 69.3 1.29* 66.3 NCP61589 RHOA Ras homolog gene family, member A 39.9 1.13* 43 NCP17475 SERPINA1 Serpin peptidase inhibitor, clade A, member 1 46.7 1.16* 25.1 NCP97524 SLC27A2 Solute carrier family 27 (fatty acid transporter), member 2 70.3 1.21*** 67.6 NCQ9ES38 SLC27A5 Solute carrier family 27 (fatty acid transporter), member 5 77.7 1.37** 52.2 NCP24008 SRD5A1 Steroid-5-�-reductase, � polypeptide 1 35.5 �1.23*** 57.9 NCP61943 ST3GAL6 ST3 �-galactoside �-2,3-sialyltransferase 6 19.9 �1.3* — —Q64550 UGT1A1 UDP glucuronosyltransferase 1 family, polypeptide A1 63.2 1.3* 57.2 NC

TABLE 3IPA-identified proteins related to lipid metabolism in insulin-treated STZ-induced GDMControl-standardized fold changes were generated for proteins linked to lipid metabolism. Differences of less than 10% were disregarded, and statistically significantdeviations from control values were identified by using T tests. �, P 0.05; ��, P 0.01; ��, P 0.001. Comparison values in the GDM group were designated as no change(NC) when they were either not statistically different from those of controls or differed from controls by less than 10%.

UniProt Symbol Entrez Gene NameSTZ STZ Insulin

Coverage Fold Change Coverage Fold Change

% %

P16970 ABCD3 ATP-binding cassette, subfamily D, member 3 56.9 NC 61.6 �1.13*Q80W57 ABCG2 ATP-binding cassette, subfamily G, member 2 — — 33.8 �1.16*Q07969 CD36 CD36 molecule (thrombospondin receptor) 41.1 �1.25** 31.8 �1.19*P70500 CDIPT CDP-diacylglycerol—inositol 3-phophatidyltransferase — — 49.8 �1.19**P04167 CYP2B6 Cytochrome P450, family 2, subfamily B, polypeptide 6 51.3 NC 40.3 �1.11*P51869 CYP4F2 Cytochrome P450, family 4, subfamily F, polypeptide 2 56.3 NC 47.3 1.14**Q9JJ46 EBP Emopamil binding protein (sterol isomerase) — — 27.8 �1.25**P35952 LDLR Low density lipoprotein receptor 21.6 NC 22 1.23*Q6AY20 M6PR Mannose 6-phosphate receptor (cation-dependent) 53.2 NC 41.4 1.19*P46720 OATP1A1 Solute carrier organic anion transporter family, member 1A1 27.2 �1.14* 28.4 �1.17*Q6RUV5 RAC1 Ras-related C3 botulinum toxin substrate 1 (rho family,

small GTP binding protein Rac1)50 NC 24 1.27*

Q9R0C9 SIGMAR1 Sigma non-opioid intracellular receptor 1 — — 48 �1.17**Q7TQM4 SOAT2 Sterol O-acyltransferase 2 (ACAT2) 37.4 NC 28.8 �1.2*P22985 XDH Xanthine dehydrogenase 27.1 NC 23.1 1.14**


at ASPE

T Journals on M


Dow

nloaded from


the GDM group consistently exhibited a higher number ofalterations to proteins associated with the activation ofthese nuclear receptors (22 unique proteins versus 10unique proteins). As predicted, many of the specific pro-teins that overlapped with nuclear receptor activation toxlists were transporters, CYPs, or other proteins of rele-vance to drug transport and metabolism. It is important tonote that the majority of proteins presented in Table 4were identified in both iTRAQ experiments.

Table 4 also includes data from two additional lists:1) CYPs with xenobiotic substrates and 2) xenobiotic me-tabolism. Including these additional lists revealed a hand-ful of differentially expressed proteins that did not appearin lipid metabolism networks or the nuclear receptor acti-vation tox lists. For the GDM group, the additional listsidentified CYP2A1, CYP2B9, and CYP2D3 as xenobioticmetabolizing enzymes. All three proteins were up-regu-lated in GDM. For the insulin-treated group, the addi-tional lists identified ALDH1L1, glutathione S-transferase

A4 and Mu 1 (GSTA4 and GSTM1) as well as UGT1A4 asxenobiotic metabolizing enzymes. UGT1A4 was down-reg-ulated, whereas ALDH1L1, GSTA4, and GSTM1 were up-regulated.

Because our evidence suggested activation of FXR andPXR via the up-regulation of their target genes (e.g.,Mdr2), the FXR/PXR “target gene” Mdr1 was examined atthe level of mRNA and protein. Mdr1 was absent from bothiTRAQ data sets, and this supplementary analysis allowedus to further test our hypothesis that hyperlipidemiawould have indirect effects on drug transporters. As Fig. 5illustrates, Mdr1b/18S standard ratios were higher inGDM at 0.26 � 0.15 compared with 0.01 � 0.004 in thevehicle group and 0.03 � 0.02 in the insulin-treated group(p 0.01). Mdr1/�-actin standard ratios were also higherin GDM at 0.41 � 0.1 compared with 0.23 � 0.07 in thevehicle group and 0.22 � 0.06 in the insulin-treated group(p 0.01). All groups had comparable Mdr1a/18S stan-dard ratios (p � 0.05).

Fig. 3. Lipid metabolism network in the GDM group. IPA software was used to graphically illustrate the molecular relationships between alteredproteins that were linked to lipid metabolism and proteins stored in the IKB. Molecules are represented as nodes, and the biological relationshipsbetween nodes are represented as connecting lines. Molecules are labeled with their Entrez gene symbols (see Tables 2 and 3). Up-regulated proteinsare red, down-regulated proteins are green, nuclear receptors are blue, and proteins inserted by IPA software are white.


at ASPE

T Journals on M


Dow

nloaded from


TABLE 4Comparisons of proteins that were altered in iTRAQ experiments to tox lists in the Ingenuity Knowledge BaseData sets were examined by IPA software for overlap with tox list proteins. Tox lists are lists of molecules that are known to be involved in particular aspects of toxicity, andour focus was on lists that addressed nuclear receptor activation and xenobiotic metabolism. Underlined entries were not common to the datasets of both experiments.

Tox List ID (Total number of listmembers)

STZ STZ Insulin

Up-Regulated Down-Regulated Up-Regulated Down-Regulated

CAR/RXR activation (28) UGT1A1 — — CYP2B6FXR/RXR activation (66) ABCB4, APOA1, G6PC,

SLC27A5FASN — —

LXR/RXR activation (55) APOA1 ACACA, CD36,FASN

LDLR CD36

Nrf2 activation (148) CYP2A1, CYP4A11,CYP4A14

HMOX1 GSTA4, GSTM1,HSP90AA1

—

PPAR�/RXR activation (118) ACADL, APOA1 CD36, FASN HSP90AA1 CD36PXR/RXR activation (57) ALDH1A7, G6PC,

HMGCS2, UGT1A1— GSTM1 CYP2B6

Cytochrome P450 panel (rat)–substrate is a xenobiotic(23)

CYP2A1, CYP2D3 — CYP2D3 CYP2B6

Xenobiotic metabolism (89) CYP2A1, CYP2B9, CYP2D3,CYP51A1, UGT1A1

— ALDH1L1, CYP2D3,CYP4F2, GSTA4,GSTM1

CYP2B6, UGT1A4

Fig. 4. Lipid metabolism network in the insulin-treated group. IPA software was used to graphically illustrate the molecular relationships betweenaltered proteins that were linked to lipid metabolism and proteins stored in the IKB. Molecules are represented as nodes, and the biologicalrelationships between nodes are represented as connecting lines. Molecules are labeled with their Entrez gene symbols (see Tables 2 and 3).Up-regulated proteins are red, down-regulated proteins are green, nuclear receptors are blue, and proteins inserted by IPA software are white.


at ASPE

T Journals on M


Dow

nloaded from


DiscussionIn this study, we used the STZ rat model of GDM to

determine whether plasma lipids compete with drugs forprotein binding in this disease. We also used this model toexamine whether GDM is associated with alterations in he-patic drug transporters and metabolic enzymes.

With respect to plasma protein binding, we first deter-mined the concentrations of two important plasma proteins:albumin and AAG. In humans, glyburide is 98% bound toalbumin (Olsen et al., 1995) and was 95% protein-bound inthis study’s vehicle group. In humans, saquinavir is 98%bound to AAG (Holladay et al., 2001) and was 90% protein-bound in this study’s vehicle group. In GDM, albumin con-centrations were decreased and AAG levels were increased.This is consistent with clinical studies involving patientswith poorly managed diabetes (Mattock et al., 2001).

If plasma protein concentrations were the sole determi-nant of protein binding, then one would expect, based onalbumin and AAG levels, decreased binding of glyburide andincreased binding of saquinavir in GDM. However, in thisstudy, the protein binding of both glyburide and saquinavirwas lower in GDM than in the control groups. Moreover, asillustrated in Fig. 1C, the extent of one drug’s protein bindingwas found to be highly predictive of the other’s. This isindicative of a contribution from a factor other than absoluteplasma protein content.

To test whether plasma lipids were, as hypothesized, com-peting with drugs for binding, we repeated our glyburideprotein binding experiments with delipidated plasma sam-ples. If plasma protein concentrations were the sole determi-nant of protein binding, then one would also expect thatdelipidation would have no effect on protein binding; how-ever, this was not the case in this study. Delipidation com-pletely normalized the protein binding of glyburide in GDM,yet it did not affect protein binding in either the vehicle orinsulin-treated groups. It has previously been reported thatthe binding capacity of human albumin for glyburide washalved by protein glycosylation, a process that can occur inGDM (Koyama et al., 1997). If glycosylation was preventingglyburide protein binding in our study, however, the plasmaprotein binding of glyburide should not have been completely

normalized by delipidation alone. Moreover, delipidation didnot affect glyburide protein binding in controls. Taken to-gether, our results provide strong evidence in support ofhyperlipidemia contributing directly to decreased plasmaprotein binding of drugs in GDM.

In our proteomics experiments, many alterations in GDMoccurred in proteins that are linked to lipid metabolism.Although studies have associated the etiology of hyperlipid-emia in STZ-induced diabetes with decreased insulin signal-ing and altered fat absorption, the contribution of hepaticproteins has remained largely unexplored.

Based on our findings in GDM, there seem to be a numberof prohyperlipidemic and antihyperlipidemic responses inthe liver, and some of these changes could greatly influenceplasma lipids. One particularly interesting prohyperlipi-demic response was the up-regulation of FDFT1 andCYP51A1 in GDM, both of which are proteins that playcritical roles in the conversion of acetyl-CoA to cholesterol. Infact, FDFT1 is a potential therapeutic target for hyperlipid-emia and an FDFT1 inhibitor, lapaquistat/TAK-475, hasbeen shown to be efficacious in preclinical and clinical studies(Nishimoto et al., 2003; Piper et al., 2006). Many of theprohyperlipidemic responses we observed are probably theresult of excess acetyl-CoA being shuttled to alternativeroutes of utilization. It is possible that, in GDM, enhancedhepatic lipid production is preferential to excess bloodglucose.

With respect to interesting antihyperlipidemic hepatic re-sponses in GDM, ACACA, ACAT1, and FASN were down-regulated, and ACADL was up-regulated. ACACA catalyzesthe carboxylation of acetyl-CoA to produce malonyl-CoA, asubstrate for the biosynthesis of fatty acids. ACAT1 catalyzesthe esterification of cholesterol, which is subsequently incor-porated into LDL and secreted. FASN s main function is tocatalyze the synthesis of acetyl-CoA and malonyl-CoA intolong-chain saturated fatty acids. Decreased expression ofACACA, ACAT1, and FASN in GDM would, therefore, beexpected to decrease the synthesis and secretion of lipids bythe liver. ACADL catalyzes the initial step in mitochondrialfatty acid oxidation, whereby fatty acids are broken down togenerate acetyl-CoA. This increased expression of ACADL inGDM would be expected to increase fatty acid utilization.Although interesting, it is important to note that these anti-hyperlipidemic hepatic responses, and those involving lipidtransporting proteins, did not correct the hyperlipidemiapresent in our model of GDM. Thus, the regulation of pro-teins linked to lipid metabolism is pathologically disrupted inGDM. Studies to determine the relative contribution of theprohyperlipidemic and antihyperlipidemic alterations de-scribed here are warranted.

An indirect effect of dysregulation in lipid metabolism reg-ulatory pathways, and one for which this study providessupport, is an activation of nuclear receptor networks thathave roles in the regulation of hepatic drug transporters andmetabolic enzymes. Studies have demonstrated that severalnuclear receptors have lipids as natural ligands (Handschinand Meyer, 2005). Activation of several of these nuclear re-ceptors is consistent with our observation that many of thedrug transporters and metabolizing enzymes that we identi-fied were up-regulated in GDM. Activation of FXR, for exam-ple, is consistent with the up-regulation of ABCB4 (Mdr2)observed in GDM. FXR also regulates, in part, Mdr1 (Sted-

Fig. 5. Hepatic Mdr1 expression. Hepatic Mdr1 was examined at the levelof mRNA via quantitative reverse transcription-PCR and at the level ofprotein via immunoblotting in the GDM and control groups. For mRNA,both genes encoding rodent Mdr1 were examined: Mdr1a and Mdr1b. Allresults are presented as mean � S.D. of five rats/group. �, Significantlydifferent from vehicle controls. #, Significantly different from GDM. �/#,p 0.05; ��/##, p 0.01.


at ASPE

T Journals on M


Dow

nloaded from


man et al., 2006), and up-regulation of this important drugtransporter was observed in GDM. Another example is theactivation of CAR and PXR, which is consistent with theup-regulation of UGT1A1 and CYP2A1 observed in GDM.These findings complement a body of literature demonstrat-ing increased drug-metabolizing capacities in experimentalanimals and humans with diabetes (Barnett et al., 1990;Song et al., 1990). It is noteworthy that OATP1A1, which canbe down-regulated in response to CAR, nuclear factor ery-throid 2-related factor 2, or PPAR� activation (Cheng et al.,2005; Maher et al., 2006), was found to be down-regulated inthe GDM and insulin-treated groups.

The finding that several ABC transporters were up-regu-lated in GDM is contrary to our previous report in nonpreg-nant, female rats with acute STZ-induced diabetes, whichdemonstrated that the hepatic expression of Mdr1a, ABCC2,and ABCG2 mRNA was down-regulated (Anger et al., 2009).In that study, rats were diabetic for half as long as the rats inthis study and had not yet developed overt hyperlipidemia, asmeasured on a lipemia index (unpublished data). Moreover,in our previous report we provided preliminary evidence thatinflammation plays a role in the observed down-regulation,but inflammatory responses are often blunted during latepregnancy (Aguilar-Valles et al., 2007), which is supportedby the fact that plasma CRP was not elevated in GDM. Wespeculate that the differences between this study and ourprevious study stem from an absence of hyperlipidemia andthe presence of inflammation in the latter. It would be inter-esting to compare the results obtained in our GDM modelwith those of a model of pregnancy complicated by maternaltype 1 diabetes, wherein the total duration of maternal dia-betes could be extended even further.

Our proteomics findings in the insulin-treated group de-serve further comment. First, insulin treatment did not nor-malize the expression of all proteins that were altered inGDM. Some of the changes in the insulin-treated group couldbe caused by our insulin treatment protocol. Daily injectionsof long-acting insulin lead to oscillating blood glucose concen-trations with the potential for brief periods of hyperglycemiapreinjection and hypoglycemia postinjection. It is possiblethat this was sufficient to stimulate some of the observedchanges, both those shared with the GDM group and thoseunique to the insulin-treated group. Although hyperlipid-emia is unlikely to have been transiently induced by thesebrief periods of abnormal blood glucose, some of the nuclearreceptors discussed here are also capable of sensing glucose.LXR is an example (Mitro et al., 2007).

Second, insulin treatment itself probably affected the ex-pression of hepatic proteins because there were proteins thatwere altered in the insulin-treated group that were not al-tered in GDM. Prohyperlipidemic alterations in the insulin-treated group that did not occur in GDM included the down-regulation of ABCD3. ABCD3 is involved in the transport oflipids, and down-regulation of this protein would be expectedto prevent the movement of plasma lipids to hepatic routes ofmetabolism or elimination. Insulin-mediated antihyperlipi-demic alterations that did not occur in GDM included thedown-regulation of SOAT2 and the up-regulation of LDLR.SOAT2 produces intracellular cholesterol esters from long-chain fatty acyl-CoA and cholesterol. As noted with ACAT1,esterified cholesterol in the liver is subsequently incorpo-rated into LDL and secreted. LDLR expression and activity is

a major determinant of plasma cholesterol content, by bind-ing to LDL and transporting it across membranes, and itsup-regulation in the insulin treatment group would be ex-pected to play an antihyperlipidemic role. An explanation forthese alterations not being observed in the GDM group couldbe that some of these proteins respond to insulin levels andare, consequently, sensitive to the exogenous administrationof insulin. LDLR is such a protein (Wade et al., 1989). Re-turning to plasma protein binding in our model of GDM,albumin production is also stimulated by insulin, which couldexplain the slightly elevated albumin concentrations andslightly increased protein binding observed in insulin-treatedrats (Lloyd et al., 1987).

In conclusion, our findings demonstrate that hyperlipid-emia in GDM contributes directly to decreased protein bind-ing of drugs and indirectly to drug transporter and metabolicenzyme up-regulation. Our findings advance our understand-ing of the impact that GDM-induced hyperlipidemia can haveon these specific drug disposition mechanisms. As men-tioned, protein binding strongly influences a drug’s distribu-tion and/or clearance. With distribution, placental transfer ofdrugs and the uptake of drugs by the liver and kidneys couldsignificantly increase. Elevated hepatic drug transportersand metabolizing enzymes in GDM would be expected tocompound changes in clearance that relate to the elevateduptake of drugs by the liver and kidneys. If confirmed inclinical populations, the effects reported in this study wouldneed to be considered when atypical therapeutic outcomesare observed in pregnancies that are complicated by GDM.

Acknowledgments

We thank Rhea Mehta (University of Toronto) for perfusing therat livers used in our proteomics experiments; Graham Robertsonand Lucy Jankova (ANZAC Research Institute, Concord, New SouthWales, Australia) and Mark S. Baker, Mark P. Malloy, Joel M. Chickand Thiri Zaw (Australian Proteome Analysis Facility, North Ryde,New South Wales, Australia) for invaluable expertise and technicalassistance with our proteomics experiments; and Sarabjit Gahir andJi Zhang for input throughout this study.

ReferencesAguilar-Valles A, Poole S, Mistry Y, Williams S, and Luheshi GN (2007) Attenuated

fever in rats during late pregnancy is linked to suppressed interleukin-6 produc-tion after localized inflammation with turpentine. J Physiol 583:391–403.

Anger GJ, Magomedova L, and Piquette-Miller M (2009) Impact of acute streptozo-tocin-induced diabetes on ABC transporter expression in rats. Chem Biodivers6:1943–1959.

Arredondo G, Suarez E, Calvo R, Vazquez JA, García-Sanchez J, and Martinez-JordaR (1999) Serum protein binding of itraconazole and fluconazole in patients withdiabetes mellitus. J Antimicrob Chemother 43:305–307.

Barnett CR, Gibson GG, Wolf CR, Flatt PR, and Ioannides C (1990) Induction ofcytochrome P450III and P450IV family proteins in streptozotocin-induced diabe-tes. Biochem J 268:765–769.

Basaran A (2009) Pregnancy-induced hyperlipoproteinemia: review of the literature.Reprod Sci 16:431–437.

Betteridge DJ (2000) Diabetic dyslipidaemia. Diabetes Obes Metab 2(Suppl 1):S31–S36.

Cham BE and Knowles BR (1976) A solvent system for delipidation of plasma orserum without protein precipitation. J Lipid Res 17:176–181.

Chauvelot-Moachon L, Tallet F, Durlach-Misteli C, and Giroud JP (1988) Delipida-tion of �1-acid glycoprotein. Propranolol binding to this glycoprotein and itsmodification by extracted material and exogenous lipids. J Pharmacol Methods20:15–28.

Cheng X, Maher J, Dieter MZ, and Klaassen CD (2005) Regulation of mouse organicanion-transporting polypeptides (Oatps) in liver by prototypical microsomal en-zyme inducers that activate distinct transcription factor pathways. Drug MetabDispos 33:1276–1282.

Couch SC, Philipson EH, Bendel RB, Pujda LM, Milvae RA, and Lammi-Keefe CJ(1998) Elevated lipoprotein lipids and gestational hormones in women with diet-treated gestational diabetes mellitus compared to healthy pregnant controls. JDiabetes Complicat 12:1–9.

de la Fuente L, Lukas JC, Jauregizar N, Vazquez JA, Calvo R, and Suarez E (2002)


at ASPE

T Journals on M


Dow

nloaded from


Prediction of unbound propofol concentrations in a diabetic population. Ther DrugMonit 24:689–695.

Ginsberg HN (1996) Diabetic dyslipidemia: basic mechanisms underlying the com-mon hypertriglyceridemia and low HDL cholesterol levels. Diabetes 45 (Suppl 3):S27–S30.

Handschin C and Meyer UA (2005) Regulatory network of lipid-sensing nuclearreceptors: roles for CAR, PXR, LXR, and FXR. Arch Biochem Biophys 433:387–396.

Harris SB, Ekoe JM, Zdanowicz Y, and Webster-Bogaert S (2005) Glycemic controland morbidity in the Canadian primary care setting (results of the diabetes inCanada evaluation study). Diabetes Res Clin Pract 70:90–97.

Holladay JW, Dewey MJ, Michniak BB, Wiltshire H, Halberg DL, Weigl P, Liang Z,Halifax K, Lindup WE, and Back DJ (2001) Elevated �-1-acid glycoprotein reducesthe volume of distribution and systemic clearance of saquinavir. Drug MetabDispos 29:299–303.

Koyama H, Sugioka N, Uno A, Mori S, and Nakajima K (1997) Effects of glycosyla-tion of hypoglycaemic drug binding to serum albumin. Biopharm Drug Dispos18:791–801.

Krentz AJ (2003) Lipoprotein abnormalities and their consequences for patients withtype 2 diabetes. Diabetes Obes Metab 5 (Suppl 1):S19–S27.

Lawrence JM, Contreras R, Chen W, and Sacks DA (2008) Trends in the prevalenceof preexisting diabetes and gestational diabetes mellitus among a racially/ethnically diverse population of pregnant women, 1999–2005. Diabetes Care 31:899–904.

Lloyd CE, Kalinyak JE, Hutson SM, and Jefferson LS (1987) Stimulation of albumingene transcription by insulin in primary cultures of rat hepatocytes. Am J PhysiolCell Physiol 252:C205–C214.

Maher JM, Slitt AL, Callaghan TN, Cheng X, Cheung C, Gonzalez FJ, and KlaassenCD (2006) Alterations in transporter expression in liver, kidney, and duodenumafter targeted disruption of the transcription factor HNF1�. Biochem Pharmacol72:512–522.

Mattock MB, Cronin N, Cavallo-Perin P, Idzior-Walus B, Penno G, Bandinelli S,Standl E, Kofinis A, Fuller JH, and EURODIAB IDDM Complications Study(2001) Plasma lipids and urinary albumin excretion rate in type 1 diabetes mel-litus: the EURODIAB IDDM Complications Study. Diabet Med 18:59–67.

Mehta R, Wong L, and O’Brien PJ (2009) Cytoprotective mechanisms of carbonylscavenging drugs in isolated rat hepatocytes. Chem-Biol Interact 178:317–323.

Mitro N, Mak PA, Vargas L, Godio C, Hampton E, Molteni V, Kreusch A, and SaezE (2007) The nuclear receptor LXR is a glucose sensor. Nature 445:219–223.

Molloy MP (2008) Isolation of bacterial cell membranes proteins using carbonateextraction. Methods Mol Biol 424:397–401.

Nanovskaya TN, Nekhayeva I, Hankins GD, and Ahmed MS (2006) Effect of humanserum albumin on transplacental transfer of glyburide. Biochem Pharmacol 72:632–639.

Nishimoto T, Amano Y, Tozawa R, Ishikawa E, Imura Y, Yukimasa H, and Sugiyama

Y (2003) Lipid-lowering properties of TAK-475, a squalene synthase inhibitor, invivo and in vitro. Br J Pharmacol 139:911–918.

Notarianni LJ (1990) Plasma protein binding of drugs in pregnancy and in neonates.Clin Pharmacokinet 18:20–36.

Olsen KM, Kearns GL, and Kemp SF (1995) Glyburide protein binding and the effectof albumin glycation in children, young adults, and older adults with diabetes.J Clin Pharmacol 35:739–745.

Perucca E, Ruprah M, and Richens A (1981) Altered drug binding to serum proteinsin pregnant women: therapeutic relevance. J R Soc Med 74:422–426.

Petrovic V, Wang JH, and Piquette-Miller M (2008) Effect of endotoxin on theexpression of placental drug transporters and glyburide disposition in pregnantrats. Drug Metab Dispos 36:1944–1950.

Piper E, Price G, and Chen Y (2006) Abstract 1493: TAK-475, a squalene synthaseinhibitor, improves lipid profile in hyperlipidemic subjects. Circulation 114:II_288.

Ruiz-Cabello F and Erill S (1984) Abnormal serum protein binding of acidic drugs indiabetes mellitus. Clin Pharmacol Ther 36:691–695.

Song BJ, Veech RL, and Saenger P (1990) Cytochrome P450IIE1 is elevated inlymphocytes from poorly controlled insulin-dependent diabetics. J Clin EndocrinolMetab 71:1036–1040.

Spector AA, Santos EC, Ashbrook JD, and Fletcher JE (1973) Influence of free fattyacid concentration on drug binding to plasma albumin. Ann NY Acad Sci 226:247–258.

Stedman C, Liddle C, Coulter S, Sonoda J, Alvarez JG, Evans RM, and Downes M(2006) Benefit of farnesoid X receptor inhibition in obstructive cholestasis. ProcNatl Acad Sci USA 103:11323–11328.

Trovik TS, Jaeger R, Jorde R, Ingebretsen O, and Sager G (1992) Plasma proteinbinding of catecholamines, prazosin, and propranolol in diabetes mellitus. EurJ Clin Pharmacol 43:265–268.

Tsen LC, Tarshis J, Denson DD, Osathanondh R, Datta S, and Bader AM (1999)Measurements of maternal protein binding of bupivacaine throughout pregnancy.Anesth Analg 89:965–968.

Wade DP, Knight BL, and Soutar AK (1989) Regulation of low-density-lipoprotein-receptor mRNA by insulin in human hepatoma Hep G2 cells. Eur J Biochem181:727–731.

Wasan KM, Ng SP, Wong W, and Rodrigues BB (1998) Streptozotocin- and alloxan-induced diabetes modifies total plasma and lipoprotein lipid concentration andcomposition without altering cholesteryl ester transfer activity. Pharmacol Toxicol83:169–175.

Worner W, Preissner A, and Rietbrock N (1992) Drug-protein binding kinetics inpatients with type I diabetes. Eur J Clin Pharmacol 43:97–100.

Address correspondence to: Dr. Micheline Piquette-Miller, Leslie Dan Fac-ulty of Pharmacy, University of Toronto, 144 College St., Toronto, Ontario,Canada, M5S 3M2. E-mail: [email protected]


at ASPE

T Journals on M


Dow

nloaded from


impact of hyperlipidemia on plasma protein binding...

Documents