Thiol Oxidative Stress Induced by Metabolic Disorders AmplifiesMacrophage Chemotactic Responses and AcceleratesAtherogenesis and Kidney Injury in LDL Receptor-Deficient Mice

Mu Qiao1,§, Qingwei Zhao1,§, Chi Fung Lee1, Lisa R. Tannock2, Eric J. Smart3, Richard G.LeBaron4, Clyde F. Phelix4, Yolanda Rangel1, and Reto Asmis1,*1Office of the Dean, School of Health Professions, University of Texas Health Science Center atSan Antonio, Texas2Division of Endocrinology and Molecular Medicine, University of Kentucky, Lexington, Kentucky3Department of Pediatrics, University of Kentucky, Lexington, Kentucky4University of San Antonio, San Antonio, Texas

AbstractBackground—Strengthening the macrophage glutathione redox buffer reduces macrophagecontent and decreases the severity of atherosclerotic lesions in LDL receptor-deficient (LDLR−/−)mice, but the underlying mechanisms were not clear. This study examined the effect of metabolicstress on the thiol redox state, chemotactic activity in vivo and the recruitment of macrophages intoatherosclerotic lesions and kidneys of LDL-R−/− mice in response to mild, moderate and severemetabolic stress.

Methods and Results—Reduced glutathione (GSH) and glutathione disulfide (GSSG) levels inperitoneal macrophages isolated from mildly, moderately and severe metabolically-stressed LDL-R−/− mice were measured by HPLC, and the glutathione reduction potential (Eh) was calculated.Macrophage Eh correlated with the macrophage content in both atherosclerotic (r2=0.346, P=0.004)and renal lesions (r2=0.480, P=0.001) in these mice as well as the extent of both atherosclerosis(r2=0.414, P=0.001) and kidney injury (r2=0.480, P=0.001). Compared to LDL-R−/− mice exposedto mild metabolic stress, macrophage recruitment into MCP-1-loaded Matrigel plugs injected intoLDL-R−/− mice increased 2.6–fold in moderately metabolically-stressed mice and 9.8–fold inseverely metabolically-stressed mice. The macrophage Eh was a strong predictor of macrophagechemotaxis (r2=0.554, P<0.001).

Conclusion—Thiol oxidative stress enhances macrophage recruitment into vascular and renallesions by increasing the responsiveness of macrophages to chemoattractants. This novel mechanismcontributes at least in part to accelerated atherosclerosis and kidney injury associated withdyslipidemia and diabetes in mice.

*Correspondence should be addressed to: Reto Asmis, Office of the Dean, School of Health Professions, University of Texas HealthScience Center at San Antonio, 7703 Floyd Curl Drive, MSC 6243, San Antonio, TX 78229; Tel.: (210) 567-2720, FAX: (210) 567-2709;asmis@uthscsa.edu.§Both authors contributed equally to this work.DISCLOSURES There are no conflicts to disclose.

NIH Public AccessAuthor ManuscriptArterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

Published in final edited form as:Arterioscler Thromb Vasc Biol. 2009 November ; 29(11): 1779–1786. doi:10.1161/ATVBAHA.109.191759.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

INTRODUCTIONMetabolic disorders such as hypercholesterolemia and diabetes are strongly associated withboth macro- and microvascular diseases, a common feature of which is the recruitment of bloodmonocyte-derived macrophages to sites of vascular injury. While most studies exploring themechanisms underlying atherosclerosis and other vascular pathologies have focused on theimpact of a dysregulated metabolism on the vasculature itself, a number of more recent studiessuggest that metabolic disorders may also directly impact monocytes and alter theirfunctionalities in ways that promote and accelerate the disease process. Phenotypicalabnormalities in blood monocytes of diabetic patients have been reported, including alteredmetabolism 1–3, phagocytosis 4;5 and cytokine release 6–8. Furthermore, peritonealmacrophages isolated from either atherosclerosis-prone mice or diabetic mice show alteredcytokine and chemokine responses compared with macrophages from healthy control mice 9;10. However, it is not yet well-understood to what extent monocyte dysfunction induced bymetabolic diseases contributes to macrophage recruitment and vascular diseases such asatherosclerosis.

The recruitment of blood monocyte-derived macrophages into the vessel wall is consideredone of the earliest events in the onset of atherosclerosis 11;12. The mechanisms that triggermacrophage recruitment are not fully understood but prolonged retention and subsequentmodification of LDL may play a critical role 13–15. Modified LDL, particularly oxidatively-modified LDL stimulates vascular cells and macrophages to secrete a vast array ofinflammatory molecules, including chemokines, which in turn promote and sustain acontinuous influx of macrophages into the vasculature 12;16. Studies in transgenic and knockoutmice revealed that at least two major chemokine/chemokine receptor systems are involved inthe recruitment of macrophages into atherosclerotic lesions, MCP-1/CCL2 and the receptorCCR2, and RANTES/CCL5 and the receptor CCR5. Deficiencies in either the macrophagechemoattractant MCP-1 or its receptor CCR2 reduced the severity of atherosclerosis indifferent mouse models of atherosclerosis, and the reduction in lesion size was accompaniedby a reduction in macrophage accumulation 17–19. Inhibiting the effect of RANTES in highcholesterol diet-fed LDL-R−/− mice with Met-RANTES, a RANTES antagonist, also reducesmacrophage infiltration and inhibited lesion formation 20. The reduction in these lesions'macrophage content was associated with a more stable plaque phenotype. Deficiency in theRANTES receptors CCR5 in ApoE−/− mice showed similar effects 21, but other studies suggestthat CCR5 contributes primarily to the later stages of lesion development 22;23.

While the individual contributions of these chemoattractants and their receptors to macrophagerecruitment and atherogenesis appear to change throughout the maturation of atheroscleroticlesions, these studies suggest that macrophage recruitment can be accelerated either byincreasing the chemotactic signals originating from the vasculature or by sensitizing monocytesto these signals, for example by increasing chemokine receptor expression levels and/oractivity. Here we examined the hypothesis that metabolic stress promotes macrophagerecruitment and atherogenesis by increasing the responsiveness of circulating blood monocytesto chemoattractants, specifically to MCP-1. We compared normolipidemic,hypercholesterolemic and dyslipidemic diabetic LDL-R−/− mice and showed that with eachincrease in the level of metabolic stress macrophage recruitment into the vasculature andkidneys increased, and the development of vascular and renal lesions accelerated accordingly.Increased monocyte responsiveness to chemoattractants appeared to account at least in part forthe observed increase in macrophage recruitment as MCP-1-loaded Matrigel plugs implantedinto these mice also showed enhanced macrophage accumulation in response to each increasein the level of metabolic stress, even though the concentration of chemoattractant was identicalin each implant. We provide evidence that increased cellular thiol oxidation induced bymetabolic stress sensitizes macrophages to MCP-1-induced chemotaxis and that thiol oxidative

Qiao et al. Page 2

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

stress contributes to the enhanced recruitment of macrophages to sites of tissue injury inmetabolically-challenged mice.

MATERIALS AND METHODSAnimals

Female LDL-R−/− mice (B6.129S7-LdlrtmIHer/J, stock no. 002207) on a C57BL/6J backgroundand C57BL/6J mice were obtained from The Jackson Laboratories (Bar Harbor, ME). Afterone week on a maintenance diet (MD, AIN-93G, BioServ), mice were randomized into threegroups and subjected to either mild (MD), moderate (HFD) or severe metabolic stress (STZ+HFD). Mice in the STZ+HFD group were rendered diabetic with intraperitoneal injection ofstreptozotocin (STZ; 60 mg × kg−1 × day−1) dissolved in citrate buffer (50 mM; pH=4.5) forfive consecutive days and, after a two-day rest, again for two consecutive days. Mice in theMD and HFD groups received a comparable volume of citrate buffer. To inducehypercholesterolemia, mice in the HFD and STZ+HFD group were fed a diet supplementedwith fat (21% wt/wt) and cholesterol (0.15% wt/wt; AIN-76A, BioServ) beginning at 3 weeksafter the first injection, for a total of 12 weeks. The remaining animals (MD group) receivedMD for 12 weeks. All studies were performed with the approval of the UTHSCSA InstitutionalAnimal Care and Use Committee.

Analysis of AtherosclerosisAfter mice were euthanatized, the right atrium was removed and hearts and aortas wereperfused with PBS through the left ventricule. Hearts were embedded in OCT and frozen ondry ice. Aortas were fixed over night with 4% paraformaldehyde in PBS, dissected from theproximal ascending aorta to the bifurcation of the iliac artery, and had their adventitial fatremoved. For en face analysis, aortas were stained with Oil Red O (ORO), openedlongitudinally, pinned flat onto black paper placed over dental wax and digitally photographedat a fixed magnification 24. Total aortic area and lesion areas were calculated using ImageProPlus 6.0 (Media Cybernetics). As a second measure of atherosclerosis, lesions of the aortic rootwere analyzed. Serial sections were cut through a 700 μm segment of the aortic root. For eachmouse, 8 sections (7 μm) separated by 70 μm were examined. Each section was stained withORO, counterstained with hematoxylin (Vector Labs, Burlingame, CA) and digitized. Lesionarea was measured using ImagePro Plus 6.0 (Media Cybernetics) and expressed as millimeterssquared 24. Macrophages were detected with anti-CD68 antibodies (Serotech) 24. In situlabeling of S-glutathionylated proteins in serial sections from the aortic root was performedusing the method described by Reynaert et al. 25.

Analysis of Kidney InjuryKidneys were frozen in OCT, and five cross cryosections separated by an interval of 50 mmwere stained with ORO or Masson-Trichrome 26;27, or were processed forimmunohistochemical analysis of macrophage content with anti-CD68 antibodies (Serotech)24.

Blood and Urine AnalysisMice were fasted overnight prior to glucose and lipid measurements. Glucose was measuredbiweekly using a Contour®meter (Bayer). For measurements of white blood cell (WBC) andmonocyte counts, blood was obtained by retro-orbital bleed after 11 weeks of diet feeding, anddifferential blood cell counts were performed by the Department of Laboratory AnimalResources at UTHSCSA on a VetScan HM II Analyzer (Abaxis). After peritoneal macrophageswere harvested, blood was drawn by cardiac puncture. Plasma cholesterol and triglyceridelevels were determined using enzymatic assay kits (Wako Chemicals). Lipoprotein cholesterol

Qiao et al. Page 3

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

distributions were analyzed in individual plasma samples (50 μl) for 8 mice in each group.Plasma samples were separated by size exclusion chromatography on tandem superose 6–10/300GL columns 28. Serum amyloid A (Biosource), leptin and insulin levels (Chrystal Chem)were measured by ELISA. Urine creatinine and albumin concentrations were measured byELISA (Exocell).

Macrophage Glutathione/Glutathione Disulfide AnalysisResident peritoneal cells were harvested by lavage, plated, and after 3 h non-adherent cellswere removed with multiple washing 24. Adherent cells were cultured overnight prior to cellharvest. Glutathione and glutathione disulfide levels were measured by HPLC and values werenormalized to cell DNA levels as described elsewhere 29. The glutathione reduction potentialEh was calculated using the Nernst equation for GSH/GSSH and E0=−264 mV for pH=7.430. The mean macrophage volume was estimated at 1.4 × 10−6 μl 31.

In Vivo Matrigel Macrophage Recruitment AssayThree days prior to the end of the study, each mouse received two Matrigel plugs, one in eachflank. Growth factor-reduced Matrigel (BD Biosciences) supplemented with either vehicle orrecombinant MCP-1 (300 nM) was injected subcutaneously into the right and left flank of eachmouse. The plugs were removed at the time of sacrifice, dissolved, and cells were counted inan automated, video-based hemocytometer (Nexcelom, Lawrence, MA). Cell staining withantibodies directed against macrosialin/CD68 confirmed that >93% of the cells recruited intothe Matrigel plugs were macrophages.

StatisticsData were analyzed using ANOVA (SPSS 16.0). Data were tested for use of parametric ornonparametric post hoc analysis, and multiple comparisons were performed by using the LeastSignificant Difference method. All data are presented as mean ± SE. Results were consideredstatistically significant at the P < 0.05 level.

RESULTSMetabolic Stress in LDL-R −/− Mice Promotes the Oxidation of the Glutathione Redox Statein Macrophages and Accelerates Atherosclerotic Lesion Formation

To examine the effect of metabolic stress on the thiol redox state of monocyte-derivedmacrophages, we determined the GSH/GSSG ratio in peritoneal macrophages isolated fromLDL-R−/− mice that were exposed for 12 weeks to either mild (MD), moderate (HFD), or severemetabolic stress (STZ-induced hyperglycemia plus HFD). With increasing metabolic stress,these mice showed a progressive decrease in their macrophage GSH/GSSG ratios, indicatinga progressive increase in intracellular (thiol) oxidative stress (Fig. 1A). As expected, increasingmetabolic stress increased plasma cholesterol levels (Table 1) and accelerated atheroscleroticlesion formation in both the aortic arch and descending aorta of LDL-R−/− mice (Fig. 1B).Lesion size showed a strong correlation with plasma cholesterol levels in these animals(r2=0.635, P<0.001, Fig, 1C). HFD feeding increased cholesterol in both the IDL/LDL and theVLDL fractions, whereas STZ treatment prior to HFD feeding resulted in only a minor,statistically not significant (P=0.19) further increase in plasma total cholesterol (Table 1),primarily as VLDL cholesterol (Supplemental Fig. I). Both plasma triglycerides and glucoselevels were also increased by the STZ treatment, but the differences did not reach statisticalsignificance.

Surprisingly, insulin levels were similar in non-diabetic and diabetic mice (Table 1). However,it is unlikely that β-cells recovered from the STZ treatment; as fasting glucose levels peaked

Qiao et al. Page 4

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

four weeks after the first STZ injection, i.e. one week after initiating HFD feeding, andplateaued thereafter. A more likely explanation for the similar insulin levels is that the depletionof β-cells was incomplete. Injecting higher doses of STZ resulted in slightly higher fastingglucose levels in this mouse model, but the survival rate of the HFD-fed animals droppedprecipitously at these higher STZ doses. Also, mice were fasted for 15 h to obtain true fastingglucose levels, and this extended fasting period likely contributed to the low insulin levels weobserved in the non-diabetic mice.

To evaluate the relationship between thiol oxidative stress in macrophages induced by low,modest or severe metabolic stress and atherosclerotic lesion formation, we calculated theglutathione reduction potential (Eh) based on the cellular GSH and GSSG concentrations wedetermined in peritoneal macrophages isolated from each animal. The correlation between themacrophage glutathione reduction potential and aortic lesion size was statistically highlysignificant (r2=0.414, P=0.001; Fig. 1D), whereas plasma triglyceride levels (r2=0.345,P=0.003) and plasma glucose concentrations (r2=0.252, P=0.011) were both poorer predictorsof atherosclerotic lesion formation.

With increasing metabolic stress we also observed an acceleration in atherosclerotic lesionformation in the aortic root (Fig. 2A, Supplemental Fig. II). Increased lesion area was paralleledby an increase in macrophage content in the vessel wall (Fig. 2B, Supplemental Fig. II).Importantly, the macrophage glutathione reduction potential was a strong predictor ofmacrophage content in the lesions (expressed as mm2: r2=0.346, P=0.004; expressed as %lesion area: r2=0.683, P<0.001), suggesting that thiol oxidative stress in blood monocytespromotes macrophage recruitment into the vascular wall.

Han et al. reported that increasing extracellular cholesterol levels by adding native LDL toTHP-1 monocytes increases their chemotactic response to MCP-1 32. We found that plasmacholesterol levels in our mouse models correlated with the macrophage glutathione reductionpotentials in these mice (r2=0.527, P=0.01), suggesting that in addition to increasing LDLdepositions within the vasculature, elevated plasma cholesterol may also affect lesion sizeindirectly by promoting thiol oxidative stress in macrophages and by increasing theirresponsiveness to chemoattractants like MCP-1.

Protein-S-glutathionylation is a reversible posttranslational protein modification involved inredox signaling, and serves as a marker of cellular (thiol) oxidative stress 33. To determinewhether thiol oxidative stress is increased in macrophages recruited into atherosclerotic lesions,we incubated aortic sections with glutaredoxin to specifically reduce protein-glutathione mixeddisulfides and labeled the released protein thiols with thiol-specific Alexa545-conjugated N-ethyl maleimide. Compared to control mice (MD), we observed increased incorporation offluorescent label into proteins of aortic lesions from moderately (HFD) or severelymetabolically-stressed (STZ+HFD) mice (Fig. 2C), indicating that protein-S-glutathionylation– and thus thiol oxidative stress – increased with increasing levels of metabolic stress.Furthermore, protein-S-glutathionylation also localized to macrophages within these lesion,and intensified with increasing metabolic stress, providing further evidence that macrophagesrecruited to aortic lesions of metabolically-stressed (STZ+HFD) mice are thiol oxidativelystressed.

Although the number of WBC appeared to increase with increasing metabolic stress, we didnot observe any statistically significant differences in blood monocyte counts between mildly(MD), moderately (HFD) or severely metabolically-stressed (STZ+HFD) mice (Table 1). Thissuggests that the increase in macrophage recruitment in these mice was not the result ofincreased blood monocyte counts.

Qiao et al. Page 5

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Hyperglycemia in Dyslipidemic LDL-R−/− Mice Accelerates Glomerular Lipid Deposition,Macrophage Accumulation and Kidney Injury

Next, we examined the kidneys of these mice to determine whether the apparent relationshipbetween the macrophage glutathione reduction potential, macrophage recruitment and theseverity of vascular lesions also extended to other sites of tissue injury. The kidneys of LDL-R−/− mice exposed to low, modest or severe metabolic stress showed progressive glomerularlipid deposition (Fig. 3A, Supplemental Fig. III), which was accompanied by increasedmacrophage recruitment into the glomeruli (Fig. 3B, Supplemental Fig. III). Like in thevasculature, we found a highly significant correlation between the macrophage glutathionereduction potential and macrophage accumulation in the kidney (r2=0.481, P=0.001,Supplemental Fig. IV). Increased renal lipid deposition and macrophage accumulation wasaccompanied by increased fibrosis (Supplemental Fig. III). Interestingly, we also observed asignificant correlation between macrophage glutathione reduction potential and renal lipidaccumulation in LDL-R−/− mice exposed to mild, modest or severe metabolic stress (r2=0.526,P<0.001, Supplemental Fig. V). Of note, mice from the STZ+HFD group showed a significantincrease in the urinary albumin/creatinine ratio (Table 1), indicating that the extent of kidneyinjury in these severely metabolically-stressed, diabetic mice had resulted in the loss of renalfunction.

Metabolic Stress Enhances Macrophage Chemotactic Activity In VivoTo examine whether metabolic stress changes the responsiveness of macrophages tochemoattractants, we injected mildly (MD), moderately (HFD) or severely metabolically-stressed (STZ+HFD) mice with Matrigel plugs loaded with either vehicle or MCP-1. After 3days, the plugs were removed and macrophage recruitment into these plugs was quantified.Increasing the level of metabolic stress in LDL-R−/− mice by feeding a HFD, increasedmacrophage recruitment into MCP-1-loaded Matrigel plugs 2.6-fold (Fig. 4A). Nomacrophages were detected in vehicle-loaded plugs removed from the opposite flank of eitherMD-or HFD-fed mice. When we further increased metabolic stress by inducing hyperglycemiain LDL-R−/− mice prior to HFD feeding, we observed an additional 3.8-fold increase inmacrophage recruitment. Even though the MCP-1 concentration in the Matrigel injected intoall three groups of mice was identical, 9.8-fold more macrophages were recruited into MCP-1-loaded Matrigel plugs isolated from severely metabolically-stressed, diabetic LDL-R−/− micethan into plugs removed from MD-fed LDL-R−/− mice (Fig. 4A). In contrast to mildly (MD)and moderately metabolically-stressed LDL-R−/− mice (HFD), we also detected small numbersof macrophages in vehicle-loaded plugs from severely metabolically-stressed LDL-R−/− mice(Fig. 4A, STZ+HFD, numbers in parentheses). This indicates that the sensitivity ofmacrophages in these animals was increased to such an extent that the cells even responded toresidual chemotactic factors present in the growth factor-poor Matrigel plugs. When wemeasured GSH and GSSG levels and calculated the glutathione reduction potential of theperitoneal macrophages, we again found a highly significant correlation (r2=0.554, P<0.001)between the macrophage glutathione reduction potential and the number of macrophagesrecruited into the MCP-1 loaded Matrigel plugs.

DISCUSSIONIn this study we examined whether metabolic stress induced by hypercholesterolemia alone orhyperlipidemia plus hyperglycemia accelerates atherosclerosis and renal injury by promotingthiol oxidative stress in macrophages and increasing the recruitment of macrophages to sitesof tissue injury. We found that with each additional level of metabolic stress, macrophageaccumulation increased in both atherosclerotic lesions and in kidneys of LDL-R−/− mice.Analysis of the macrophage glutathione reduction potential (Eh) revealed that metabolic stresspromotes intracellular thiol oxidation and that oxidation of the macrophage glutathione

Qiao et al. Page 6

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

reduction state correlates with both accelerated macrophage recruitment and increased lesionseverity. Macrophage recruitment into MCP-1-loaded Matrigel plugs implanted in these micewas also accelerated with each additional level of metabolic stress, suggesting that increasedcellular thiol oxidation induced by metabolic stress amplifies macrophage responses tochemotactic signals. These studies identified a novel thiol-dependent mechanism thatcontributes to macrophage recruitment and the development of atherosclerotic lesions.

Macrophages are continually recruited into atherosclerotic lesions, and macrophageaccumulation appears to increase in proportion to lesion size 34. Although the rate ofmacrophage recruitment depends on 1) the nature of chemoattractants released by the injuredtissue and 2) the size of chemotactic gradient, our data suggest that there is a third key factordetermining the extent of macrophage accumulation and thus lesion size: the intensity of themonocytes' response to a chemotactic signal. The increase in chemotactic activity we observedin response to increased metabolic stress could not be fully accounted for by a singlecardiovascular risk factor, e.g. plasma cholesterol, triglycerides, or glucose levels, suggestingthat the overall metabolic state rather than any individual metabolite determines the monocyteresponsiveness to chemoattractants. Monocytes appear to act as sensors of metabolic stressand their chemotactic activity may reflect cardiovascular risk, a concept, if confirmed, thatwould have important diagnostic and therapeutic implications.

Changes in the cellular redox environment not only initiate (or inhibit) individual signalingpathways but dictate a cell's fate with regard to function, differentiation, proliferation andsurvival 35. The cellular glutathione reduction potential serves as a key indicator of a cell'sredox environment 35. Previously, we showed that increased expression of macrophageglutathione reductase activity reduces the severity of atherosclerosis in LDL-R−/− mice,providing evidence that the glutathione redox state in macrophages plays a critical role in thedevelopment and progression of atherosclerotic lesions 24. The strong correlations we observedin this study between the macrophage glutathione reduction potential (Eh) and 1) macrophagechemotactic activity in vivo, 2) macrophage accumulation in vascular and renal lesions, and 3)the severity of atherosclerotic lesions and renal injury, confirm that the level of thiol oxidationin macrophages appears to be a critical determinant for both the extent of macrophagerecruitment and the rate of lesion development.

Our in vivo chemotaxis assay demonstrated that macrophage chemotactic activity increasedwith increasing levels of metabolic stress, even though the concentration of chemoattractant,i.e. MCP-1, was identical in all Matrigel plugs. Two mechanisms could have contributed tothis increase in macrophage recruitment: increased numbers of blood monocytes and/orincreased responsiveness of monocytes to chemoattractants. We observed no statisticallysignificant increases in blood monocyte counts in metabolically-stressed mice, indicating thatchronic metabolic stress rendered monocytes-derived macrophages hyperresponsive toMCP-1. Increased cell surface expression of the MCP-1 receptor CCR2 could account for thedetected increase in macrophage responsiveness to MCP-1. In support of this hypothesis, Hanet al. reported that CCR2 transcript levels are elevated 2-fold in human blood monocytesisolated from hypercholesterolemic subjects as compared with monocytes fromnormolipidemic subjects, although cell surface expression of CCR2 protein was not determinedin this study 32. Increased surface expression of CCR2, however, was detected on monocytesfrom diabetic patients 36. The lack of a commercially available antibody directed againstmurine CCR2 prevented us from directly testing whether CCR2 surface expression inmacrophages was increased in our metabolically-stressed mice, but the increase in macrophagechemotactic activity we observed in moderately and severely metabolically-stressed micewould be consistent with increased CCR2 surface expression.

Qiao et al. Page 7

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Alternatively, thiol oxidative stress may increase CCR2 sensitivity toward MCP-1 and/orsignaling in response to MCP-1 binding. An example of how thiol oxidative stress and protein-S-glutathionylation can affect chemokine signaling was illustrated by Kanda et al. for PDGFreceptor β (PDGFRβ) 37. PDGF-B, a PDGFRβ agonist and mitogen, is also a potent monocytechemoattractant involved in tissue repair and wound healing. The authors show that S-glutathionylation of active site cysteine residues of LMW-PTP inhibits the ability of thephosphatase to counterregulate PDGFRβ activation, thus promoting hyperactivation of thereceptor and its downstream signaling pathways in response to agonist binding. MCP-1signaling through CCR2 and the subsequent internalization of the receptor is regulated by G-protein-related kinases (GRK) GRK2 and GRK3 38. Both GRK2 activity and protein levelsare attenuated by oxidative stress, suggesting a potential role for thiol oxidation in theregulation of both CCR2 surface expression and MCP-1-dependent signaling 39. Loss of GRK2activity and the subsequent dysregulation of the protein kinase/phosphatase steady statecontrolling CCR2 activity would be expected to promote CCR2 surface retention and tohyperactivate the receptor in response to MCP-1. Amplification of MCP-1-induced signalingby (thiol) oxidative stress would provide an additional mechanism for the enhancedchemotactic activity of macrophages observed in metabolically-stressed mice.

In summary, these studies provide evidence that increased macrophage responsiveness tochemoattractants induced by thiol oxidative stress is a novel mechanism contributing toincreased macrophage recruitment and accelerated atherosclerosis and renal injury associatedwith metabolic disorders. Our data underscore the critical role of the macrophage glutathioneredox state in regulating macrophage chemotaxis and identify the glutathione-dependentantioxidant system in monocytes as a potential therapeutic target for the prevention andtreatment of atherosclerosis.

Supplementary MaterialRefer to Web version on PubMed Central for supplementary material.

ACKNOWLEDGEMENTSWe would like to thank Wuqiong Ma for her technical assistance and Kirsten Gallagher for her comments and criticalreading of the manuscript.

FUNDING SOURCES This work was supported by grants to R.A. from the NIH (HL-70963) and the American HeartAssociation (0855011F).

ABBREVIATIONSEh, glutathione reduction potentialGrx, glutaredoxinGSH, reduced glutathioneGSSG, glutathione disulfide (oxidized glutathione)HFD, high fat dietHPLC, high performance liquid chromatographyMCP-1, monocyte chemotactic protein-1MD, maintenance dietORO, Oil Red OPBS, phosphate-buffered salineROS, reactive oxygen speciesSAA, serum amyloid ASTZ, streptozotocinWBC, white blood cells

Qiao et al. Page 8

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

REFERENCES1. Noritake M, Katsura Y, Shinomiya N, Kanatani M, Uwabe Y, Nagata N, Tsuru S. Intracellular hydrogen

peroxide production by peripheral phagocytes from diabetic patients. Dissociation betweenpolymorphonuclear leucocytes and monocytes. Clin Exp Immunol 1992;88:269–274. [PubMed:1572091]

2. Josefsen K, Nielsen H, Lorentzen S, Damsbo P, Buschard K. Circulating monocytes are activated innewly diagnosed type 1 diabetes mellitus patients. Clin Exp Immunol 1994;98:489–493. [PubMed:7994912]

3. Hill HR, Hogan NA, Rallison ML, Santos JI, Charette RP, Kitahara M. Functional and metabolicabnormalities of diabetic monocytes. Adv Exp Med Biol 1982;141:621–628. [PubMed: 6283839]

4. Katz S, Klein B, Elian I, Fishman P, Djaldetti M. Phagocytotic activity of monocytes from diabeticpatients. Diabetes Care 1983;6:479–482. [PubMed: 6443809]

5. Geisler C, Almdal T, Bennedsen J, Rhodes JM, Kolendorf K. Monocyte functions in diabetes mellitus.Acta Pathol Microbiol Immunol Scand [C] 1982;90:33–37.

6. Luger A, Schernthaner G, Urbanski A, Luger TA. Cytokine production in patients with newlydiagnosed insulin-dependent (type I) diabetes mellitus. Eur J Clin Invest 1988;18:233–236. [PubMed:3138125]

7. Ohno Y, Aoki N, Nishimura A. In vitro production of interleukin-1, interleukin-6, and tumor necrosisfactor-alpha in insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1993;77:1072–1077.[PubMed: 8408455]

8. Salvi GE, Collins JG, Yalda B, Arnold RR, Lang NP, Offenbacher S. Monocytic TNF alpha secretionpatterns in IDDM patients with periodontal diseases. J Clin Periodontol 1997;24:8–16. [PubMed:9049792]

9. Netea MG, Demacker PN, Kullberg BJ, Boerman OC, Verschueren I, Stalenhoef AF, van der MeerJW. Increased interleukin-1alpha and interleukin-1beta production by macrophages of low-densitylipoprotein receptor knock-out mice stimulated with lipopolysaccharide is CD11c/CD18-receptormediated. Immunology 1998;95:466–472. [PubMed: 9824512]

10. Zykova SN, Jenssen TG, Berdal M, Olsen R, Myklebust R, Seljelid R. Altered cytokine and nitricoxide secretion in vitro by macrophages from diabetic type II-like db/db mice. Diabetes2000;49:1451–1458. [PubMed: 10969828]

11. Libby P. Inflammation in atherosclerosis. Nature 2002;420:868–874. [PubMed: 12490960]12. Glass CK, Witztum JL. Atherosclerosis. the road ahead. Cell 2001;104:503–516. [PubMed:

11239408]13. Williams KJ, Tabas I. Lipoprotein retention--and clues for atheroma regression. Arterioscler Thromb

Vasc Biol 2005;25:1536–1540. [PubMed: 16055756]14. Cushing SD, Berliner JA, Valente AJ, Territo MC, Navab M, Parhami F, Gerrity R, Schwartz CJ,

Fogelman AM. Minimally modified low density lipoprotein induces monocyte chemotactic protein1 in human endothelial cells and smooth muscle cells. Proc Natl Acad Sci U S A 1990;87:5134–5138. [PubMed: 1695010]

15. Quinn MT, Parthasarathy S, Fong LG, Steinberg D. Oxidatively modified low density lipoproteins:a potential role in recruitment and retention of monocyte/macrophages during atherogenesis. ProcNatl Acad Sci U S A 1987;84:2995–2998. [PubMed: 3472245]

16. Steinberg, D.; Witztum, JL. Lipoproteins, lipoprotein oxidation, and atherogenesis. In: Chien, KR.,editor. Molecular Basis of Cardiovascular Disease. W.B. Sanders Co.; Philadelphia, PA: 1998.

17. Boring L, Gosling J, Cleary M, Charo IF. Decreased lesion formation in CCR2−/− mice reveals arole for chemokines in the initiation of atherosclerosis. Nature 1998;394:894–897. [PubMed:9732872]

18. Gosling J, Slaymaker S, Gu L, Tseng S, Zlot CH, Young SG, Rollins BJ, Charo IF. MCP-1 deficiencyreduces susceptibility to atherosclerosis in mice that overexpress human apolipoprotein B. J ClinInvest 1999;103:773–778. [PubMed: 10079097]

19. Gu L, Okada Y, Clinton SK, Gerard C, Sukhova GK, Libby P, Rollins BJ. Absence of monocytechemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice.Mol Cell 1998;2:275–281. [PubMed: 9734366]

Qiao et al. Page 9

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

20. Veillard NR, Kwak B, Pelli G, Mulhaupt F, James RW, Proudfoot AE, Mach F. Antagonism ofRANTES receptors reduces atherosclerotic plaque formation in mice. Circ Res 2004;94:253–261.[PubMed: 14656931]

21. Braunersreuther V, Zernecke A, Arnaud C, Liehn EA, Steffens S, Shagdarsuren E, Bidzhekov K,Burger F, Pelli G, Luckow B, Mach F, Weber C. Ccr5 but not Ccr1 deficiency reduces developmentof diet-induced atherosclerosis in mice. Arterioscler Thromb Vasc Biol 2007;27:373–379. [PubMed:17138939]

22. Kuziel WA, Dawson TC, Quinones M, Garavito E, Chenaux G, Ahuja SS, Reddick RL, Maeda N.CCR5 deficiency is not protective in the early stages of atherogenesis in apoE knockout mice.Atherosclerosis 2003;167:25–32. [PubMed: 12618265]

23. Potteaux S, Combadiere C, Esposito B, Lecureuil C, Ait-Oufella H, Merval R, Ardouin P, Tedgui A,Mallat Z. Role of bone marrow-derived CC-chemokine receptor 5 in the development ofatherosclerosis of low-density lipoprotein receptor knockout mice. Arterioscler Thromb Vasc Biol2006;26:1858–1863. [PubMed: 16763157]

24. Qiao M, Kisgati M, Cholewa J, Zhu W, Smart EJ, Sulistio MS, Asmis R. Increased expression ofcytosolic and mitochondrial glutathione reductase in macrophages inhibits atherosclerotic lesiondevelopment in LDL receptor-deficient mice. Arterioscler Thromb Vasc Biol 2007;27:1375–1382.[PubMed: 17363688]

25. Reynaert NL, Ckless K, Guala AS, Wouters EF, Van D, V, Janssen-Heininger YM. In situ detectionof S-glutathionylated proteins following glutaredoxin-1 catalyzed cysteine derivatization. BiochimBiophys Acta 2006;1760:380–387. [PubMed: 16515838]

26. Zhao Q, Egashira K, Inoue S, Usui M, Kitamoto S, Ni W, Ishibashi M, Hiasa KK, Ichiki T, ShibuyaM, Takeshita A. Vascular endothelial growth factor is necessary in the development ofarteriosclerosis by recruiting/activating monocytes in a rat model of long-term inhibition of nitricoxide synthesis. Circulation 2002;105:1110–1115. [PubMed: 11877364]

27. Ni W, Egashira K, Kitamoto S, Kataoka C, Koyanagi M, Inoue S, Imaizumi K, Akiyama C, NishidaKI, Takeshita A. New anti-monocyte chemoattractant protein-1 gene therapy attenuatesatherosclerosis in apolipoprotein E-knockout mice. Circulation 2001;103:2096–2101. [PubMed:11319201]

28. Daugherty A, Rateri DL. Presence of LDL receptor-related protein/alpha 2-macroglobulin receptorsin macrophages of atherosclerotic lesions from cholesterolfed New Zealand and heterozygousWatanabe heritable hyperlipidemic rabbits. Arterioscler Thromb 1994;14:2017–2024. [PubMed:7526898]

29. Wang Y, Qiao M, Mieyal JJ, Asmis LM, Asmis R. Molecular mechanism of glutathione-mediatedprotection from oxidized LDL-induced cell injury in human macrophages: Role of glutathionereductase and glutaredoxin. Free Radic Biol Med 2006;41:775–785. [PubMed: 16895798]

30. Jones DP. Redox potential of GSH/GSSG couple: assay and biological significance. MethodsEnzymol 2002;348:93–112. [PubMed: 11885298]

31. Poulter LW, Turk JL. Rapid quantitation of changes in macrophage volume induced by lymphokinein vitro. Clin Exp Immunol 1975;19:193–199. [PubMed: 1106913]

32. Han KH, Tangirala RK, Green SR, Quehenberger O. Chemokine receptor CCR2 expression andmonocyte chemoattractant protein-1-mediated chemotaxis in human monocytes. A regulatory rolefor plasma LDL. Arterioscler Thromb Vasc Biol 1998;18:1983–1991. [PubMed: 9848893]

33. Shelton MD, Mieyal JJ. Regulation by reversible S-glutathionylation: molecular targets implicatedin inflammatory diseases. Mol Cells 2008;25:332–346. [PubMed: 18483468]

34. Swirski FK, Pittet MJ, Kircher MF, Aikawa E, Jaffer FA, Libby P, Weissleder R. Monocyteaccumulation in mouse atherogenesis is progressive and proportional to extent of disease. Proc NatlAcad Sci U S A 2006;103:10340–10345. [PubMed: 16801531]

35. Schafer FQ, Buettner GR. Redox environment of the cell as viewed through the redox state of theglutathione disulfide/glutathione couple. Free Radic Biol Med 2001;30:1191–1212. [PubMed:11368918]

36. Mine S, Okada Y, Tanikawa T, Kawahara C, Tabata T, Tanaka Y. Increased expression levels ofmonocyte CCR2 and monocyte chemoattractant protein-1 in patients with diabetes mellitus. BiochemBiophys Res Commun 2006;344:780–785. [PubMed: 16631114]

Qiao et al. Page 10

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

37. Kanda M, Ihara Y, Murata H, Urata Y, Kono T, Yodoi J, Seto S, Yano K, Kondo T. Glutaredoxinmodulates platelet-derived growth factor-dependent cell signaling by regulating the redox status oflow molecular weight protein-tyrosine phosphatase. J Biol Chem 2006;281:28518–28528. [PubMed:16893901]

38. Aragay AM, Mellado M, Frade JM, Martin AM, Jimenez-Sainz MC, Martinez A, Mayor F Jr.Monocyte chemoattractant protein-1-induced CCR2B receptor desensitization mediated by the Gprotein-coupled receptor kinase 2. Proc Natl Acad Sci U S A 1998;95:2985–2990. [PubMed:9501202]

39. Lombardi MS, Kavelaars A, Penela P, Scholtens EJ, Roccio M, Schmidt RE, Schedlowski M, MayorF Jr. Heijnen CJ. Oxidative stress decreases G protein-coupled receptor kinase 2 in lymphocytes viaa calpain-dependent mechanism. Mol Pharmacol 2002;62:379–388. [PubMed: 12130691]

Qiao et al. Page 11

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Figure 1. Correlation of Aortic Lesion Surface Area with Plasma Cholesterol and MacrophageGlutathione Reduction Potential(A) GSH/GSSG ratios in peritoneal macrophages isolated from LDL-R−/− mice exposed toeither mild (MD, n=8), moderate (HFD, n=8) or severe metabolic stress (STZ+HFD, n=10)and (B) lesion surface area in aortas from these metabolically-stressed LDL-R−/− mice. (C)Correlation between plasma cholesterol and aortic lesion surface area (r2=0.635, P<0.001) and(D) between the glutathione reduction potential (Eh) of peritoneal macrophages and aorticlesion surface area (r2=0.414, P=0.001) in mildly (MD, Ο, n=7), modestly (HFD, ◆, n=7) andseverely metabolically-stressed LDL-R−/− mice (STZ+HFD, ▴, n=9).

Qiao et al. Page 12

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Qiao et al. Page 13

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Figure 2. Lesion Area, Macrophage Content ProteinS-Glutathionylation in the Aortic Root(A) Lesion area in the aortic roots from LDL-R−/− mice exposed to either mild (MD, Ο, n=8),moderate (HFD, ♦, n=8) or severe metabolic stress (STZ+HFD, ▴, n=10) was measured in eight7 μm sections separated by 70 μm after ORO staining. Lesion size is expressed mm2. (B)Macrophage content was assessed in adjacent sections by immunohistochemistry withantibodies directed against macrosialin/CD68 and is expressed as macrophage-containing areain mm2 (see Supplement, Figure S1B for representative images). (C) Representative imagesof three sections taken from three mice per group from the aortic root stained with ORO (upperpanels), and enlarged images of selected areas (white box, upper panels) on adjacent sectionslabeled in situ for S-glutathionylated proteins (red), and stained with a macrophage-specificantiserum directed against CD68 (green) and DAPI (blue) to identify nuclei (lower panels).Colocalization of protein-S-glutathionylation with CD68-positive areas (macrophages), areshown in yellow. Sections in which the glutaredoxin 1-mediated reduction of protein-glutathione mixed disulfides (PSSG) was omitted (w/o Grx1) served as controls for the insitu labeling procedure. Magnification: 600X

Qiao et al. Page 14

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Figure 3. Glomerular Lipid Accumulation and Macrophage Content(A) Lipid accumulation in glomeruli was assessed in ORO-stained cross cryosections ofkidneys isolated from LDL-R−/− mice exposed to either mild (MD, Ο, n=8), moderate (HFD,♦, n=8) or severe metabolic stress (STZ+HFD, ▴, n=10). Results are expressed as percent ORO-positive glomerular area. (B) Macrophage recruitment was assessed in adjacent sections byimmunohistochemistry, with antibodies directed against macrosialin/CD68 (see Supplement,Figure S2 for representative images). Results are expressed as numbers of macrophages perglomerulus.

Qiao et al. Page 15

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Figure 4. Macrophage Chemotactic Activity in Metabolically Stress Mice(A) Matrigel plugs filled with either vehicle (left bar, number of cells given in parentheses) orMCP-1 (300 nM, right bar) were injected into the left and right flanks, respectively, of LDL-R−/− mice exposed to either mild (MD, □ n=5), moderate (HFD, ■, n=5) or severe metabolicstress (STZ+HFD, ■, n=10). After three days the plugs were removed, dissolved and cells werecounted. Immunostaining with antibodies directed against macrosialin/CD68 confirmed that> 93% of the cells recruited into the Matrigel plugs were macrophages. (B) Correlation betweenglutathione reduction potential (Eh) of peritoneal macrophages isolated from mildly (MD, Ο,n=5), modestly (HFD, ♦, n=5) and severely metabolically-stressed LDL-R−/− mice (STZ

Qiao et al. Page 16

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

+HFD, ▴, n=10) and macrophage numbers recruited into Matrigel plugs implanted for 3 daysinto the same animals (r2=0.554, P<0.001).

Qiao et al. Page 17

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

NIH

-PA Author Manuscript

Qiao et al. Page 18

Table 1Blood and Plasma Parameters For LDL-R−/− Mice Exposed to Either Mild (MD), Moderate (HFD) or Severe MetabolicStress (STZ+HFD).

Parameter MD (n=8) HFD (n=8) STZ + HFD (n=10)

Weight (g) 19.8 ± 0.9 25.6 ± 1.2* 21.6 ± 0.8§§Plasma total cholesterol (mg/dl) 223 ± 20 612 ± 53* 784 ± 70*Plasma triglycerides (mg/dl) 41.9 ± 6.4 58.2 ± 7.2 177.3 ± 53.4Glucose (mg/dl) 62.8 ± 0.6 82.0 ± 4.0** 149.4 ± 25.4**Albumin/Creatinine (μg/mg) 2.2 ± 0.6 1.7 ± 0.2 23.9 ± 13.1**SAA (ng/ml) 0.78 ± 0.42 0.89 ± 0.50 2.53 ± 0.89*Leptin (ng/ml) 15.2 ± 2.2 12.6 ± 1.5 9.0 ± 2.2§Insulin (ng/ml) 0.14 ± 0.03 0.14 ± 0.03 0.17 ± 0.05WBC (109/l) 5.7 ± 0.7 7.0 ± 1.4 10.9 ± 1.2**,§Monocytes (109/l) 0.31 ± 0.06 0.26 ± 0.06 0.37 ± 0.06

Results are expressed as mean ± SE for 8 – 10 mice.

Lipid, hormone and SAA measurements were performed in plasma samples from overnight-fasted mice. Glucose was measured in whole blood with aglucometer after overnight fasting. White blood cell (WBC) and monocyte counts were obtained with a VetScan HM II Analyzer and determined in bloodobtained by retro-orbital bleed after 11 weeks of diet feeding.

*: P < 0.05 versus MD

**: P < 0.01 versus MD

§: P < 0.05 versus HFD

§§: P < 0.01 versus HFD.

Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2010 November 1.