hdl of patients with type 2 diabetes mellitus elevates the ... · human cancer biology hdl of...

Human Cancer Biology

HDL of Patients with Type 2 Diabetes Mellitus Elevates theCapability of Promoting Breast Cancer Metastasis

Bing Pan1, Hui Ren1, Yubin He2, Xiaofeng Lv2, Yijing Ma1, Jing Li3, Li Huang3, Baoqi Yu1,Jian Kong4, Chenguang Niu1, Youyi Zhang1, Wen-bing Sun4, and Lemin Zheng1

AbstractPurpose:Epidemiologic studies suggested complicated associationsbetween type2diabetesmellitus and

breast cancer. High-density lipoprotein (HDL) is inversely associated with the risk and mortality of breast

cancer. Our study is to determine the different effects of normal and diabetic HDL on breast cancer cell

metastasis.

Experimental Design:MDA-MB-231 and MCF7 cells were treated with N-HDL, D-HDL, G-HDL, and

Ox-HDL. Cell metastasis potency was examined using a tail-vein injection model, and cell adhesion

abilities to human umbilical vein endothelial cells (HUVEC) and extracellular matrix (ECM) were

determined in vitro. Integrin expression and protein kinase C (PKC) activity were evaluated, and PKC

inhibitor was applied.

Results: D-HDL dramatically promoted cell pulmonary metastasis (103.6% increase at P < 0.001 for

MDA-MB-231 with 1� 105 cell injection; 157.1% increase at P < 0.05 for MCF7with 4� 105 cell injection)

and hepaticmetastasis (18.1-fold increase at P < 0.001 forMCF7with 4� 105 cell injection), and stimulated

higher TC-HUVECs adhesion (21.9% increase at P < 0.001 for MDA-MB-231; 23.6% increase at P < 0.05 forMCF7) and TC-ECM attachment (59.9% and 47.9% increase, respectively, for MDA-MB-231 and MCF7,

both at P < 0.01) compared withN-HDL. D-HDL stimulated higher integrin (b1, b2, b3, andan) expressionon cell surface and induced higher PKC activity. Increased TC-HUVECs and TC-ECM adhesion induced by

D-HDL, G-HDL, and Ox-HDL could be inhibited by staurosporine.

Conclusions:Our study showed that glycation and oxidation of HDL in diabetic patients could lead to

abnormal actions on breast cancer cell adhesion to HUVECs and ECM, thereby promoting metastasis

progression of breast cancer. This will largely draw the attention of HDL-based treatments in the diabetes

patients with breast cancer. Clin Cancer Res; 18(5); 1246–56. �2012 AACR.

IntroductionBoth cancer and diabetes are prevalent diseases, whose

incidence is increasing globally, and have a tremendousimpact onhealthworldwide (1–3). Epidemiologic evidence

suggests that people with diabetes are at a higher risk ofmany forms of cancer (including breast, colon, pancreas,and etc.; ref. 1). Moreover, some studies show that patientsdiagnosed with cancer who have preexisting diabetes are atincreased risk for long-term, all-cause mortality comparedwith those without diabetes (4). In terms of breast cancer,diabetes was associated with a close to 40% increase inmortality within the first 5 years following breast cancer (5).

Metastasis of breast cancer is the major cause of tumor-related morbidity and mortality (6, 7), which involvesmultistep processes and various cytophysiologic changes,including altered intercellular interaction between circulat-ing tumor cells and the endothelial cells of blood vessels,and changed adhesion ability between tumor cells andsubendothelial extracellular matrix (ECM; ref. 8). Theseinteractions are considered as important events duringtumor metastasis (9). Multiple and diverse adhesion mole-cules play a pivotal role in the intercellular and cell–ECMinteractions of cancer (10), such as integrins, the immuno-globulin supergene family, selectins, cadherines, and etc(11). Integrins,which consist of ana subunit noncovalentlylinked to a b subunit, are members of a glycoprotein family

Authors' Affiliations: 1The Institute of Cardiovascular Sciences and KeyLaboratory of Molecular Cardiovascular Sciences, Ministry of Education,Peking University Health Science Center; 2The Department of Cardiovas-cular Medicine and the Department of Endocrinology, the Military GeneralHospital of Beijing; 3The Department of Cardiovascular Medicine, China-Japan Friendship Hospital; and 4Department of Hepatobiliary Surgery,West Campus, Beijing Chao-yang Hospital Affiliated to Capital MedicalUniversity, Beijing, China

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

B. Pan and H. Ren contributed equally to this work.

Corresponding Author: Lemin Zheng, The Institute of CardiovascularSciences and Key Laboratory of Molecular Cardiovascular Sciences,Ministry of Education, Peking University Health Science Center, Beijing100191, China. Phone: 086-010-8280-5452; Fax: 086-010-8280-2769;E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-11-0817

�2012 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 18(5) March 1, 20121246

on February 15, 2021. © 2012 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 18, 2012; DOI: 10.1158/1078-0432.CCR-11-0817

http://clincancerres.aacrjournals.org/

that forms heterodimeric receptors via which cells attach toECM, to the surfaces of each other or to different cell types(12). They have been implicated in a wide variety of cellularfunctions, including cell adhesion, migration, invasion,proliferation, and survival (12, 13). Protein kinase C (PKC)activity has been shown to be important in the regulation ofintegrins expression, localization (14, 15), andactivity (16).High-density lipoprotein (HDL) contributes importantly

to cardiovascular disease risk, with a significant inverserelationship between HDL levels and the risk of cardiovas-cular disease (17–20). Similarly, it has been reported thatHDL levels are inversely related to the rates of incidentcancer, including breast cancer (21–24). However, emerg-ing evidence indicates that HDL can be modified undercertain circumstances, such as in diabetes (25) and cardio-vascular diseases (26), and lose its protective effects or evenbecome proatherogenic (25, 27). HDL isolated frompatients with type 2 diabetes mellitus (T2DM) can bemodified into glycated HDL and oxidized HDL (25, 28)

and exhibits deficient activities in comparing with normalHDLof healthy subjects (25).Due to the difference betweenHDL from healthy subjects and T2DM patients, we specu-late that normal anddiabeticHDLmayhavedifferent effectsin themetastasis of breast cancer cells and finally lead to theincreased mortality of breast cancer in patients withdiabetes.

Materials and MethodsPatient characteristics

Written informed consent was obtained from all partici-pants, and the study protocol was approved by the localethics committee. Each patient volunteer underwent amed-ical history, physical examination, screening laboratorytests, and a 75 g oral glucose tolerance test. Patients werestable on antihypertensive, hypoglycemic, and lipid-lower-ing medications. There was no treatment change during thestudy. Thehealthy subjects hadno family history of diabetesand had normal glucose tolerance. Characteristics of thestudy participants are shown in Table 1.

AnimalsFour-week-old female BALB/c nude mice were obtained

from the animal house, Academy of Military MedicalSciences, China. Throughout the experiments, mice weremaintainedwith free access topellet food andwater. Animalwelfare and experimental procedures were carried out strict-ly in accordance with the related ethical regulations.

Cell lines and cell cultureThehormone-independentMDA-MB-231 andhormone-

dependent MCF7 human breast cancer cell lines were fromCell Resource Center, Institute of Basic Medical Sciences,Chinese Academy of Medical Sciences, and cultured inDulbecco’s modified Eagle’s medium (DMEM; GIBCO)containing 10% FBS (GIBCO) in a humidified incubatorat 37�C with an atmosphere of 5% CO2. Human umbilicalvein endothelial cells (HUVEC) were isolated by collage-nase digestion of umbilical veins from fresh cords (29). Thecells were plated on gelatin-coated culture dishes in endo-thelial cellmedium(Sciencell) consistingof 500mLofbasal

Table 1. Patient characteristics

Healthy controls Patients with T2DM

Characteristic (n ¼ 102) (n ¼ 107) P

Age, y 46 � 19 60 � 13 <0.001Sex, male/female 50:52 49:58 >0.05Fasting glucose, mmol/L 5.10 � 0.56 8.95 � 3.95 <0.001Triglycerides, mmol/L 1.05 � 0.35 1.56 � 0.95 <0.01Total cholesterol, mmol/L 4.65 � 0.75 4.77 � 1.34 >0.05HDL-C, mmol/L 1.36 � 0.30 1.22 � 0.31 <0.01LDL-C, mmol/L 2.55 � 0.60 2.94 � 1.18 <0.01

NOTE: Values are expressed as mean � SD.Abbreviations: HDL-C, HDL cholesterol; LDL-C, LDL cholesterol.

Translational RelevanceOur study has shown that modifications (such as

glycation and oxidation) are associated with the abnor-mal functions of diabetic high-density lipoprotein(HDL) in type 2 diabetes on promoting breast cancercell metastasis potency in vivo and the abilities of adhe-sion to human umbilical vein endothelial cells andextracellular matrix in vitro, involving integrin upregula-tion and protein kinase C activation in the process. Thisprovides substantial supporting evidence that type 2diabetes is closely related to breast cancer development,especially the metastasis progression, which may bepartly due to the modifications of HDL in the diabeticpatients. It is suggested that diabetic HDL can be con-sidered as potential risk markers and targets in thetreatment of breast cancer with type 2 diabetes and thatprospective HDL-based strategies for therapy are inurgent need.

Diabetic HDL Promotes Breast Cancer Metastasis

www.aacrjournals.org Clin Cancer Res; 18(5) March 1, 2012 1247




medium, 25 mL of FBS, 5 mL of endothelial cell growthsupplement, and 5 mL of penicillin/streptomycin solutionand then cultured in a humidified incubator at 37�C withan atmosphere of 5% CO2. HUVECs were used at passages2 to 5.

Isolation, glycation, and oxidation of HDLFresh, fasting plasma was separated by centrifugation

from peripheral blood obtained from healthy subjects(n ¼ 102) and type 2 diabetic patients (n ¼ 107). Low-density lipoprotein (LDL; 1.019–1.063) and HDL (1.063–1.210) were isolated from fresh plasma by ultracent-rifugation as previously described (30). HDL from eachindividual were dialyzed against 3 � 1 L of endotoxin-freePBS (pH ¼ 7.4) and 100 mmol/L diethylenetriamine pen-taacetic acid (Sigma), sterilized with 0.22-mm filter, storedin sealed tubes at 4�C in dark and used within 2months. Tocreate glycated HDL, fresh human HDL (5 mg protein) wasincubated with 25 mmol/L glucose in PBS under sterileconditions for1weekat37�C(referred toasG-HDL; ref. 31).Glycation of the apoA-I components of HDL was measuredusing mass spectrometry. Oxidative modification of HDLwas done by dialysis at lipoprotein concentration of 0.8 mgprotein/mL against PBS containing 5 mmol/L CuSO4 for 24hours at 37�C (referred to as Ox-HDL; ref. 32). Lipidperoxidation of normal, diabetic, and oxidized HDL wasdetermined by thiobarbituric acid reactive substances(TBARS; Nanjing Jiancheng Bioengineering Institute, Nanj-ing, China) kit which quantifies the malondialdehyde(MDA) content. TBARS were expressed as nanomoles ofMDA per milligram apoA-I.

Tail vein metastasis assayMDA-MB-231 and MCF7 cells were pretreated with nor-

mal, diabetic, glycated, and oxidized HDL for 24 hours. Toproduce experimental metastasis, BALB/c nude mice wereinjected intravenously withMDA-MB-231 cells (1� 105) orMCF7 cells (1 � 105 or 4 � 105) in 0.2 mL DMEM via tailveins. After 20 days, themicewere sacrificed, their lungs andlivers were resected and photographs were taken after fix-ation in buffered formalin. The numbers of metastaticnodules on the surface of the lungs and livers were counted.

HistologyTissueswere fixed in buffered formalin for 24 to 48 hours.

After washing in fresh PBS, fixed tissues were processed andembedded in paraffin. Sections (5 milli micron) were col-lected on microscope slides, deparaffinized, and stainedwith hematoxylin and eosin (H&E) stain. The images werecaptured at 10� magnification.

TC-HUVECs adhesion assayTC-HUVECs adhesion was measured using rose Bengal

stain as previously describedmethod (33). Briefly, HUVECswere seeded onto 96-well plates at a density of 5� 104 cellsper well. Then tumor cells (pretreated with or without HDLfor 6 hours) were plated (5 � 104 cells per well) andincubated for 30 minutes at 37�C. Then unattached cells

were gently washed twicewith 10%FBS-containingDMEM,and 100mLof 0.25% rose Bengal (Sigma)was added. After 5minutes, cells were gently washed twice and then 200 mLethanol/PBS (1:1)was added to eachwell. After 30minutes,the absorbance at 570 nm was recorded. Five parallel wellswere set up for each group.

TC-ECM adhesion assayCell adhesion was measured using MTT assay as previ-

ously described method (34). Briefly, 96-well flat bottomplates were coated with 2 mg per well basement membranematrix (Matrigel; BD) for an hour at 37�C, then blockedwith 2% bovine serum albumin (BSA) for 2 hours at 37�Cfollowed by washing twice. Tumor cells seeded on 12-wellplate were treated with serum-free DMEM alone or HDL(normal, diabetic, glycated, or oxidized) for 6 hours. Afterdetachment with trypsin, the cells (1 � 104 per well) wereplated on ECM-precoated 96-well plates and then washedtwice with PBS to remove nonadherent cells after 45 min-utes. MTT colorimetric assays at 490 nm were employed tomeasure the absorbance of adhesive cells. Four parallelwells were set up for each group.

Antibody-blocking experimentsTo determine the contribution of individual integrins to

the interaction of MDA-MB-231 and MCF7 cells withHUVECs and ECM, monoclonal antibodies (mAb) againstintegrin b1, integrin b2, integrin b3, and integrin av(Abcam) were applied. Human polyclonal antibody of IgG(Boster) was served as negative control. The detached cellswere treatedwith 5 mg/mLmAbs for 30minutes at 37�Candthen allowed to adhere to HUVECs and ECM for 30 min-utes. Five parallel wells were set up for each group.

Determination of cell surface expression of adhesionmolecules by enzyme immunoassay

Cells in 96-well plates were grown till 70% confluency,then treated with HDL (normal, diabetic, glycated, or oxi-dized) in serum-freemedium for 6hours at 37�C.CellswerewashedwithPBS andfixedwithmethanol for 5minutes andthen washedwith PBS 3 times and blockedwith 2%BSA for2 hours at 37�C. Cell surface expression of adhesion mole-cules were determined by primary binding with specificmAbs for integrin b1, integrin b2, integrin b3, and integrinav (1:200; Abcam), followed by secondary binding with ahorseradish peroxidase–conjugated antibody (1:3,000;Boster) as described previously (35). Quantification wasdone by determination of colorimetric conversion at anoptical density (OD) at 450 nm of 3,3,5,5-tetramethylben-zidine using TMB peroxidase EIA substrate kit (Bio-Medinnovation).

PepTag assay for nonradioactive detection of PKCactivity

The PKC activity in the MDA-MB-231 and MCF7 celllysates was determined by nonradioactive detection kit ofprotein kinase (PepTag Corporation), following the man-ufacturer’s instructions. Briefly, PKC in the HDL-treated cell

Pan et al.

Clin Cancer Res; 18(5) March 1, 2012 Clinical Cancer Research1248




lysates was separated by column of DEAE cellulose and wasincubated with the brightly colored, fluorescent peptidesubstrates. Then phosphorylation by PKC of its specificsubstrate alters the net charge of peptide from þ1 to �1,which allows the phosphorylated and nonphosphorylatedversions of the substrate to be rapidly separated on anagarose gel.

Statistical analysisThe results of multiple observations are presented as the

means� SD and as a representative result of 2 or 3 differentseparate experiments, unless otherwise stated. Data wereanalyzed using Student t test and ANOVA test and valueswere considered significant at P < 0.05.

ResultsParticipantsBaseline demographic and clinical characteristics of

healthy subjects and patients with T2DM are shownin Table 1. Participants were included in the study betweenMay 2009 and Jul 2011.

Pretreatmentwithdiabetic, glycated, andoxidizedHDLpromotes metastasis of MDA-MB-231 and MCF7 cellsWe previously found that diabetic HDL has elevated

ability in promoting breast cancer cells proliferation,migra-tion, and invasion comparedwithnormalHDL in vitro (36),so we speculated that diabetic HDL may promote breastcancer cell metastasis progression. To test our hypothesis,we first checked themetastatic ability of HDL-treatedMDA-MB-231 andMCF7 cells in nudemice by tail vein injection,and 6 mice were set for each group (Fig. 1A). Due to thehigher levels of glycation and oxidization, as we previouslydetermined in the diabetic HDL (36), we also tested gly-cated and oxidized HDL to partly mimic HDL in diabetesindividuals. Twenty days after the injectionof 1�105MDA-MB-231 cells, it was observed that D-HDL, G-HDL, andOx-HDL significantly enhanced the lung homing of MDA-MB-231 cells as counted by the number of colonies on thesurface of whole lung, with 84.3%, 64.2%, and 71.6%increase as compared with control (all at P < 0.001) andwith 103.6%, 81.4%, and 89.6% increase as compared withN-HDL-treated cells (all at P < 0.001). By contrast, N-HDLreduced metastasis by 9.5% as compared with control,though no statistic significance was found (Fig. 1B). How-ever, no visible nodules on the livers were found in themicetreated with 1 � 105 MDA-MB-231 cells (data not shown).In addition, no pulmonary and hepatic metastasis wasshown in the groups injected with 1� 105MCF7 cells (datanot shown). Then 4 � 105 MCF7 cells were injected forfurther study. It is suggested that D-HDL, G-HDL, and Ox-HDL significantly promoted both pulmonary and hepaticmetastasis of MCF7 cells. The nodules on the lungs ofD-HDL, G-HDL, and Ox-HDL groups increased by111.8%, 168.6%, and 149.1% as compared with control(P < 0.05 for D-HDL; P < 0.01 for G-HDL andOx-HDL) andby 157.1%, 226.1%, and 202.4% as compared with N-HDL

group (P < 0.05 for D-HDL, P < 0.001 for G-HDL, P < 0.01for Ox-HDL; Fig. 1C); the nodules on the livers of D-HDL,G-HDL, and Ox-HDL groups increased by 275.3%,229.2%, and 320.2% as compared with control (all atP < 0.01) and by 18.1-, 15.7-, and 20.4-fold as comparedwith N-HDL group (P < 0.01 for G-HDL; P < 0.001 for D-HDL and Ox-HDL; Fig. 1D). On the contrary, normalHDL reduced metastasis of MCF7 cells by 7.1% in thelung and 95.9% in the liver as compared with control,though without statistic significance. Consistent resultswere obtained in the H&E-stained sections (Fig. 1E).Weight of mice during the study was shown (Supplemen-tary Fig. S1A and S1B).

The diabetic, glycated, and oxidized HDL-treatedMDA-MB-231 and MCF7 cells have increased capacities ofadhesion to HUVECs and attachment to ECM

Tumor cells adhesion to vascular endothelial cells andattachment to ECM are considered as important steps in themetastatic processes of malignant tumor cells. Therefore,corresponding experiments were carried out. Adhesion ofMDA-MB-231 (hormone independent) and MCF7 (hor-mone dependent) cells pretreated with diabetic, glycated,and oxidized HDL, but not pretreated with normal HDL, toboth HUVECs (Fig. 2A) and ECM (Fig. 2B) markedlyincreased. In the TC-HUVECs adhesion assay, the adhesionof MDA-MB-231 cells pretreated with D-HDL, G-HDL, andOx-HDL increased by 24.4%, 32.8%, and 32.1% as com-pared with control, respectively (all at P < 0.001), andincreased by 21.9%, 30.2%, and 29.5% as compared withN-HDL, respectively (all at P < 0.001); the adhesion ofMCF7 cells pretreated with D-HDL, G-HDL, and Ox-HDLincreased by 22.7%, 31.0%, and 32.6% as compared withcontrol, respectively (all at P < 0.01), and increased by23.6%, 31.9%, and 33.6% as compared with N-HDL,respectively (P < 0.05 for D-HDL; P < 0.01 for G-HDL andOx-HDL). In the TC-ECMadhesion assay, the attachment ofMDA-MB-231 cells treated with D-HDL, G-HDL, and Ox-HDL has increased by 33.7%, 40.3%, and 39.0% as com-pared with control, respectively, without statistical signifi-cance and have increased by 59.9%, 67.8%, and 66.2% ascompared with N-HDL, respectively (all at P < 0.01); theattachment of MCF7 cells pretreated with D-HDL, G-HDL,and Ox-HDL increased by 29.4%, 22.1%, and 29.0% ascompared with control, respectively, without statisticalsignificance, and increased by 47.9%, 39.6%, and 47.4%as comparedwithN-HDL, respectively (P < 0.05 forG-HDL;P <0.01 forD-HDL andOx-HDL). By contrast, normalHDLreducedMDA-MB-231 andMCF7 cells adhesion to ECMby16.4% and 12.5%, respectively, as compared with control,though no statistic significance was found. Furthermore,HDL from 6 normal controls and 6 diabetic patients at 100mg/mL apoA-I were used for confirmation. It was suggestedthat normalHDLhardly stimulateMDA-MB-231 andMCF7cells adhesion to both HUVECs and ECM, whereas diabeticHDL substantially enhanced the 2 adhesion capabilities.The adhesion of MDA-MB-231 and MCF7 cells which wereinduced by diabetic HDLwas 40.6% (P < 0.001) and 20.5%






Figure1. Effects of thedifferent typesofHDLon thepulmonary andhepaticmetastasis of humanbreast cancer cells. A,MDA-MB-231andMCF7 cellswere leftuntreated (C), or treated with normal (N), diabetic (D), glycated (G), and oxidized (Ox) HDL. Control and HDL-treated cells (1� 105 for MDA-MB-231 cells and4 � 105 for MCF7 cells) were intravenously injected into BALB/c nude mice via tail vein. After 20 days, lungs and livers were resected and analyzed formetastasis. Representative pictures of lungs and livers are shown. B–D, quantitative evaluation of macroscopically detectable metastasis nodules on thesurface of the whole lungs and livers. Data are expressed as mean � SD. (�, P < 0.05; ��, P < 0.01; ��, P < 0.001 by one-way ANOVA). E, representativehistologic photomicrographs of lung and liver tissue sections stained with H&E (10�). Arrows indicate tumor islands.

Pan et al.





(P < 0.01) higher, respectively, in the TC-HUVECs adhesionassay (Fig. 2C), and the attachment of MDA-MB-231 andMCF7 cells induced by diabetic HDL was 27.6% (P < 0.01)and 23.5% (P < 0.05) higher, respectively, in the TC-ECMadhesion assay (Fig. 2D).

Anti-integrin antibodies inhibit MDA-MB-231 andMCF7 cell adhesion to HUVECs and to ECMBlocking antibodies to integrins were used to determine

the roles of integrinb1, integrinb2, integrinb3, and integrinan in the adhesion of MDA-MB-231 and MCF7 cells toHUVECs and ECM. Adhesion of MDA-MB-231 and MCF7cells to HUVECs (Fig. 3A) and ECM (Fig. 3B) was signifi-cantly inhibited in the presence of anti-integrin antibodies.The adhesion of untreated cells was considered as 100%.The adhesion of MDA-MB-231 cells treated with mAbs ofintegrin b1, integrin b2, integrin b3, and integrin an to

HUVECs was (44.51 � 19.25)%, (62.09 � 4.75)%, (45.43�18.48)%, and (43.24�29.15)%atP¼0.004,P¼0.0031,P ¼ 0.0038 and P ¼ 0.0135, respectively; the adhesion ofMCF7 cells treated with mAbs of integrin b1, integrin b2,integrin b3, and integrin an to HUVECs was (68.91 �6.59)%, (59.97 � 16.08)%, 68.49 � 14.29)%, and(71.25 � 5.75)% at P ¼ 0.0472, P ¼ 0.019, P ¼ 0.0395and P ¼ 0.0292, respectively. The adhesion of MDA-MB-231 cells treated with mAbs of integrin b1, integrin b2,integrin b3, and integrin an to ECM was (86.56 � 5.92)%,(83.42 � 10.62)%, (76.37 � 8.29)% and (87.33 � 7.71)%at P ¼ 0.0092, P ¼ 0.0264, P ¼ 0.0021 and P ¼ 0.026,respectively; the adhesion of MCF7 cells treated with mAbsof integrin b1, integrin b2, integrin b3, and integrin an toECM was (71.53 � 5.71)%, (69.77 � 10.10)%, (67.56 �8.25)%, and (79.25 � 14.52)% at P ¼ 0.0024, P ¼ 0.0038,P ¼ 0.0017 and P ¼ 0.0468, respectively.

Figure 2. Diabetic, glycated, andoxidized HDL promote more MDA-MB-231andMCF7breast cancer celladhesion to HUVECs andattachment to ECM. A, MDA-MB-231 and MCF7 cells were pretreatedwith normal, diabetic, glycated, andoxidized HDL at 100 mg/mL apoA-Ifor 6 hours and then seeded ontoHUVEC-coated 96-well plate for 30minutes. Relative cell adhesion wasdetermined by Rose Bengal assayand expressed as percentage ofHDL-treated cells in comparisonwithcontrol. Data are expressed as mean�SDwith 5parallel wells. (�,P < 0.05;��,P<0.01; ��,P<0.001byone-wayANOVA). B,MDA-MB-231andMCF7cells were pretreated with normal,diabetic, glycated, and oxidized HDLat 100 mg/mL apoA-I for 6 hours andthen plated ontoMatrigel-coated 96-well plate for 30minutes. Relative celladhesion was determined by MTTassay and expressed as percentageof HDL-treated cells in comparisonwith control. Data are expressed asmean � SD with 4 parallel wells.(�, P < 0.05; ��, P < 0.01 by one-wayANOVA). C, MDA-MB-231 andMCF7cellswerepretreatedwithHDLfrom 6 normal subjects and 6patients with type 2 diabetes for 6hours at 100 mg/mL apoA-I. Celladhesion to HUVEC stimulated bynormal HDL versus diabetic HDL isshown. (��,P < 0.01; ��, P < 0.001 byStudent t test). D, MDA-MB-231 andMCF7cellswerepretreatedwithHDLfrom 6 normal subjects and 6patients for 6 hours at 100 mg/mLapoA-I. Cell adhesion to ECMstimulated by normal HDL versusdiabetic HDL is shown. (�, P < 0.05;��, P < 0.01 by Student t test).






Diabetic, glycated, and oxidized HDL induce higherintegrin expression on the MDA-MB-231 andMCF7 cellsurface

Integrins have been reported to play a key role in TC-HUVECs and TC-ECM adhesion. Diabetic, glycated, andoxidized HDL induced elevated expressions of integrin b1(Fig. 4A), integrin b2 (Fig. 4B), integrin b3 (Fig. 4C), andintegrin av (Fig. 4D) compared with normal HDL on theMDA-MB-231 andMCF7 cell surface. Specifically, forMDA-MB-231 cells, integrin b1 was increased by 21.1%, 22.2%,and 24.5%, respectively (all at P < 0.001); integrin b2 wasincreased by 21.2%, 19.7%, and 27.9%, respectively (all atP < 0.001); integrin b3 was increased by 20.0%, 21.8%, and19.3%, respectively (all at P < 0.01); integrin av wasincreased by 18.7%, 17.3%, and 18.1%, respectively (allat P < 0.01); and for MCF7 cells, integrin b1 was increasedby 30.1%, 27.8%, and 41.1%, respectively (P < 0.05 forG-HDL, P < 0.01 for D-HDL and Ox-HDL); integrin b2

was increased by 24.3%, 18.4%, and 17.1%, respectively (P< 0.001 for D-HDL, P < 0.01 for G-HDL and Ox-HDL);integrin b3 was increased by 25.5%, 23.1%, and 27.1%,respectively (all at P < 0.001); integrin av was increased by35.6%, 40.6%, and 52.1%, respectively (all at P < 0.001).Furthermore, 6 normal and 6 diabetic HDL samples at 100mg/mL apoA-I were used for confirmation. It was suggestedthat normal HDL hardly stimulated integrins expression,whereas diabetic HDL induced higher integrins expressionboth on the hormone-independent and hormone-depen-dent human breast cancer cells. For MDA-MB-231 cells,diabeticHDL induced 17.4%higher integrinb1 (P<0.001),33.5%higher integrin b2 (P< 0.001), 23.4%higher integrinb3 (P < 0.001), and 25.5% higher integrin av (P < 0.01)compared with normal HDL, respectively; for MCF7 cells,diabetic HDL stimulated 30.3% higher integrin b1 (P <0.01), 17.5% higher integrin b2 (P < 0.001), 25.5% higherintegrin b3 (P < 0.001), and 17.6% higher integrin av (P <0.001) compared with normal HDL, respectively (Fig. 4E).

Diabetic, glycated, and oxidized HDL-inducedMDA-MB-231 and MCF7 cell metastasis involvesPKC pathway

It is reported that PKC activity is associated with theregulation of integrin-related adhesion. Therefore, PKCactivity induced by normal, diabetic, glycated, and oxidizedHDL was examined. It was shown that diabetic, glycated,andoxidizedHDL stimulatedhigher PKCactivity comparedwith normal HDL both in MDA-MB-231 and MCF7 cells(Fig. 5A). So far, the results indicated that diabetic, glycated,and oxidized HDL may promote MDA-MB-231 and MCF7cell metastasis via PKC pathway. Therefore, MDA-MB-231and MCF7 cells were treated with staurosporine, a PKCinhibitor, to determine whether the suppression of the PKCpathway would result in the inhibition of breast cancer celladhesion promoted by diabetic, glycated, and oxidizedHDL. Staurosporine presented to be powerful in inhibitingHDL-induced (D-HDL,G-HDL, andOx-HDL) breast cancercell adhesion to HUVECs (Fig. 5B) and to ECM (Fig. 5C).After pretreatment with staurosporine, MDA-MB-231 andMCF7 cell adhesion to both HUVECs and ECM was sub-stantially reduced, and no statistical significance was foundamong the groups of staurosporine-treated cells in both ofthe adhesion assays. It was suggested that the increasedabilities of diabetic, glycated, and oxidized HDL in promot-ing MDA-MB-231 andMCF7 cell adhesion to HUVECs andECMweremainly owing to the increased PKC activity. Thennext, the inhibitory effects of staurosporine on integrinsexpression on theMDA-MB-231 andMCF7 cell surfacewereexamined. Integrin b1 (28.41%decrease, P < 0.01 forMDA-MB-231 cells; 26.49% decrease, P < 0.01 for MCF7 cells),integrin b2 (33.27% decrease, P < 0.01 for MDA-MB-231cells; 57.57% decrease, P < 0.001 for MCF7 cells), integrinb3 (15.15% decrease, P < 0.01 for MDA-MB-231 cells;57.11% decrease, P < 0.001 for MCF7 cells), and integrinan (18.36% decrease, P < 0.01 for MDA-MB-231 cells;22.47% decrease, P < 0.05 for MCF7 cells) were all signif-icantly suppressed by staurosporine (Fig. 5D).

Figure 3. Anti-integrin antibodies inhibit MDA-MB-231 and MCF7 breastcancer cell adhesion to HUVECs and attachment to ECM. MDA-MB-231and MCF7 cells were incubated with no antibody, negative controlantibody, or anti-integrin (b1, b2, b3, and av) antibodies for 30 minutes at37�C before the adhesion to HUVECs (A) and attachment to ECM (B).Antibody-blocked groups were compared with control (antibody-untreatedgroups). Data are expressedasmean�SDwith 5parallel wells.(�, P < 0.05; ��, P < 0.01 by Student t test).

Pan et al.





DiscussionPrevious study suggests complex associations between

T2DM and breast cancer (37). Diabetes may have directbiologic effects on breast cancer risk, clinical and pathologiccharacteristics and outcome. Moreover, certain antidiabetictherapies may have direct activity against breast cancer.Diabetes may also affect breast cancer outcome indirectlyand has been shown to influence medical decision makingabout screening and management of breast cancer (37).In addition, some studies pointed that it is possible thatcancer is well known to be a proinflammatory state,whereas the decreased level of HDL is inefficient to exertits anti-inflammatory effects, which allows tumor cell pro-liferation, survival, and migration (38, 39). Our studyprovides a novel view and shows the significant differencebetween diabetic and normal HDL on breast cancer cellmetastasis progression.

We previously found that HDL in T2DM can beglycated and oxidized, and diabetic HDL was found topromote proliferation, migration, and invasion of breastcancer cells in vitro as compared with normal HDL (36).Glycated and oxidized HDL produced in vitro were usedto partly mimic diabetic HDL. Diabetic, glycated, andoxidized HDL could induce higher synthesis and secre-tion of VEGF-C, matrix metalloproteinase (MMP)-2, andMMP-9 from MDA-MB-231 cells (36). It was indicatedthat diabetic, glycated, and oxidized HDL promoteMDA-MB-231 cell migration and invasion throughextracellular signal–regulated kinase (ERK) and p38mitogen-activated protein kinase (MAPK) pathways,and Akt pathway plays an important role as well inMDA-MB-231 cell invasion (36). It drove us to speculatethat diabetic HDL may have elevated ability of promot-ing breast cancer cell metastasis.

Figure 4. Diabetic, glycated, andoxidized HDL are more efficient instimulating MDA-MB-231 andMCF7cell surface integrins expression.MDA-MB-231 and MCF7 cells weretreated with normal, diabetic,glycated, and oxidized HDL at 100mg/mL apoA-I for 6 hours. Integrin b1(A), integrin b2 (B), integrin b3 (C), andintegrin av (D) levels on the cellsurfaceweremeasured by cell ELISAwith 5 parallel wells. Results wereexpressed as percentage ofHDL-treated cells in comparisonwithcontrol and data are expressed asmean � SD. (�, P < 0.05; ��, P < 0.01;��,P<0.001byone-wayANOVA). E,MDA-MB-231 and MCF7 cells weretreated with 6 normal and 6 diabeticHDL for 6 hours at 100 mg/mLapoA-I.Relative integrin expression on thecell surface stimulated by normalHDL versus diabetic HDL is shown.(��,P <0.01; ��,P<0.001byStudentt test).






In this study, we have shown that diabetic, glycated, andoxidized HDL, which were compared with normal HDL,could promote breast cancer cells metastasis in vivo and celladhesion to HUVECs and ECM in vitro in similar ways andcould inducemuchhigher expressionof integrins on the cellsurface and increase the activity of PKC as well. Moreover,we observed that staurosporine, a PKC inhibitor, inhibitedbreast cancer cell adhesion to HUVECs and ECM whichwere induced by diabetic, glycated, and oxidized HDL, aswell as the expression of integrins. The present data showedthat the increased capabilities of diabetic, glycated, andoxidized HDL in promoting breast cancer cell adhesion toHUVECs and ECM is mainly due to the elevated PKCactivity. The activated PKC could in turn stimulate secre-tions of integrin b1, integrin b2, integrin b3, and integrinan, which are of vital importance in promoting breast

cancer cell metastasis. Our study used a tail-vein injectionmodel to investigate the effects of normal, diabetic, gly-cated, and oxidized HDL on breast cancer cell metastasis invivo. A spontaneous metastatic model can be furtheremployed to study the effects of different forms of HDLon breast cancer development in nude mice, includingtumor growth and metastasis (40). In addition, fluores-cence imaging techniques can be applied to visualize theinteraction of breast cancer cells with vasculature and ECMin nude mice, including tumor cell mobility, invasion,metastasis, and angiogenesis (41, 42). All the studies werecarried out on both hormone-independent (MDA-MB-231)and hormone-dependent (MCF7) cell lines. We found thatMDA-MB-231 cells presented to have stronger abilities ofmetastasis both in vivo and in vitro for the reason that MCF7cells seldom have lung or liver homing after injected with

Figure 5. Diabetic, glycated, andoxidized HDL stimulate MDA-MB-231 and MCF7 cell adhesion toHUVECs and to ECM involvingPKC pathway. A, MDA-MB-231and MCF7 cells were treated withnormal, diabetic, glycated, andoxidized HDL, and then cell lysateswere subjected to the PepTagnonradioactive PKC assay. B andC, MDA-MB-231 and MCF7 cellswere pretreated with 5 nmol/Lstaurosporine, a PKC inhibitor, for3 hours and then coincubated withnormal, diabetic, glycated, andoxidized HDL at 100 mg/mL apoA-I,respectively, for 6 hours. Celladhesion to HUVECs (B) andattachment to ECM (C) weremeasured. (�, P < 0.05; ��, P < 0.01;��, P < 0.001 by Student t test).D, MDA-MB-231 and MCF7 cellswere pretreated with 5 nmol/Lstaurosporine for 6 hours and thenintegrins expression on the cellsurface was measured by cellELISA. (�, P < 0.05; ��, P < 0.01;��, P < 0.001 by Student t test).

Pan et al.





1� 105 cells and that the absolute OD value of the adhesiveMDA-MB-231 cells in both TC-HUVECs and TC-ECM assaywas higher than that of MCF7 cells (data not shown). It hasbeen reported that the effects of HDL on breast cancer cellproliferation have a higher response in the hormone-inde-pendent cells such asMDA-MB-231 cell line (43). However,our study observed similar effects of diabetic, glycated, andoxidized HDL on the cell adhesion to HUVECs and to ECMbetween hormone-independent (MDA-MB-231) cells andhormone-dependent (MCF7) cells compared with normalHDL.HDL plays an extremely important role of protecting

the cardiovascular system from atherosclerosis, includingreversing cholesterol transport, anti-inflammatory, anti-oxidant properties, antithrombotic, and etc. For instance,HDL can mediate reverse cholesterol transport through itsreceptors of SR-BI and ABCAI (44). Also, HDL is potentialcarrier of enzymatic proteins, such as paraoxonase andplatelet-activating factor-acetyl hydrolase, which canhydrolyze phospholipid peroxides and cholesteryl esterperoxides (44). In addition, HDL may inhibit adhesionmolecule expression through SR-BI and S1P receptors(especially S1P-1) when an inflammatory cytokine suchas TNF-a is a proatherogenic stimulant (45). Further-more, a recent study found that HDL can mediate signaltransduction of ERK1/2 and Akt by an ABCA1-dependentmechanism, leading to increased proliferation and migra-tion of prostate cancer cells (46). However, HDL is foundto be modified in many ways under some diseases andsuch modification could impair the cardiovascular-pro-tective abilities of HDL. For example, glycated HDL maylead to the deterioration of vascular function throughaltered production of reactive oxygen species (ROS) andreactive nitrogen species in endothelial cells (47); oxi-dized HDL induces a dose-dependent increase in ROSproduction and elicits a marked increase in the activationof the NF-kB pathway in the endothelial cells, a key playerin the inflammatory response (48). Furthermore, oxi-dized HDL can promote the activation of a network ofintracellular kinases, including ERK1/2 and p38 MAPK, inthe endothelial cells (48). In this study, we found thatglycated and oxidized HDL, as well as diabetic HDL, haveincreased capability of promoting breast cancer metasta-sis through upregulating the PKC pathway–related integ-rins expression. All the experimental evidence allows us tospeculate that the promoting effects of diabetic HDL on

breast cancer cell metastasis, at least partially, may beattributed to the glycation and oxidation of HDL. Owingto the complicated composition of HDL, it needs furtherintensive research to determine the altered parts of HDL(both in quality and in quantity) in diabetic patients andhow they act in different ways to advance breast cancerdevelopment.

Evidence from epidemiologic studies also shows thatsome pathologic alterations including changes of lipidsprofile levels (22, 49) in T2DM could raise higher risk ofbreast cancer (43). Notably, low level of HDL, frequentlypresented in the patients with T2DM and metabolic syn-drome, is strongly linked with breast cancer (49, 50). Ourstudy here provided supporting biologic evidence (Supple-mentary Fig. S2A and S2B) in vitro that low concentration ofnormal HDL could promote MDA-MB-231 breast cancercell capacities of adhesion to HUVECs and attachment toECM,whereas high concentration of normalHDL is slightlyinhibitory.

All the evidence substantially supported the concept thatT2DM is closely related to breast cancer, which could at leastbe partially attributed to the alterations of HDL in T2DM,not only to the declined quantity but also to the alteredstructures and compositions. Our study shows that themodifications of HDL in the diabetic patients could leadto accelerated breast cancermetastasis progression. Thiswillbring more attention on the therapeutic strategies based onHDL function, especially in the patients of diabetes withbreast cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Grant SupportThis project was supported by grant 2010CB912504 and 2011CB503900

from "973" National S&T Major Project; by grant 30900595, 30821001,81172500, 81170101 from the National Natural Science Foundation ofChina; by grant 7102104, 7122106 from the Natural Science Foundation ofBeijing; by grant 2009J01112 from theNatural Science Foundation of Fujian,and by grant 70103-007 from Peking University Health Science Center. ThisProject was sponsored by the programs for "The Returned Overseas ChineseScholars" and "New Century Excellent Talents in Universities" and grant20090001120039 and20100001120034, State EducationMinistry ofChina.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

ReceivedMarch 28, 2011; revisedDecember 23, 2011; acceptedDecember30, 2011; published OnlineFirst January 18, 2012.

References1. Giovannucci E, Harlan DM, Archer MC, Bergenstal RM, Gapstur SM,

Habel LA, et al. Diabetes and cancer: a consensus report. CA Cancer JClin 2010;60:207–21.

2. Jemal A, Siegel R, Ward E, Murray T, Xu J, ThunMJ. Cancer statistics,2007. CA Cancer J Clin 2007;57:43–66.

3. Boyle JP, Thompson TJ, Gregg EW, Barker LE, Williamson DF. Pro-jection of the year 2050 burden of diabetes in the US adult population:dynamic modeling of incidence, mortality, and prediabetes preva-lence. Popul Health Metr 2010;8:29.

4. Barone BB, Yeh HC, Snyder CF, Peairs KS, Stein KB, Derr RL, et al.Long-term all-cause mortality in cancer patients with preexisting

diabetes mellitus: a systematic review and meta-analysis. JAMA2008;300:2754–64.

5. Lipscombe LL, Goodwin PJ, Zinman B, McLaughlin JR, Hux JE. Theimpact of diabetes on survival following breast cancer. Breast CancerRes Treat 2008;109:389–95.

6. Wittekind C, Neid M. Cancer invasion and metastasis. Oncology2005;69 Suppl 1:14–6.

7. GuptaGP,Massagu�e J. Cancermetastasis: building a framework. Cell2006;127:679–95.

8. Lee YB, Ko KC, Shi MD, Liao YC, Chiang TA, Wu PF, et al. Alpha-Mangostin, a novel dietary xanthone, suppresses TPA-mediated






MMP-2 andMMP-9 expressions through theERKsignalingpathway inMCF-7 human breast adenocarcinoma cells. J Food Sci 2010;75:H13–23.

9. Geng JG. Interaction of vascular endothelial cells with leukocytes,platelets and cancer cells in inflammation, thrombosis and cancergrowth and metastasis. Acta Pharmacol Sin 2003;24:1297–300.

10. Nizamutdinova IT, Lee GW, Lee JS, Cho MK, Son KH, Jeon SJ, et al.Tanshinone I suppresses growth and invasion of human breast cancercells, MDA-MB-231, through regulation of adhesion molecules.Carcinogenesis 2008;29:1885–92.

11. Hynes RO, Lander AD. Contact and adhesive specificities in theassociations, migrations, and targeting of cells and axons. Cell1992;68:303–22.

12. RathinamR,Alahari SK. Alahari important role of integrins in the cancerbiology. Cancer Metastasis Rev 2010;29:223–37.

13. CohenMB, Griebling TL, Ahaghotu CA, Rokhlin OW, Ross JS. Cellularadhesion molecules in urologic malignancies. Am J Clin Pathol 1997;107:56–63.

14. Ng T, Shima D, Squire A, Bastiaens PI, Gschmeissner S, HumphriesMJ, et al. PKC alpha regulates beta1 integrin-dependent cell motilitythrough association and control of integrin traffic. EMBO J 1999;18:3909–23.

15. ParsonsM, KepplerMD, Kline A,Messent A, HumphriesMJ, Gilchrist R,et al. Site-directed perturbation of protein kinase C- integrin interactionblocks carcinoma cell chemotaxis. Mol Cell Biol 2002;22:5897–911.

16. Kolanus W, Seed B. Integrins and inside-out signal transduction:Converging signals from PKC and PIP3. Curr Opin Cell Biol 1997;9:725–31.

17. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. Highdensity lipoprotein as a protective factor against coronary heart dis-ease. The Framingham Study. Am J Med 1977;62:707–14.

18. Wilson PW, Garrison RJ, Castelli WP, Feinleib M, McNamara PM,Kannel WB. Prevalence of coronary heart disease in the FraminghamOffspring Study: role of lipoprotein cholesterols. Am J Cardiol1980;46:649–54.

19. Gordon DJ, Probstfield JL, Garrison RJ, Neaton JD, Castelli WP, KnokeJD, et al. High-density lipoprotein cholesterol and cardiovascular dis-ease. Four prospective American studies. Circulation 1989;79:8–15.

20. Barter P, Gotto AM, LaRosa JC, Maroni J, Szarek M, Grundy SM, et al.HDLcholesterol, very low levels of LDLcholesterol, and cardiovascularevents. N Engl J Med 2007;357:1301–10.

21. Jafri H, Alsheikh-Ali AA, Karas RH. Baseline and on-treatment high-density lipoprotein cholesterol and the risk of cancer in randomizedcontrolled trials of lipid-altering therapy. J Am Coll Cardiol 2010;55:2846–54.

22. Moorman PG, Hulka BS, Hiatt RA, Krieger N, Newman B, VogelmanJH, et al. Association between high-density lipoprotein cholesterol andbreast cancer varies by menopausal status. Cancer Epidemiol Bio-markers Prev 1998;7:483–8.

23. Kucharska-NewtonAM,RosamondWD,MinkPJ, AlbergAJ, Shahar E,Folsom AR. HDL-cholesterol and incidence of breast cancer in theARIC cohort study. Ann Epidemiol 2008;18:671–7.

24. Furberg AS, Veierød MB, Wilsgaard T, Bernstein L, Thune I. Serumhigh-density lipoprotein cholesterol, metabolic profile, and breastcancer risk. J Natl Cancer Inst 2004;96:1152–60.

25. Kontush A, Chapman MJ. Why is HDL functionally deficient in type 2diabetes? Curr Diab Rep 2008;8:51–9.

26. Zheng L,NukunaB,BrennanML, SunM,GoormasticM, SettleM, et al.Apolipoprotein A-I is a selective target for myeloperoxidase-catalyzedoxidation and functional impairment in subjects with cardiovasculardisease. J Clin Invest 2004;114:529–41.

27. Feng H, Li XA. Dysfunctional high-density lipoprotein. Curr OpinEndocrinol Diabetes Obes 2009;16:156–62.

28. Lapolla A, Brioschi M, Banfi C, Tremoli E, Bonfante L, Cristoni S, et al.On the search for glycated lipoprotein ApoA-I in the plasma of diabeticand nephropathic patients. J Mass Spectrom 2008;43:74–81.

29. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of humanendothelial cells derived from umbilical veins. Identification by mor-phologic and immunologic criteria. J Clin Invest 1973;52:2745–56.

30. Chung BH, Wilkinson T, Geer JC, Segrest JP. Preparative and quan-titative isolation of plasma lipoproteins: rapid, single discontinuousdensity gradient ultracentrifugation in a vertical rotor. J Lipid Res1980;21:284–91.

31. Hedrick CC, Thorpe SR, Fu MX, Harper CM, Yoo J, Kim SM, et al.Glycation impairs high-density lipoprotein function. Diabetologia2000;43:312–20.

32. Valiyaveettil M, Kar N, Ashraf MZ, Byzova TV, Febbraio M, Podrez EA.Oxidized high-density lipoprotein inhibits platelet activation andaggregation via scavenger receptor BI. Blood 2008;111:1962–71.

33. Gamble JR, Vadas MA. A new assay for the measurement of theattachment of neutrophils and other cell types to endothelial cells.J Immunol Methods 1988;109:175–84.

34. Ratheesh A, Ingle A, Gude RP. Pentoxifylline modulates cell surfaceintegrin expression and integrin mediated adhesion of B16F10 cells toextracellular matrix components. Cancer Biol Ther 2007;6:1743–52.

35. Kimura T, Tomura H, Mogi C, Kuwabara A, Ishiwara M, Shibasawa K,et al. Sphingosine 1-phosphate receptors mediate stimulatory andinhibitory signalings for expression of adhesion molecules in endo-thelial cells. Cell Signal 2006;18:841–50.

36. Pan B, Ren H, Ma Y, Liu D, Yu B, Ji L, et al. HDL of patients with type 2diabetes mellitus elevates the capability of promoting migration andinvasion of breast cancer cells. Int JCancer 2011 Jul 29 [Epub aheadofprint].

37. Rubinek IWT Diabetes mellitus and breast cancer. In. Masur K, Th�eve-nod F, Z€anker KS (eds): Diabetes and cancer. Epidemiological evi-dence and molecular links. Front diabetes. Basel, Karger: PlenumPress; 2008;19:97–113.

38. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860–7.

39. Pikarsky E, Porat RM, Stein I, Abramovitch R, Amit S, Kasem S, et al.NF-kappaB functions as a tumour promoter in inflammation-associ-ated cancer. Nature 2004;431:461–6.

40. HoffmanRM.Orthotopicmetastaticmousemodels for anticancer drugdiscovery and evaluation: a bridge to the clinic. Invest New Drugs1999;17:343–59.

41. Hoffman RM. The multiple uses of fluorescent proteins to visualizecancer in vivo. Nat Rev Cancer 2005;5:796–806.

42. Hoffman RM, Yang M. Color-coded fluorescence imaging of tumor-host interactions. Nat Protoc 2006;1:928–35.

43. Rotheneder M, Kostner GM. Effects of low- and high-density lipopro-teins on the proliferation of human breast cancer cells in vitro: differ-ences between hormone-dependent and hormone-independent celllines. Int J Cancer 1989;43:875–9.

44. Martinez LO, Jacquet S, Terc�e F, Collet X, Perret B, Barbaras R. Newinsight on the molecular mechanisms of high-density lipoproteincellular interactions. Cell Mol Life Sci 2004;61:2343–60.

45. Kimura T, Tomura H, Mogi C, Kuwabara A, Damirin A, Ishizuka T, et al.Role of scavenger receptor class B type I and sphingosine 1-phos-phate receptors in high density lipoprotein-induced inhibition of adhe-sion molecule expression in endothelial cells. J Biol Chem 2006;281:37457–67.

46. Sekine Y, Demosky SJ, Stonik JA, Furuya Y, Koike H, Suzuki K, et al.High-density lipoprotein induces proliferation and migration of humanprostate androgen-independent cancer cells by anABCA1-dependentmechanism. Mol Cancer Res 2010;8:1284–94.

47. MatsunagaT,NakajimaT,Miyazaki T, Koyama I, Hokari S, Inoue I, et al.Glycated high-density lipoprotein regulates reactive oxygen speciesand reactive nitrogen species in endothelial cells. Metabolism2003;52:42–9.

48. Norata GD, Pirillo A, Catapano AL. Modified HDL: biological andphysiopathological consequences. Nutr Metab Cardiovasc Dis 2006;16:371–86.

49. Furberg AS, Espetvedt S, Emaus A, KhanN, Thune I. Low high-densitylipoprotein cholesterol may signal breast cancer risk: recent findingsand new hypotheses. Biomark Med 2007;1:121–31.

50. Kumar K, SachdanandamP, ArivazhaganR. Studies on the changes inplasma lipids and lipoproteins in patients with benign and malignantbreast cancer. Biochem Int 1991;23:581–9.

Pan et al.





2012;18:1246-1256. Published OnlineFirst January 18, 2012.Clin Cancer Res Bing Pan, Hui Ren, Yubin He, et al. Capability of Promoting Breast Cancer MetastasisHDL of Patients with Type 2 Diabetes Mellitus Elevates the

Updated version

10.1158/1078-0432.CCR-11-0817doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2012/01/18/1078-0432.CCR-11-0817.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/18/5/1246.full#ref-list-1

This article cites 48 articles, 8 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/18/5/1246.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/18/5/1246To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-11-0817

http://clincancerres.aacrjournals.org/content/suppl/2012/01/18/1078-0432.CCR-11-0817.DC1

http://clincancerres.aacrjournals.org/content/18/5/1246.full#ref-list-1

http://clincancerres.aacrjournals.org/content/18/5/1246.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/18/5/1246


hdl of patients with type 2 diabetes mellitus elevates the ... · human cancer biology hdl of...

Documents