research article carbohydrate microarrays identify blood...

Research ArticleCarbohydrate Microarrays Identify Blood GroupPrecursor Cryptic Epitopes as Potential ImmunologicalTargets of Breast Cancer

Denong Wang1 Jin Tang1 Shaoyi Liu2 and Jiaoti Huang3

1Tumor Glycomics Laboratory SRI International Biosciences Division 333 Ravenswood Avenue Menlo Park CA 94025 USA2Department of Pharmacology Weill Medical College of Cornell University 1300 York Avenue New York NY 10065 USA3Department of Pathology David Geffen School of Medicine University of California in Los Angeles 570 Westwood PlazaLos Angeles CA 90095 USA

Correspondence should be addressed to Denong Wang denongwangsricom

Received 25 April 2015 Revised 1 August 2015 Accepted 6 August 2015

Academic Editor Paola Nistico

Copyright copy 2015 Denong Wang et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Using carbohydratemicroarrays we explored potential natural ligands of antitumormonoclonal antibodyHAE3This antibodywasraised against a murinemammary tumor antigen but was found to cross-react with a number of human epithelial tumors in tissuesOur carbohydrate microarray analysis reveals that HAE3 is specific for an O-glycan cryptic epitope that is normally hidden in thecores of blood group substances Using HAE3 to screen tumor cell surface markers by flow cytometry we found that the HAE3glycoepitope gpHAE3 was highly expressed by a number of human breast cancer cell lines including some triple-negative cancersthat lack the estrogen progesterone and Her2neu receptors Taken together we demonstrate that HAE3 recognizes a conservedcryptic glycoepitope of blood group precursors which is nevertheless selectively expressed and surface-exposed in certain breasttumor cellsThepotential of this class ofO-glycan cryptic antigens in breast cancer subtyping and targeted immunotherapywarrantsfurther investigation

1 Introduction

Recognition of abnormal glycosylation in almost any cancertype has raised great interest in the exploration of the tumorglycome for biomarker discovery [1ndash3] In this study weexplored potential glycan markers that are overexpressed onthe surfaces of breast cancers A key immunological probeof this investigation is an antitumor monoclonal antibody(mAb) HAE3 This mAb was raised against epiglycanin themajor sialomucin glycoprotein (sim500 kDa) of murine mam-mary adenocarcinoma TA3 cells [4] It was initially calledAE3 but was later renamed HAE3 [5] to avoid confusionwith a commonly used anti-cytokeratin antibody in cancerresearch [6 7] Interestingly HAE3 was found to stronglycross-react with a number of human epithelial tumors intissues including lung prostate bladder esophagus andovarian cancers [5 8ndash10]

This cross-species tumor binding profile suggests thepossibility that HAE3 may recognize a conserved tumorglycan marker that is coexpressed by both mouse- andhuman-derived epithelial cancers HAE3 was initially sug-gested to resemble lectin peanut agglutinin (PNA) whichrecognizes the T disaccharide (Gal12057313GalNAc1205721-) linked toSerThr [11] However the antibody differed from PNA inthat the concentration of the blood group T disacchariderequired for inhibition of binding to epiglycanin was 104times greater than for PNA Moreover a T-specific mAbHH8 was found to be negative with epiglycanin in ELISAmicrotiter plates [12] HH8 is specific for the T-terminaldisaccharide moiety expressed by asialoglycophorin A [13]Consistent with these observations Palma et al reportedrecently that the neoglycolipid conjugates that display a singleT disaccharide epitope were negative with HAE3 [14] Ofnote mAb HAE3 was cited as AE3 in the report

Hindawi Publishing CorporationJournal of Immunology ResearchVolume 2015 Article ID 510810 9 pageshttpdxdoiorg1011552015510810

2 Journal of Immunology Research

Given that HAE3 has been analyzed using a large col-lection of synthetic neoglycoconjugates (119899 = 492) [14]this study focused on identification of the potential naturalligands of HAE3 In essence we produced a comprehensivecarbohydrate microarray using a large collection of purifiednatural carbohydrate antigens for screening These includeA B O Lewisab I and i and the blood group precursorsof various biological origins As summarized below we haverevealed that HAE3 is specific for a blood group precursorcryptic epitope that is normally hidden in the cores or internalchains of blood group substances but becomes differentiallyexpressed in human breast cancer cells

2 Material and Methods

21 Antigens and Antibodies A preparation of humancarcinoma-associated antigen (HCA) (1) was kindly pro-vided by Dr Zeqi Zhou of Egenix (Millbrook NY) Themurine hybridoma IgM antibody HAE3 was produced bymouse immunization (C57BLJ) with asialoepiglycanin (85)[4 5] A preparation of purifiedHAE3 protein was purchasedfrom RA Biosources Inc (Belmont CA) Carbohydrateantigens applied in this study are listed in Supplemen-tary Table 1 in Supplementary Material available online athttpdxdoiorg1011552015510810

22 Cell Lines Cancer cell lines used include breast-derived(T-47D SK-BR-3 MCF-7 BT-549 MD-AMB-231 and MD-AMB-468) lung-derived (A549) or prostate-derived (PC3)epithelial tumor cell lines and a skin-derived melanomaSKMEL-28 All tumor cell lines were acquired from ATCC

23 Carbohydrate Microarrays Microarray assays were per-formed as described [15] In brief a microarray robot(PIXSYS 5500C Cartesian Technologies Irvine CA) wasused to spot antigen preparations onto glass slides precoatedwith nitrocellulose polymer (FAST Slides Schleicher ampSchuell KeeneNH)Theprintedmicroarrayswere incubatedat room temperature with HAE3 (IgM) antibody at 5 120583gmLin 1 (wtvol) BSA in PBS containing 005 (wtvol) NaN

3

and 005 (volvol) Tween-20 An R-phycoerythrin- (R-PE-) conjugated affinity-purified F(ab1015840) fragment of goatanti-mouse IgM secondary antibody preparation (RocklandImmunochemicals Inc PA) was applied at 20 120583gmL toreveal HAE3-specific staining signal Fluorescence intensityvalues for each array spot and its background were calcu-lated using ScanArray Express software SAS Institutersquos JMP-Genomics software package (httpwwwjmpcom CaryNC) was used for microarray data standardization andstatistical analysis Results of the microarray assay are shownas the means of fluorescent intensities (MFIs) of triplicatedetections of given antigen preparations (Figure 1(a) andTable 1)

24 Fluorescence-Activated Cell Sorting (FACS) AnalysisHAE3 (IgM 50 120583gmL) and an isotype control mAb 9147(IgM anti-120572(16)dextran 50 ugmL) [16] were applied tostain tumor cell lines The R-PE-conjugated goat anti-mouse

IgM antibody preparation described above was applied inthe second staining step to reveal the antigen-captured IgMFACS data were collected with LSR-II (BD Bioscience SanJose CA) and analyzed with FlowJo (TreeStar Inc AshlandOR)

25 Enzyme-Linked Immunosorbent Assay (ELISA) andELISA Inhibition Assays Carbohydrate-specific ELISA andELISA inhibition assays were performed as described [1517] In brief glycoprotein antigen preparations were dilutedin 01M sodium bicarbonate buffer solution pH 96 forcoating on ELISA microplates (NUNC MaxiSorp ThermoScientific Rochester NY) followed by blocking using 1 BSAand 1X Phosphate Buffered Saline Tween-20 (PBST) MAbHAE3 (IgM) (25120583gmL) and biotinylated PNA (20120583gmL)were diluted in 1 BSA PBST for the ELISA binding assayThe bound HAE3 and PNA were revealed by an alkalinephosphatase- (AP-) conjugated goat anti-mouse IgM and anAP-streptavidin conjugate (Sigma Chemical Co St LouisMO) respectively ELISA inhibition assays were performedwith EPGN (85) (1 120583UmL) coated to display the nativegpHAE3 epitope and a series of blood group reference antigensas potential inhibitors (250120583gmLeach) to compete with theHAE3 (10 120583gmL) binding of EPGN Percent inhibition wascalculated as follows inhibition = ((standard A minus blank A)minus (A with inhibitor minus A))(standard A minus blank A)

26 Bio-Gel Chromatography Bio-Gel P-10 filtration wasperformed following the manufacturersquos manual (Bio-RadLaboratories Hercules CA) with minor modifications Inbrief Tij II (20 from 2nd 10) substance was fractioned ina precalibrated Bio-Gel P-10 column The sizes of the Tij IIsubstance were measured based on the neutral sugar elutionprofile with reference to the calibrated saccharide molecularweight standards as shown in Figure 4

3 Results

31 Carbohydrate Microarray Identifies Blood Group Precur-sors as the Natural Ligands of HAE3 As shown in Figure 1and Table 1 an HCA preparation (ID 1 and 2) was spotted asa positive control forHAE3 activityThisHAE3+ glycoproteinpreparation was affinity-purified from cultural supernatantof the lung cancer cell line A549 using an HAE3-agarosecolumn (Egenix Millbrook NY) Given that the simple O-glycan core T disaccharide (Gal12057313GalNAc-) and its peptideconjugates were found to weakly but significantly inhibitHAE3 binding to epiglycanin we characterized a panel ofblood group substances that carry O-glycan cores in thismicroarray screening A number of blood group precursors(29ndash32) were plotted from 3 to 32 which was followedby other antigens from 33 to 78 and microarray printingand scanning calibration controls in 79 and 80 Blood groupsubstance reference reagents [18] used include the followingCyst9 and Cyst14 A active Beach phenol insoluble B activeHog H active JS phenol insoluble H and Leb active andN-1 20 from the second 10 Lea active Key blood groupprecursor references includeOGTij II BeachP1 andMcDon

Journal of Immunology Research 3

1 HCA 01

2 HCA 05

29 Beach P1 (Ii)

30 McDon P131 Tij II (Ii)

32 OG (Ii) 80 Dex-FITCmarker

Blood groupprecursors (29ndash32)

3ndash32 blood group antigens 33ndash76 other autoantigens and microbial antigens

2 3 4 5 6 7 8 91 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 8010 60

58

61

64

36373839404142434445464748495051525354555657

34

59

3332

6263

35

656667686970717273747576777879

3031

Antigen ID numberAntigenBackground

0

2000

4000

6000

8000

10000

12000

14000

16000

MFI

(a)

31 Tij II29 Beach P1

30 McDon P1

32 OG

2 HCA 05

1 HCA 01

80 Dex-FITC

(b)

Schematic of an O-glycan cryptic antigen the internal carbohydrate moiety of blood group substances

Branch 1

Branch 2 Branch 3

Branch 4

A

B

Fuc

H

I (II)

darr 1205721 2

Galdarr 1205731 4

GlcNAcdarr 1205731 6

Gal


Galuarr 1205731 4

GlcNAcuarr 1205731 6

darr 1205731 43

II120573

II120573 II120573I120573

GalNAc1205721 rarr 3

Gal1205721 rarr 3

Gal1205731 rarr 3(4)GlcNAc1205731 rarr 3Gal1205731 rarr 3GlcNAc1205731 rarr 3Gal1205731 rarr 3GalNAc1205721 rarr SerThr

(c)

Figure 1 Carbohydrate microarray analysis of antiepiglycanin mAb HAE3 Seventy-six glycoproteins glycoconjugates and polysaccharideswere spotted in triplicates in 1 to 2 dilutions to yield the customizedmicroarrays for antibody screening (a)Microarray detections were shownas the mean fluorescent intensities (MFIs) of each microspot with antigen-binding signal in red and background reading in blue Each errorbar is constructed using one standard deviation from the mean of triplicate detections The labeled antigens include HCA (ID 1 and ID2) a number of blood group precursors (29ndash32) and a microarray spotting marker (80) (b) Images of a microarray stained with HAE3(5120583gmL) (c) Schematic of a blood group substance structure with the conserved O-glycan core highlighted

P1These precursor substances were prepared to removemostof the 120572-L-fucosyl end groups that are essential for bloodgroups A B and H or Lewis active side chains but possessthe internal domains or core structures of blood groupsubstances A large panel of other autoantigens andmicrobialpolysaccharides were spotted in the same microarrays toexamine potential polyreactivity of this IgM antibody

Figures 1(a) and 1(b) illustrate a representative resultof multiple microarray screening assays In Figure 1(a) theMFIs of carbohydratemicroarray detections ofHAE3bindingsignal (red column) are plotted with corresponding localbackground reading (blue column) as an overlay plot Eachdata point represents the mean of triplicate detectionsthese are shown in the Figure 1(b) microarray image withthe number of positive antigens labeled Each error bar is

constructed using one standard deviation from the mean Asillustrated HAE3 is strongly positive with HCA (1 and 2) asexpected Importantly this antibody selectively binds to fourblood group precursor antigens Beach P1 (29) McDon P1(30) Tij II (31) and OG (32) By contrast HAE3 has nodetectable cross-reactivity with blood group substances A BO or Lewis antigens or the large panel of other carbohydrateantigens spotted in the same array

Figure 1(c) is a schematic of blood group substancestructure with the common blood group precursor corestructure highlighted The four-branched structure in thecircle represents the internal portion of the carbohydratemoiety of blood group substances which was proposed basedon extensive immunochemical characterization of precursorOG and other P1 fractions of blood group precursors that


Table 1 Dataset from a carbohydrate microarray analysis of mAb HAE3 (Figure 1)lowastAntigen spotted Fluorescent intensities (INT) Microarray scores (log 2-INT)

(ID) antigen name Replicates Means StDev AgBg Means StDev lowastlowast119905-test (119901 value)

(1)HCA(A549) 1 5 3 2306 214 209 11167304 0132563 666755E minus 05(2) HCA (A549) 3 8537 265 773 13058996 0044597 10582E minus 07(3) Cyst9 (A) 3 1065 5 096 1005708 00062627 0695270476(4) JS (A) 3 1059 2 096 10048485 00027246 0630836897(5) HGM-BGS (A + H) 3 1060 12 096 10049785 00166116 0641720796(6) Hog76 (A) 3 1095 6 099 10096264 00072603 0982754025(7) Hog (A) 3 996 7 090 99604633 00096233 0195275968(8) MSS (A) 3 1139 134 103 10147131 01651035 0701231922(9) Hog39 B2 (A) 3 1046 10 095 10031084 00136387 0512794031(10) Cyst14 (A2) 3 1144 116 104 10155053 01435773 06354829(11) WG (A2) 3 1021 13 092 99952239 00176786 0320320938(12) Cyst11 3 1011 2 091 99810904 0002181 0261637986(13) Cow21 3 1083 9 098 10080783 00122256 0886644093(14) Beach (B) 3 1152 83 104 10167059 01023248 051871479(15) Cow28 (B) 3 1065 27 096 10056322 00369858 0695618692(16) Cow43 3 1012 7 092 99824975 00094843 0267256771(17) Cow26 3 1093 56 099 10092395 00732857 0985749244(18) Hog5 3 1057 4 096 10046209 00047907 0614466244(19) Hog 10 3 1076 8 097 10071437 00104506 0809324608(20) Hog6 (H) 3 1462 595 132 10441803 05435961 0384564394(21) Hog67 3 1067 13 097 10058822 00176694 0709880521(22) Hog (H) 3 1132 10 102 10144623 00123178 0598292687(23) Hog30 3 1071 40 097 10064074 00539668 0760857847(24) Cow21 (I-Ma) 3 1031 7 093 10009339 00099403 0386688505(25) Cow25 3 1051 24 095 10036836 0033184 0555735986(26) Cow26 (I) 3 1008 13 091 9977683 0017886 0250673214(27) N-1 10 2X (Lea) 3 1139 10 103 10153512 00131909 0537783247(28) N-1 IO4- (Lea) 3 1044 18 094 100273 00250417 0491711028(29) Beach P1 (Ii) 3 2119 22 192 11049118 001476 0000114629(30) McDon P1 3 7675 604 695 12903032 0113579 491829E minus 08(31) TijII (Ii) 3 7050 220 638 12782943 0044774 183548E minus 07(32) OG (Ii) 3 11166 433 1010 13446047 0056595 250433E minus 08(33) LNT-BSA (Type I) 3 1092 12 099 10092703 00152473 09870917(34) Pn XIV (Type II) 3 1108 17 100 10113624 002265 0838256847(35) ASOR (Trim-II) 3 1128 26 102 10138873 00330974 0644502626(36) AGOR (Trim-Gn) 3 1042 5 094 1002559 00068188 0477740236(37) iAFGP 3 1024 5 093 10000458 00070924 0342706487(38) Chondroitin-SO4-A 3 1038 13 094 10019053 00179351 0440853108(39) Chondroitin-SO4-B 3 1128 10 102 10139939 00128602 0631919148(40) Chondroitin-SO4-C 3 1047 14 095 10031963 00189058 0519441046(41) Hyaluronic Acid 3 1096 9 099 10098441 00113582 0964347497(79) Bg 6 1105 188 100 1009423 0219076(80) Dex70K-FITC 3 63225 1067 5722 15948071 00243011 823802119864 minus 09

lowastAntigenrsquos initial spotting concentration was 05 120583g120583L The positive results are emphasized with bold Microbial antigens tested (42ndash78) were negative withHAE3 (AgBg lt 120 119901 gt 020 data not shown)lowastlowast119905-test microarray scores that is the log 2-transformed microarray values from triplicate spots for each antigen were applied in a 119905-test to examine the

differences of significance between each probe and Bg (microarray reading background) which is the mean fluorescent intensity of six spots of Av-Cy3Cy5(ID 79) in the FITC channel


HAE3 PNA

TTij II TnBgO-core natural antigens

0

05

10

15

20

25

30

OD405

nm


0

10

20

30

40

OD405

nm

(a) ELISA binding assays

Tij II

0

20

40

60

80

100

Inhi

bitio

n (

)

B-P1

(29)

Tij I

I (31

)

H (2

2)

OG

(32)

EPG

N (8

5)

B (1

4)

Tn (8

3)

H (4

)

HCA

(1)

A (3

)

T (8

4)

Inhibitors(b) ELISA inhibition assays

Figure 2 Carbohydrate-specific ELISA and ELISA inhibition assays validate binding specificities of HAE3 (a) An antigen-specific ELISAdistinguished HAE3 binding specificity from the T-antigen-specific PNA ELISA plates were coated with O-core antigen Tij II (31) T (81)and Tn (82) at 10120583gmL to react with either HAE3 (50 120583gmL) or PNA (10 120583gmL) (b) ELISA inhibition assays with a series of carbohydrateantigens as competitors (250 120583gmL) to inhibit interaction between EPGN (10120583UmL) and anti-HCA (HAE3 10120583gmL) These antigensinclude HCA (1) EPGN (85) which is the immunogen of HAE3 and blood group substances Tn (83) T (84) H (4 or 22) A (3) B(14) Beach P1 (29) Tij II (31) and OG (32) Results are shown as percent inhibition in the presence of an inhibitor

were isolated from ovarian cancer cyst fluids [19ndash22] Selec-tive detection of these blood group precursors from a largepanel of blood group substances by HAE3 illustrated that thisantibody is specific for a shared cryptic glycoepitope of theseprecursor substances

32 Carbohydrate-Specific ELISA and ELISA Inhibition AssaysValidate Binding Specificity of HAE3 We further exam-ined whether the observed HAE3 binding reactivity can beattributed to cross-reactivity with TTn glycoepitopes thatare often expressed as components of blood group precursorsubstances For this purpose we tested ELISA binding ofHAE3 or T-specific lectin PNA with blood group precursorTij II (31) and two TTn-positive glycoconjugates asialo-PSM (T) (81) and asialo-OSM(Tn) (82) Unlike PNAwhichbinds to both Tij II and asialo-PSM (T) HAE3 specificallybinds to Tij II without cross-reacting with asialo-PSM (T)or asialo-OSM (Tn) (Figure 2(a)) Thus HAE3 binding of TijII is irrelevant to the native TTn-glyco-epitopes expressedby these glycoconjugates ELISA inhibition assays furtherdemonstrated that blood group precursors Tij II (31) andOG (32) but not other blood group antigens includingA (3) B (14) H (22) T (84) or Tn (83) significantlyinhibitedHAE3 binding to the immunogen asialoepiglycanin(85) (Figure 2(b))

33 FACS Analysis Detects Tumor Cell Surface Expressionof gp1198671198601198643 We examined whether the HAE3+ glycoepitopeswere expressed as cell surface tumor markers In the firstset of experiments we screened a panel of four tumor celllines by cell surface staining in flow cytometryThese include(a) a breast cancer line T-47D which was selected owing tothe fact that breast cancer patients were found to producesubstances in circulation that are highly effective in inhibiting

HAE3 binding of epiglycanin [12 23] (b) a lung cancerline A549 which is known to produce an HAE3-positivesubstance in cell culture (c) a prostate cancer line PC3whichis found to express a blood group B-related F77 glycoepitope[24 25] and (d) a melanoma cell line SKMEL-28 whichis derived from skin but not epithelial tissue As shown inFigure 3(a) melanoma SKMEL-28 and prostate cancer PC3were negative for HAE3 The A549 lung cancer cell line wasweakly positive By contrast the breast cancer cell line T-47Dwas strongly positive in HAE3-cell surface staining

Given these results we extended the FACS analysis to apanel of seven human breast cancer cell lines including twoER+PR+ lines (T-47D and MCF-7) one ER+ (SK-BR-3) andfour triple-negative cancers (BT-549 Hs 578T MDA-MB-231 andMDA-MB-468) Figure 3(b) shows that two ER+PR+lines T-47D and MCF-7 and two triple-negative lines BT-549 and MDA-MB-468 are gpHAE3 strongly positive SK-BR-3 is gpHAE3 intermediately positive By contrast the tworemaining triple-negative cell lines Hs578T and MDA-MB-231 were HAE3 negative

4 Discussion

Expression of gpHAE3 by human breast cancer has been exam-ined in this study FACS analyses (Figure 3) revealed that fiveof the seven breast cancer cell lines are HAE3 positive Theseare T-47DMCF-7 SK-BR-3 BT-549 andMDA-MB-468 BT-549 cells are triple-negativebasal-B mammary carcinomaMDA-MB-468 cells are known as triple-negativebasal-Amammary carcinoma The two remaining gpHAE3minus triple-negative cell lines Hs578T and MDA-MB-231 were basal-Bmammary carcinoma It is important to extend this investiga-tion to a cohort of breast cancer patients to examine whetherthis marker was significantly associated with metastatic


HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

(PCa)(BCa)T47D

(melanoma)(LCa)A549 PC3

Max

()

HAE3R-PEHAE3R-PE101 102 103 104100 101 102 103 104100

HAE3R-PE101 102 103 104100

HAE3R-PE101 102 103 104100

0

20

40

60

80

100SKMEL-28

(a)

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

(TN) (TN) (TN) (TN)Hs 578T MDA-MB-468MDA-MB-231BT-549

PR+ PR+ER+ ER+

MCF7T-47DHER2+

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100SK-BR-3

(b)

Figure 3 HAE3 cell surface staining detected selective expression of the HAE3-cryptic glycan markers in human cancer cell lines (a) Fourtumor cell lines T-47D A549 PC3 and SKMEL-28 were stained with the C1 preparation of HAE3 (IgM) at 1 6 dilution or with an isotypecontrol IgM 9147 (50 120583gmL) (b) Seven breast cancer cell lines were stained with purified mAb HAE3 (50120583gmL) or 9147 (50 120583gmL)These cell lines are T-47D MCF-7 SK-BR-3 BT-549 Hs578T MDA-MB-231 and MDA-MB-468 An R-PE-conjugated goat anti-mouse IgMantibody was applied to quantify the cell surface-captured IgM antibodies Blue line HAE3 stain Red line 9147 IgM isotype control

breast cancer and breast circulating tumor cells especiallyin patients with triple-negative cancer cells that lack specificsurface biomarkers

Our carbohydrate microarray analysis has identifiedblood group precursor substances as the natural ligandsof antibody HAE3 As shown in Figure 1 four well-characterized blood group precursor reference antigensBeach P1 (29) McDon P1 (30) Tij II (31) and OG (32)were HAE3 positive OG [19] and Tij II [26] antigens were

prepared by pepsin digestion ethanol precipitation andsolubilization in 90 phenol followed by fractional ethanolprecipitation from phenol Beach P1 [27] and McDon P1 [28]were obtained as the nondialyzable O-cores from partiallyhydrolyzed blood group antigen Beach B and McDon A1respectively Thus these blood group precursor substanceswere prepared to eliminate peripheral glycoepitopes suchas A B H or Lewis antigen-specific epitopes but preservetheir O-glycan core structures leaving a number of cryptic


chromatographyBio-Gel P-10Tij II antigen

20 10 5 IM5MW (kDa)

40 6020 800

Fraction number (05mL)

0

1

2

3

4

5

6

7

8

A490

Figure 4 A Bio-Gel P-10 plot of Tij II antigen Tij II (20 from2nd 10) substance was fractioned in a precalibrated Bio-Gel P-10 column at 05mL per fraction The neutral sugar content ineach fraction was determined by phenol-sulfuric acid color reactionand quantitatively measured at OD490 nm The sizes of the TijII substance were measured based on the neutral sugar elutionprofile with reference to the calibrated saccharide molecular weightstandards as indicated in the plot IM

5stands for isomaltopentaose

O-core epitopes exposed for antibody recognition Selectivedetection of these blood group precursors from a large panelof blood group substances by HAE3 demonstrated that thisantibody is specific for a shared cryptic glycoepitope of theseprecursor substances

The native blood group precursor substances are oftenmore complex in carbohydrate structures than the mostknown model O-cores [29] As determined by Bio-Gel P-10filtration the sizes of the Tij II blood group precursor sub-stance were distributed in a range of molecular weights from7KDa to 15 KDa approximately (Figure 4) Since this antigencontains only approximately 2 peptide sequences its massis apparently made up predominately of carbohydrates [2630 31] Figure 1(c) is a postulated blood group precursor corestructure which was proposed based on extensive glycanstructural analyses and immunochemical studies of bloodgroup substances [19ndash22] The four-branched structure inthe circle represents the internal portion of the carbohydratemoiety of blood group substances

Chemical synthesis of such complex blood group pre-cursors is technically challenging However it is not impos-sible to rationally design and produce HAE3-positive com-pounds by stabilizing relatively simple O-cores via specificstructural modifications For example a recent microarrayscreening revealed an unpredicted binding of this anti-body to a sulfated glycolipid SM1a Gal1205731-3GalNAc1205731-4(3-O-sulfate)Gal1205731-4GlcCer [14] which can be viewed as anO-corederivative SM1a naturally occurs in small amounts in normalkidney [32] but such a carbohydrate sequence has not beendescribed in tumor glycome

Tumor-associated overexpression of blood group-relatedautoantigens is not limited to breast cancer Gao et al recentlyreported that the natural ligand of a prostate cancer-specificmAb F77 is in fact blood group H which is built on a6-linked branch of a poly-N-acetyllactosamine backbone

[24 25] Overexpression of gpF77 in prostate cancers mayreflect increased blood group H expression together withupregulated expression of branching enzymes As illustratedin Figures 1ndash3 HAE3 differs from F77 in glycan bindingspecificities and tumor binding profiles Unlike F77 whichis blood group H-specific and stains prostate cancer cell linePC3 HAE3 has neither reactivity with blood groupH nor thecell surface targets of PC3

BothHAE3 and F77 studies call our attention to epithelialtumor expression of blood group substance-related autoanti-gens It is noteworthy that blood group substance antigensmay also serve as the natural ligands of C-type lectinDC-SIGN one of the key glycan-binding receptors of theconserved innate immune system [33ndash35] Our preliminarydata indicates that the HAE3-positive TijII antigen is likelya DC-SIGN ligand (data not shown) Potential of this classof tumor glycoantigens as costimulators of the immune cellsin both innate and acquired immune systems for tumorvaccine development and targeted immunotherapy is yet tobe explored

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Disclaimer

Thecontent is solely the responsibility of the authors and doesnot necessarily represent the official views of the NationalInstitutes of Health

Acknowledgments

The authors acknowledge John L Daiss and Zeqi (Joe)Zhou for valuable discussions Narayanan Parthasarathy andXiaohe Liu for technical assistance and the Kabat Collectionof Carbohydrate Antigens at SRI International for a panelof carbohydrate antigens that were applied in this studyThis work is supported in part by NIH Grants U01CA128416(Denong Wang) and R56AI108388 (Denong Wang) and SRIInternational IRampD funds (Denong Wang)

References

[1] S Hakomori ldquoAberrant glycosylation in tumors and tumor-associated carbohydrate antigensrdquoAdvances inCancer Researchvol 52 pp 257ndash331 1989

[2] M Fukuda ldquoPossible roles of tumor-associated carbohydrateantigensrdquo Cancer Research vol 56 no 10 pp 2237ndash2244 1996

[3] D H Dube and C R Bertozzi ldquoGlycans in cancer andinflammationmdashpotential for therapeutics and diagnosticsrdquoNature Reviews Drug Discovery vol 4 no 6 pp 477ndash488 2005

[4] J F Codington B H Sanford and RW Jeanloz ldquoGlycoproteincoat of the TA3 cell Isolation and partial characterization of asialic acid containing glycoprotein fractionrdquo Biochemistry vol11 no 14 pp 2559ndash2564 1972

[5] R Li J L Yao P A Bourne P A Di SantrsquoAgnese and J HuangldquoFrequent expression of human carcinoma-associated antigen


a mucin-type glycoprotein in cells of prostatic carcinomardquoArchives of Pathology amp Laboratory Medicine vol 128 no 12pp 1412ndash1417 2004

[6] K I Al-Shibli H A Mohammed and K S Mikalsen ldquoSentinellymph nodes and breast carcinoma analysis of 70 cases byfrozen sectionrdquo Annals of Saudi Medicine vol 25 no 2 pp 111ndash114 2005

[7] J Akimoto H Namatame J Haraoka and M Kudo ldquoEpithe-lioid glioblastoma a case reportrdquo Brain Tumor Pathology vol22 no 1 pp 21ndash27 2005

[8] S Liang J Yao P A Bourne P A DiSantrsquoAgnese J Huang andJ-Y Lei ldquoOverexpression of human carcinoma-associated anti-gen in esophageal adenocarcinoma and its precursor lesionsrdquoAmerican Journal of Clinical Pathology vol 122 no 5 pp 747ndash751 2004

[9] T Thingstad S Haavik K Hansen K Sletten J F Codingtonand H Barsett ldquoHuman carcinoma-associated antigen (HCA)isolated from the endometrial carcinoma cell line KLE-1 andascitic fluid of a patient with ovarian carcinoma comparisonwith epiglycaninrdquo European Journal of Pharmaceutical Sciencesvol 6 no 2 pp 121ndash129 1998

[10] J L Yao P A Bourne Q Yang J Lei P A Di SantrsquoAgneseand J Huang ldquoOverexpression of human carcinoma-associatedantigen in urothelial carcinoma of the bladderrdquo Archives ofPathology and Laboratory Medicine vol 128 no 7 pp 785ndash7872004

[11] S Haavik M Nilsen T Thingstad et al ldquoSpecificity studiesof an antibody developed against a mucin-type glycoproteinrdquoGlycoconjugate Journal vol 16 no 3 pp 229ndash236 1999

[12] J F Codington S Haavik N Nikrui et al ldquoImmunologic quan-titation of the carcinoma specific human carcinoma antigen inclinical samplesrdquo Cancer vol 94 no 3 pp 803ndash813 2002

[13] H Clausen M Stroud J Parker G Springer and H Sen-Itiroh ldquoMonoclonal antibodies directed to the blood group aassociated structure galactosyl-A specificity and relation to thethomsen-friedenreich antigenrdquoMolecular Immunology vol 25no 2 pp 199ndash204 1988

[14] A S Palma Y Liu R A Childs et al ldquoThe human epithelialcarcinoma antigen recognized by monoclonal antibody AE3 isexpressed on a sulfoglycolipid in addition to neoplasticmucinsrdquoBiochemical and Biophysical Research Communications vol408 no 4 pp 548ndash552 2011

[15] DWang S Liu B J Trummer C Deng and AWang ldquoCarbo-hydratemicroarrays for the recognition of cross-reactivemolec-ular markers of microbes and host cellsrdquo Nature Biotechnologyvol 20 no 3 pp 275ndash281 2002

[16] D N Wang H T Chen J Liao et al ldquoTwo families ofmonoclonal antibodies to alpha(1ndash6)dextran VH1912 andVH9147 show distinct patterns of J kappa and JH minigeneusage and amino acid substitutions in CDR3rdquo The Journal ofImmunology vol 145 pp 3002ndash3010 1990

[17] DWang L Dafik RNolley et al ldquoAnti-oligomannose antibod-ies as potential serum biomarkers of aggressive prostate cancerrdquoDrug Development Research vol 74 no 2 pp 65ndash80 2013

[18] E A Kabat Blood Group Substances Their Chemistry andImmunochemistry Academic Press New York NY USA 1956

[19] G Vicari and E A Kabat ldquoImmunochemical studies onblood groups XLV Structures and activities of oligosaccharidesproduced by alkaline degradation of a blood group substancelacking ABHLea and Leb specificitiesrdquo Biochemistry vol 9no 17 pp 3414ndash3421 1970

[20] T Feizi E A Kabat G Vicari B Anderson and W L MarshldquoImmunochemical studies on blood groups XLIX The I anti-gen complex specificity differences among anti-I sera revealedby quantitative precipitin studies partial structure of the I deter-minant specific for one anti-I serumrdquo Journal of Immunologyvol 106 no 6 pp 1578ndash1592 1971

[21] T Feizi E A Kabat G Vicari B Anderson and W L MarshldquoImmunochemical studies on blood groups XLVII The Iantigen complexmdashprecursors in the A B H Lea and leb bloodgroup systemmdashhemagglutination-inhibition studiesrdquo Journal ofExperimental Medicine vol 133 no 1 pp 39ndash52 1971

[22] A M Wu K-H Khoo S-Y Yu Z Yang R Kannagi and WM Watkins ldquoGlycomic mapping of pseudomucinous humanovarian cyst glycoproteins identification of Lewis and sialylLewis glycotopesrdquo Proteomics vol 7 no 20 pp 3699ndash3717 2007

[23] J F Codington NMass S Haavik and D Norway ldquoAntibodiesto human carcinoma antigenrdquo US Patent no 5693763 1997

[24] C Gao Y Liu H Zhang et al ldquoCarbohydrate sequence of theprostate cancer-associated antigen F77 assigned by a mucin O-glycomedesigner arrayrdquoTheJournal of Biological Chemistry vol289 no 23 pp 16462ndash16477 2014

[25] M Nonaka M N Fukuda C Gao et al ldquoDetermination ofcarbohydrate structure recognized by prostate-specific F77monoclonal antibody through expression analysis of glycosyl-transferase genesrdquoThe Journal of Biological Chemistry vol 289no 23 pp 16478ndash16486 2014

[26] F Maisonrouge-McAuliffe and E A Kabat ldquoImmunochemicalstudies on blood groups Fractionation heterogeneity andchemical and immunochemical properties of a blood groupsubstancewith B I and I activities purified fromhumanovariancyst fluidrdquo Archives of Biochemistry and Biophysics vol 175 no1 pp 71ndash80 1976

[27] P Z Allen and E A Kabat ldquoImmunochemical studies on bloodgroups XXII Immunochemical studies on the nondialyzableresidue from partially hydrolyzed blood group A B and O(H)substances (P1 fractions)rdquo Journal of immunology vol 82 no 4pp 340ndash357 1959

[28] S K Sikder E A Kabat D D Roberts and I J GoldsteinldquoImmunochemical studies on the combining site of the bloodgroup A-special lima bean lectinrdquo Carbohydrate Research vol151 pp 247ndash260 1986

[29] I Brockhausen ldquoMucin-type O-glycans in human colon andbreast cancer glycodynamics and functionsrdquo EMBO Reportsvol 7 no 6 pp 599ndash604 2006

[30] F Maisonrouge-McAuliffe and E A Kabat ldquoImmunochemicalstudies on blood groups Structures and immunochemicalproperties of oligosaccharides from two fractions of bloodgroup substance from human ovarian cyst fluid differing in B Iand i activities and reactivity toward concanavalin Ardquo Archivesof Biochemistry and Biophysics vol 175 no 1 pp 90ndash113 1976

[31] F Maisonrouge-McAuliffe and E A Kabat ldquoImmunochemicalstudies on blood groups Heterogeneity of oligosaccharides lib-erated by degradation with alkaline borohydride of two humanovarian cyst fractions differing in B I and i activities and inreactivity toward concanavalin Ardquo Archives of Biochemistry andBiophysics vol 175 no 1 pp 81ndash89 1976

[32] K Tadano and I Ishizuka ldquoIsolation and characterization of thesulfated gangliotriaosylceramide from rat kidneyrdquo The Journalof Biological Chemistry vol 257 no 3 pp 1482ndash1490 1982

[33] T B H Geijtenbeek R Torensma S J van Vliet et al ldquoIden-tification of DC-SIGN a novel dendritic cell-specific ICAM-3


receptor that supports primary immune responsesrdquo Cell vol100 no 5 pp 575ndash585 2000

[34] H Feinberg D A Mitchell K Drickamer and W I WeisldquoStructural basis for selective recognition of oligosaccharidesby DC-SIGN and DC-SIGNRrdquo Science vol 294 no 5549 pp2163ndash2166 2001

[35] A S Powlesland E MWard S K Sadhu Y Guo M E Taylorand K Drickamer ldquoWidely divergent biochemical propertiesof the complete set of mouse DC-SIGN-related proteinsrdquo TheJournal of Biological Chemistry vol 281 no 29 pp 20440ndash20449 2006

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


Given that HAE3 has been analyzed using a large col-lection of synthetic neoglycoconjugates (119899 = 492) [14]this study focused on identification of the potential naturalligands of HAE3 In essence we produced a comprehensivecarbohydrate microarray using a large collection of purifiednatural carbohydrate antigens for screening These includeA B O Lewisab I and i and the blood group precursorsof various biological origins As summarized below we haverevealed that HAE3 is specific for a blood group precursorcryptic epitope that is normally hidden in the cores or internalchains of blood group substances but becomes differentiallyexpressed in human breast cancer cells

2 Material and Methods

21 Antigens and Antibodies A preparation of humancarcinoma-associated antigen (HCA) (1) was kindly pro-vided by Dr Zeqi Zhou of Egenix (Millbrook NY) Themurine hybridoma IgM antibody HAE3 was produced bymouse immunization (C57BLJ) with asialoepiglycanin (85)[4 5] A preparation of purifiedHAE3 protein was purchasedfrom RA Biosources Inc (Belmont CA) Carbohydrateantigens applied in this study are listed in Supplemen-tary Table 1 in Supplementary Material available online athttpdxdoiorg1011552015510810

22 Cell Lines Cancer cell lines used include breast-derived(T-47D SK-BR-3 MCF-7 BT-549 MD-AMB-231 and MD-AMB-468) lung-derived (A549) or prostate-derived (PC3)epithelial tumor cell lines and a skin-derived melanomaSKMEL-28 All tumor cell lines were acquired from ATCC

23 Carbohydrate Microarrays Microarray assays were per-formed as described [15] In brief a microarray robot(PIXSYS 5500C Cartesian Technologies Irvine CA) wasused to spot antigen preparations onto glass slides precoatedwith nitrocellulose polymer (FAST Slides Schleicher ampSchuell KeeneNH)Theprintedmicroarrayswere incubatedat room temperature with HAE3 (IgM) antibody at 5 120583gmLin 1 (wtvol) BSA in PBS containing 005 (wtvol) NaN

3

and 005 (volvol) Tween-20 An R-phycoerythrin- (R-PE-) conjugated affinity-purified F(ab1015840) fragment of goatanti-mouse IgM secondary antibody preparation (RocklandImmunochemicals Inc PA) was applied at 20 120583gmL toreveal HAE3-specific staining signal Fluorescence intensityvalues for each array spot and its background were calcu-lated using ScanArray Express software SAS Institutersquos JMP-Genomics software package (httpwwwjmpcom CaryNC) was used for microarray data standardization andstatistical analysis Results of the microarray assay are shownas the means of fluorescent intensities (MFIs) of triplicatedetections of given antigen preparations (Figure 1(a) andTable 1)

24 Fluorescence-Activated Cell Sorting (FACS) AnalysisHAE3 (IgM 50 120583gmL) and an isotype control mAb 9147(IgM anti-120572(16)dextran 50 ugmL) [16] were applied tostain tumor cell lines The R-PE-conjugated goat anti-mouse

IgM antibody preparation described above was applied inthe second staining step to reveal the antigen-captured IgMFACS data were collected with LSR-II (BD Bioscience SanJose CA) and analyzed with FlowJo (TreeStar Inc AshlandOR)

25 Enzyme-Linked Immunosorbent Assay (ELISA) andELISA Inhibition Assays Carbohydrate-specific ELISA andELISA inhibition assays were performed as described [1517] In brief glycoprotein antigen preparations were dilutedin 01M sodium bicarbonate buffer solution pH 96 forcoating on ELISA microplates (NUNC MaxiSorp ThermoScientific Rochester NY) followed by blocking using 1 BSAand 1X Phosphate Buffered Saline Tween-20 (PBST) MAbHAE3 (IgM) (25120583gmL) and biotinylated PNA (20120583gmL)were diluted in 1 BSA PBST for the ELISA binding assayThe bound HAE3 and PNA were revealed by an alkalinephosphatase- (AP-) conjugated goat anti-mouse IgM and anAP-streptavidin conjugate (Sigma Chemical Co St LouisMO) respectively ELISA inhibition assays were performedwith EPGN (85) (1 120583UmL) coated to display the nativegpHAE3 epitope and a series of blood group reference antigensas potential inhibitors (250120583gmLeach) to compete with theHAE3 (10 120583gmL) binding of EPGN Percent inhibition wascalculated as follows inhibition = ((standard A minus blank A)minus (A with inhibitor minus A))(standard A minus blank A)

26 Bio-Gel Chromatography Bio-Gel P-10 filtration wasperformed following the manufacturersquos manual (Bio-RadLaboratories Hercules CA) with minor modifications Inbrief Tij II (20 from 2nd 10) substance was fractioned ina precalibrated Bio-Gel P-10 column The sizes of the Tij IIsubstance were measured based on the neutral sugar elutionprofile with reference to the calibrated saccharide molecularweight standards as shown in Figure 4

3 Results

31 Carbohydrate Microarray Identifies Blood Group Precur-sors as the Natural Ligands of HAE3 As shown in Figure 1and Table 1 an HCA preparation (ID 1 and 2) was spotted asa positive control forHAE3 activityThisHAE3+ glycoproteinpreparation was affinity-purified from cultural supernatantof the lung cancer cell line A549 using an HAE3-agarosecolumn (Egenix Millbrook NY) Given that the simple O-glycan core T disaccharide (Gal12057313GalNAc-) and its peptideconjugates were found to weakly but significantly inhibitHAE3 binding to epiglycanin we characterized a panel ofblood group substances that carry O-glycan cores in thismicroarray screening A number of blood group precursors(29ndash32) were plotted from 3 to 32 which was followedby other antigens from 33 to 78 and microarray printingand scanning calibration controls in 79 and 80 Blood groupsubstance reference reagents [18] used include the followingCyst9 and Cyst14 A active Beach phenol insoluble B activeHog H active JS phenol insoluble H and Leb active andN-1 20 from the second 10 Lea active Key blood groupprecursor references includeOGTij II BeachP1 andMcDon


1 HCA 01

2 HCA 05

29 Beach P1 (Ii)





2 3 4 5 6 7 8 91 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 8010 60

58

61

64

36373839404142434445464748495051525354555657

34

59

3332

6263

35

656667686970717273747576777879

3031


0

2000

4000

6000

8000

10000

12000

14000

16000

MFI

(a)


30 McDon P1

32 OG

2 HCA 05

1 HCA 01

80 Dex-FITC

(b)


Branch 1

Branch 2 Branch 3

Branch 4

A

B

Fuc

H

I (II)

darr 1205721 2

Galdarr 1205731 4


Gal


Galuarr 1205731 4


darr 1205731 43

II120573

II120573 II120573I120573


Gal1205721 rarr 3


(c)













HAE3 PNA


0

05

10

15

20

25

30

OD405

nm


0

10

20

30

40

OD405

nm


Tij II

0

20

40

60

80

100

Inhi

bitio

n (

)

B-P1

(29)

Tij I

I (31

)

H (2

2)

OG

(32)

EPG

N (8

5)

B (1

4)

Tn (8

3)

H (4

)

HCA

(1)

A (3

)

T (8

4)








4 Discussion



HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

(PCa)(BCa)T47D


Max

()

HAE3R-PEHAE3R-PE101 102 103 104100 101 102 103 104100

HAE3R-PE101 102 103 104100

HAE3R-PE101 102 103 104100

0

20

40

60

80

100SKMEL-28

(a)

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control


PR+ PR+ER+ ER+

MCF7T-47DHER2+

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100SK-BR-3

(b)







20 10 5 IM5MW (kDa)

40 6020 800


0

1

2

3

4

5

6

7

8

A490











Disclaimer


Acknowledgments


References













































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















1 HCA 01

2 HCA 05

29 Beach P1 (Ii)





2 3 4 5 6 7 8 91 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 8010 60

58

61

64

36373839404142434445464748495051525354555657

34

59

3332

6263

35

656667686970717273747576777879

3031


0

2000

4000

6000

8000

10000

12000

14000

16000

MFI

(a)


30 McDon P1

32 OG

2 HCA 05

1 HCA 01

80 Dex-FITC

(b)


Branch 1

Branch 2 Branch 3

Branch 4

A

B

Fuc

H

I (II)

darr 1205721 2

Galdarr 1205731 4


Gal


Galuarr 1205731 4


darr 1205731 43

II120573

II120573 II120573I120573


Gal1205721 rarr 3


(c)













HAE3 PNA


0

05

10

15

20

25

30

OD405

nm


0

10

20

30

40

OD405

nm


Tij II

0

20

40

60

80

100

Inhi

bitio

n (

)

B-P1

(29)

Tij I

I (31

)

H (2

2)

OG

(32)

EPG

N (8

5)

B (1

4)

Tn (8

3)

H (4

)

HCA

(1)

A (3

)

T (8

4)








4 Discussion



HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

(PCa)(BCa)T47D


Max

()

HAE3R-PEHAE3R-PE101 102 103 104100 101 102 103 104100

HAE3R-PE101 102 103 104100

HAE3R-PE101 102 103 104100

0

20

40

60

80

100SKMEL-28

(a)

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control


PR+ PR+ER+ ER+

MCF7T-47DHER2+

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100SK-BR-3

(b)







20 10 5 IM5MW (kDa)

40 6020 800


0

1

2

3

4

5

6

7

8

A490











Disclaimer


Acknowledgments


References













































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of























HAE3 PNA


0

05

10

15

20

25

30

OD405

nm


0

10

20

30

40

OD405

nm


Tij II

0

20

40

60

80

100

Inhi

bitio

n (

)

B-P1

(29)

Tij I

I (31

)

H (2

2)

OG

(32)

EPG

N (8

5)

B (1

4)

Tn (8

3)

H (4

)

HCA

(1)

A (3

)

T (8

4)








4 Discussion



HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

(PCa)(BCa)T47D


Max

()

HAE3R-PEHAE3R-PE101 102 103 104100 101 102 103 104100

HAE3R-PE101 102 103 104100

HAE3R-PE101 102 103 104100

0

20

40

60

80

100SKMEL-28

(a)

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control


PR+ PR+ER+ ER+

MCF7T-47DHER2+

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100SK-BR-3

(b)







20 10 5 IM5MW (kDa)

40 6020 800


0

1

2

3

4

5

6

7

8

A490











Disclaimer


Acknowledgments


References













































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

(PCa)(BCa)T47D


Max

()

HAE3R-PEHAE3R-PE101 102 103 104100 101 102 103 104100

HAE3R-PE101 102 103 104100

HAE3R-PE101 102 103 104100

0

20

40

60

80

100SKMEL-28

(a)

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control

HAE3Isotype control


PR+ PR+ER+ ER+

MCF7T-47DHER2+

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100

Max

()

0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE0 103 104 105minus103

HAE3R-PE

0

20

40

60

80

100SK-BR-3

(b)







20 10 5 IM5MW (kDa)

40 6020 800


0

1

2

3

4

5

6

7

8

A490











Disclaimer


Acknowledgments


References













































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of


















20 10 5 IM5MW (kDa)

40 6020 800


0

1

2

3

4

5

6

7

8

A490











Disclaimer


Acknowledgments


References













































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















research article carbohydrate microarrays identify blood...

Documents