ceacam1-3s drives melanoma cells into nk cell-mediated ...tumor and stem cell biology ceacam1-3s...

Tumor and Stem Cell Biology

CEACAM1-3S Drives Melanoma Cells into NKCell-Mediated Cytolysis and Enhances PatientSurvivalNico Ullrich1,2, Anja Heinemann1,2, Elena Nilewski3, Inka Scheffrahn4, Joachim Klode1,2,Andr�e Scherag5, Dirk Schadendorf1,2, Bernhard B. Singer3, and Iris Helfrich1,2

Abstract

CEACAM1 is a widely expressed multifunctional cell–celladhesion protein reported to serve as a poor prognosis markerin melanoma patients. In this study, we examine the functionaland clinical contributions of the four splice isoforms of CEA-CAM1. Specifically, we present in vitro and in vivo evidence thatthey affect melanoma progression and immune surveillance ina negative or positive manner that is isoform specific in action.In contrast with isoforms CEACAM1-4S and CEACAM1-4L,expression of isoforms CEACAM1-3S and CEACAM1-3L isinduced during disease progression shown to correlate withclinical stage. Unexpectedly, overall survival was prolonged in

patients with advanced melanomas expressing CEACAM1-3S.The favorable effects of CEACAM1-3S related to enhancedimmunogenicity, which was mediated by cell surface upregu-lation of NKG2D receptor ligands, thereby sensitizing mela-noma cells to lysis by natural killer cells. Conversely, CEA-CAM1-4L downregulated cell surface levels of the NKG2Dligands MICA and ULBP2 by enhanced shedding, therebypromoting malignant character. Overall, our results define thesplice isoform-specific immunomodulatory and cell biologicfunctions of CEACAM1 in melanoma pathogenesis. Cancer Res;75(9); 1897–907. �2015 AACR.

IntroductionMelanoma is one of themost aggressive types of cancer, and its

prevalence has risen faster than any othermalignant disease in thewestern world. Tumor progression starts very early, resulting in amedian survival of 6 to 12 months in patients with advancedmelanoma (1). Melanoma progression is a complex multistepprocess orchestrated by a variety of cellular factors, includingdysregulation of cell adhesion molecules (2). Evidence hasamassed that expression of the multifunctional carcinoembryo-nic antigen (CEA)-related cell adhesion molecule 1 (CEACAM1)may be involved during this process (3). CEACAM1 belongs tothe CEA family within the immunoglobulin superfamily (4)and is expressed in human epithelial (5, 6), endothelial (7), andhematopoietic cells (8, 9). Downregulation of CEACAM1 has

been reported in colon (10), prostate (11), and breast cancer(12). Melanocytes are CEACAM negative (13), while highexpression levels of CEACAM1 have been detected in melano-ma (14) and adenocarcinoma (15). CEACAM1 was reported toinhibit cell proliferation in several tumor entities, exceptingmelanoma (13, 16, 17). On the tumor cell surface, the CEA-CAM1 protein has been shown to interact directly with CEA-CAM1 on immune cells, leading to functional inhibition ofthose cells (18–21), or to indirectly modulate cancer cellimmunogenicity by downregulating their cell surface expres-sion of MHC class I-related molecule A and B (MICA/B) andUL-16–binding protein1 (ULBP; ref. 22), both ligands for thenatural killer (NK) gene complex group 2 member D (NKG2D)receptor expressed on malignant cells (23, 24).

Although 12 different splice variants of the human CEA-CAM1 gene have been reported, only four are shown to beexpressed at mRNA level (25, 26). The CEACAM1-4 variantsconsist of four, CEACAM1-3 of three heavily glycosylated extra-cellular domains. Both isoforms are trans-membrane anchoredand carry either the long (L, 73 aa) or short (S, 10 aa) cyto-plasmic domain (27).

CEACAM1 has been controversially discussed as tumor sup-pressor and driver of invasion (28, 29). Most studies contributingto this discussion were focused on total CEACAM1 or CEACAM1-4L/S. In 2002, Thies and colleagues showed that the CEACAM1protein expression in primary cutaneous melanoma predicts thedevelopment of metastatic disease (30). Furthermore, levels ofsoluble CEACAM1 in sera from melanoma patients have beenshown to inversely correlate with overall survival (31, 32). Thisspurred discussion of CEACAM1 as a more specific and sensitivebiomarker than those currently used, including melan-A, S100b,and HMB45, and has implicated CEACAM1 as a potential noveltherapeutic target (14). Nevertheless, none of these studiesaddressed the impact of the four CEACAM1 splice variants.

1Skin Cancer Unit of the Dermatology Department, Medical Faculty,West German Cancer Center, University Duisburg-Essen, Essen, Ger-many. 2German Cancer Consortium (DKTK), Heidelberg, Germany.3Institute of Anatomy, University Hospital, University Duisburg-Essen,Essen, Germany. 4Institute for Gastroenterology and Hepatology,Medical Faculty, University Duisburg-Essen, Essen, Germany. 5ClinicalEpidemiology, Integrated Research and Treatment Center, Center forSepsis Control and Care (CSCC), Jena University Hospital, Jena,Germany.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

B.B. Singer and I. Helfrich contributed equally to this article.

Corresponding Authors: Iris Helfrich, Skin Cancer Unit of the DermatologyDepartment, Medical Faculty, University Duisburg-Essen, Hufelandstrasse 55,D-45147 Essen, Germany. Phone: 49-201-723-1648; Fax: 49-201-723-5525;E-mail: [email protected]; and Bernhard B. Singer, E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-14-1752

�2015 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 1897

on April 22, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 5, 2015; DOI: 10.1158/0008-5472.CAN-14-1752

http://cancerres.aacrjournals.org/

Furthermore, the clinical impact and the precise mechanism bywhich the four CEACAM1 variants modulate melanoma progres-sion are completely unknown. Our present study analyzed thebiologic function and clinical relevance of the individual CEA-CAM1 splice variants in humanmelanoma biopsies of increasingdisease stages and in cell lines, established from patient's metas-tases. We show for the first time that CEACAM1-3S, CEACAM1-3L, CEACAM1-4S, and CEACAM1-4L differentially affects cellularfunction and melanoma progression. Furthermore, we demon-strate that the expression of CEACAM1-3S correlates significantlywith the clinical stage and strikingly with a prolonged patientoverall survival. Finally, we provide strong evidences that CEA-CAM1-3S triggers melanoma cells for NK cell-mediated cytolysisby upregulating cell surface expression of MICA and ULBP2,whereas CEACAM1-4L causes the contrary effect due to enhancedshedding of both NKG2D ligands (NKG2DLs).

Materials and MethodsSomeMaterials andMethods are detailed in the Supplementary

Data.

Tissues and cell cultureMalignant melanoma cell lines and biopsies were obtained

from the Skin Cancer Biobank of the Dermatology Department,University Hospital Essen, Germany. Informed patient consentand the appropriate Institutional Review Board approval wereobtained for all patients. Clinical information including age,gender, stage of disease, tumor load, and survival time wasdocumented and retrieved from the electronic database (Achi-ver Anyware Medical, Achiver Software). Disease staging wasperformed according to the staging criteria of the AmericanJoint Committee on Cancer (AJCC; ref. 33). Cell lines wereestablished from malignant melanoma as described before(34, 35). RPMI-1640 supplemented with 10% FCS, 1% peni-cillin/streptomycin, and 1% L-glutamine (all from PAA Labo-ratories) was used as culture medium. All cell lines weremaintained at 37�C and 5% CO2 and regularly tested formycoplasma infection.

Plasmid constructs and transfectionThe coding sequences for CEACAM1-3L (NM_001184813.1),

CEACAM1-3S(NM_001184816.1),CEACAM1-4L(NM_001712.4),and CEACAM1-4S (NM_001024912.2) were cloned into thepcDNA3.1(�) Neo plasmid (Invitrogen) and verified bysequencing. Constructs were transfected into the Ma-Mel-86acell line using Metafectene (Biontex) according to the manu-facturer's protocol, and single clones selectively grown in medi-um containing 1 mg/mL G418 (Carl Roth).

xCELLigence SystemThe Real-Time Cell Analyzer System was used to analyze

cellular functions (36) and cytotoxicity (37). The experimentswere performed as described by the manufacturer's instructions.In short, the half-maximum cell index (CI50) between 0 hour and70 hours (CI50) was used for determination of statistical differ-ences calculated by the Student t test. Cytotoxicity results arepresented as percentages of cytolysis determined fromnormalizedcell index (nCI): % of specific lysis ¼ [nCI (no effector) �nCI(effector)]/nCI (no effector)� 100. Experiments were performedin duplicates.

Please refer to the detailed Supplementary ExperimentalProcedures.

ImmunofluorescenceMelanoma cells on coverslips were grown to 90% con-

fluency. Cells were fixed either with methanol or 4% PFA,blocked with 3% BSA/PBS, then incubated at 4�C overnightwith primary antibodies in 1.5% BSA/PBS. Primary antibodieswere visualized by fluorescent labeled secondary antibodies.Nuclei were counterstained with DAPI (Carl Roth), and fluo-rescence microscopy was conducted on a Leica SP8 confocalmicroscope (Leica), Zeiss AxioObserver.Z1 with Apotome andZeiss ELYRA PS.1 using SIM technology (Zeiss). Antibodyinformation is provided in Supplementary Materials and Meth-ods section.

Flow cytometrySurface expressions were analyzed using a FACScalibur flow

cytometer and the CellQuest Pro software (BD Biosciences),Gallios system (Beckman Coulter) equipped with FlowJo soft-ware and Accuri C6 flow cytometer (BD Biosciences). Cells wereharvested, incubated with primary antibodies and fluorescence-conjugated secondary antibodies. Isotype matched control anti-body staining served as negative control. Dead cells were excludedfrom measurements by propidium iodide staining. Antibodyinformation is provided in Supplementary Materials and Meth-ods section.

MICA ELISASolubleMICA (sMICA)was quantified, using theDuoSet ELISA

Development System (R&D Systems) according to the manufac-turer's protocol. Melanoma cells were starved for 24 hours.Supernatant was collected and centrifuged to remove cellulardebris. Levels of sMICA were normalized to cell number at timeof harvesting.

Enrichment of primary polyclonal NK cellsCD3�CD56þ NK cells were enriched from PBMCs of healthy

donors usingMACS technology (Miltenyi Biotec) according to themanufacturer's protocol. Enrichment of CD3�CD56þ NK cellswas confirmed by flow cytometry and ranged between 90% and95%. NK cells were cultured in the presence of 200 IU/ml IL2(Chiron) before analyses.

FACS-based cytotoxicity assayTo determine specific lysis of melanoma cells by NK cells, flow-

cytometric analysis was performed (38). In brief, melanoma cells(5 � 106 cells/mL) were labeled with 2 nmol/L CFSE for 10minutes (Invitrogen). Then antibody pretreated NK cells wereadded to 5 � 104 of CFSE-labeled melanoma cells at variouseffector to target ratios for 3 hours. 7-AAD (Cayman Chemicals)was added to each sample according to the manufacturer's pro-tocol. Probes were measured directly in a Gallios flow cytometer(Beckman Coulter). Gating strategy was used as described inHeinemann and colleagues (38).

Statistical analysisQuantitative expression values between two groups were

compared using Wilcoxon-Mann–Whitney tests. In the casethat comparisons were to a group with no observable

Ullrich et al.

Cancer Res; 75(9) May 1, 2015 Cancer Research1898




expression values (stage I/II of CEACAM1-3S), we applied theone-sided t test (against an expected value of 0). We used theStudent t test to compare two experimental conditions infunctional experiments. Spearman correlations (r) were usedto assess associations between gene expression levels and tumorstages. Time-to-event data (from melanoma diagnoses untildeath or the last observation in case of censoring) were dis-played using the method of Kaplan–Meier. Given the problemsof selecting an appropriate cutoff/cut point value, which isoften data driven, we decided to display our results for the"presence of any transcripts" in contrast to "no transcripts."Subgroups in Kaplan–Meier plots were compared using log-rank statistics; in addition, we used Cox regression to therelated estimate HR. All reported P values are nominal andtwo sided. We applied a significance level of 5% and did notadjust for multiple testing.

ResultsAnalysis of CEACAM1 variants in malignant melanomaidentified correlation of CEACAM1-3S with patient clinicalstage and overall survival

Total CEACAM1 expression has been shown to be signifi-cantly higher in melanomas compared with benign nevi (39).To examine the presence of CEACAM1-3S, CEACAM1-3L, CEA-CAM1-4S, and CEACAM1-4L in melanoma, we analyzed 46cell lines established from human melanoma metastases usingRT-PCR. We detected CEACAM1 variant expression in 33 of46 cell lines (72%, Supplementary Table S1). Remarkably, allCEACAM1-positive cell lines expressed CEACAM1-4L in com-bination with other variants. We assessed relative variantexpression by densitometric quantification. For each case (bothCEACAM1-4 and CEACAM1-3), the longer (L)-variant wasexpressed at a higher level in the melanoma cell lines (Sup-plementary Table S1).

Nextwe assessed the impact ofCEACAM1splice variants duringmelanoma progression, by analyzing variant expression in 51biopsies from melanoma patients that spanned stages I–IVaccording to the AJCC. First, in 45 of 51 (88%) biopsies, weobserved CEACAM1 variant expression (Table 1 and Supplemen-tary Table S2). Surprisingly, both CEACAM1-3 variants wereabsent in stage I/II melanomas, with only one exception thatexpressed CEACAM1-3L (Fig. 1A and Supplementary Table S2).CEACAM1-3S and CEACAM1-3L expression was induced duringmelanoma progression, and significantly positively correlatedwith clinical stage, reaching the highest expression levels in stageIV tumors (CEACAM1-3S stage III: P¼ 0.031, stage IV: P¼ 0,002;CEACAM1-3L stage III: P¼ 0.002, stage IV: P < 0.0001 vs. stage I/II; Fig. 1A). Furthermore, expression of both CEACAM1-3 iso-forms was even higher in late-stage melanoma (stage IV) com-pared with stage III biopsies (CEACAM1-3S P ¼ 0.004; CEA-CAM1-3L P ¼ 0.015, Fig. 1A). CEACAM1-4S expression wasdetected in 40 of 51 (78%) biopsies but did not vary accordingto stage (Fig. 1A and Supplementary Table S2). CEACAM1-4Lexpressionwas detected in all CEACAM1-positive patient biopsies(88%; Supplementary Table S2) and was significantly inducedduring early disease progression from stages I/II to stage III (P ¼0.030, Fig. 1A). Moreover, CEACAM1-4L was either expressedalone or in combination with other isoforms, while no otherisoform was expressed without CEACAM1-4L (SupplementaryTable S2). In correspondence to our in vitro observations, CEA-CAM1-3S exhibits the lowest expression intensities, whereasCEACAM1-4L was predominantly expressed (Fig. 1A and Sup-plementary Table S2).

Then we applied the CEACAM1 isoforms data to analyze theprognostic power. Surprisingly, patients with advanced melano-mas expressing CEACAM1-3S showed significantly (P ¼ 0.039)prolonged overall survival compared with patients with melano-mas lacking CEACAM1-3S expression (HR, 0.43; 95% confidenceinterval, 0.19–0.98; Fig. 1B), whereas expression of CEACAM1-

Table 1. Characteristics of melanoma patients

Biopsies positive for specific CEACAM1 splice formsTotal (N ¼ 51) CEACAM1-3S (N ¼ 26) CEACAM1-3L (N ¼ 38) CEACAM1-4S (N ¼ 40) CEACAM1-4L (N ¼ 45)

Age at biopsy removal, yMean 62 62 59 60 61Range 34–90 34—85 34–85 34–90 34–90

SexMale 28 13 20 22 24Female 23 13 18 18 21

Clinical stagea

Stage I/II 8 0 1 4 4Stage III 21 8 17 17 20Stage IV 22 18 20 19 21

Breslow's depth (mm)pT1 (�1.00) 9 3 3 4 5pT2 (1.01–2.00) 13 10 11 11 13pT3 (2.01–4.00) 14 6 12 12 12pT4 (> 4.00) 9 5 7 8 9Unknown 6 2 5 5 6

LocalizationSkin 42 21 30 33 36Other 9 5 8 7 9

UlcerationYes 18 7 13 15 17No 13 10 11 10 11Unknown 20 9 14 15 17

Abbreviations: y, years; mm, millimeter; N, number of biopsies.aAccording to AJCC criteria 2010.

CEACAM1 Splice Variants in Melanoma

www.aacrjournals.org Cancer Res; 75(9) May 1, 2015 1899




3L, CEACAM1-4S, and CEACAM1-4L did not affected the clinicaloutcome (data not shown).

CEACAM1 variants localize to different intracellularcompartments in melanoma cells

To explore CEACAM1 isoform functionality, we stably trans-fected CEACAM1-3S, CEACAM1-3L, CEACAM1-4S, or CEA-CAM1-4L, respectively, into the CEACAM-negative cell line Ma-Mel-86a, established from a stage III metastases (Fig. 2). Emptyvector Ma-Mel-86a-transfection served as a negative control.Expression of exogenous CEACAM1 isoforms was identified byRT-PCR (Fig. 2A), Western blot analysis (Fig. 2B), and flowcytometry (Fig. 2C).

Next, we investigated the subcellular localization of each CEA-CAM1 isoform using IHC. Beside their surface expression, theCEACAM1-3S and CEACAM1-3L isoforms were primarilydetected in vesicle-like structures that accumulated around thenucleus (Fig. 3A, I and II), whereas the CEACAM1-4S and CEA-

CAM1-4L isoforms were predominantly localized to sites of cell–cell contact on the cell surface (Fig. 3A, III and IV, arrowheads). NoCEACAM1stainingwas detectable in control cells transfectedwithempty vector (data not shown). Interestingly, CEACAM1-3S- andCEACAM1-3L-positive vesicle-like structures were arranged like a"string of pearls" orientated toward sites of cell–cell contact (Fig.3A, I and II, arrows), pointing to a recruitment of both variants toareas of cellular interactions. To analyze how this linear orienta-tion is achieved, CEACAM1-3L and CEACAM1-3S transfectantswere immunostained for total CEACAM1 and F-actin. Costainingof CEACAM1-3S transfectants revealed that CEACAM1-positivevesicular structures associated with F-actin fibers in cell protru-sions extending toward contact points with adjacent cells (Fig.3B). Similar observations were made in CEACAM1-3L transfec-tants (data not shown). Taken together, all CEACAM1 isoformswere predominantly expressed at cell–cell contacts, but cellularlocalization of the CEACAM1 isoforms varied andwas dependenton the presence of the extracellular A2 domain.

A

B

CEACAM1-3L

0

10

20

30

40*

*****

Exp

ress

ion

rela

tive

to to

tal C

EA

CA

M1

(%)

CEACAM1-3S

0

5

10

15

20 **

*°**°

Exp

ress

ion

rela

tive

to to

tal C

EA

CA

M1

(%)

CEACAM1-4S

0

20

40

60

80

100

Exp

ress

ion

rela

tive

to to

tal C

EA

CA

M1

(%)

CEACAM1-4L

0

20

40

60

80

100

Exp

ress

ion

rela

tive

to to

tal C

EA

CA

M1

(%)

*

Time from diagnosis (years)

Ove

rall

surv

ival

pro

babi

lity

(%)

0 2 4 6 8 100

20

40

60

80

100

ExpressedNot expressed

P = 0.039

Figure 1.CEACAM1-3S expression correlateswith clinical stage and overall survivalin patients with melanoma. A, scatterplot for expression of indicatedCEACAM1 splice variant relative tototal CEACAM1. Patients are groupedaccording to their clinical stages,based on the AJCC system (stage I/II,N ¼ 8; stage III, N ¼ 21; stage IV,N ¼ 22). Mann–Whitney Test; � , one-sampled t test to hypothetical value¼0; � ,P<0.05; �� ,P<0.01; ��,P<0.001;error bars, mean � SEM. B, Kaplan–Meier curve of patients (N ¼ 51) with(N¼ 26, solid line) or without (N¼ 25,dotted line) CEACAM1-3S expression.Short vertical lines represent censoredobservations. Statistical significancewas calculated by log-rank test.

Ullrich et al.





CEACAM1 isoforms differently affectmigration and invasion ofmelanoma cells

CEACAM1-4L expression has been shown to enhance themigratory capacity and invasive potential of melanoma cellswithout affecting cell proliferation (13). We found that expres-sion of the different CEACAM1 isoforms transfected into Ma-Mel-86a did also not alter proliferation (data not shown). Toinvestigate the specific contributions of each CEACAM1 iso-form to cellular motility, we monitored the CEACAM1 trans-fectants in real time by using xCELLigence impedance mea-surement. Expression of CEACAM1-4S (P ¼ 0.009) and CEA-CAM1-4L (P ¼ 0.025) had the strongest migration enhancinginfluence on melanoma cells, compared with control cells (Fig.4A and B). Also, CEACAM1-3L significantly increased migratorycapacity, but to a lesser extent (P ¼ 0.023, Fig. 4A and B). Inaccordance to this finding, expression of CEACAM1-4S (P ¼0.006) and CEACAM1-4L (P ¼ 0.005) resulted in enhancedinvasive behavior, which was only slightly increased by CEA-CAM1-3L (Fig. 4C and D). Interestingly, CEACAM1-3S expres-sion significantly decreased both cell migration (P ¼ 0.031, Fig.4A and B) and invasion (P ¼ 0.020, Fig. 4C and D) comparedwith control cells. Analyses of a second set of independentlygenerated CEACAM1 transfectants confirmed these results(data not shown). Now we hypothesized that activity of matrix

metalloproteinases (MMP), enzymes known to be key regula-tors of invasive potential in melanoma cells (40, 41), couldbe involved in modulating the cellular function of the indi-vidual CEACAM1 isoforms. Consequently, we treated ourCEACAM1 transfectants with the MMP inhibitor marimastat(42–44), and reassessed invasive capacity after inhibition ofMMP activity. In accordance to our hypothesis, marimastattreatment impaired the invasive promoting effect of CEA-CAM1-4S or CEACAM1-4L (Fig. 5).

CEACAM1-3S and CEACAM1-4L direct tumor immunogenicityby deregulating MICA and ULBP2 expression

It has been reported that the CEACAM1 expression influencesthe immunogenicity of cancer cells by modulating the surfaceexpression of NKG2D ligands. These data prompted us to analyzethe impact of the four CEACAM1 isoforms on the expression ofNKG2D ligand MICA and ULBP2 by flow cytometry. Althoughexpression of CEACAM1-3L or CEACAM1-4S did not modulateMICA andULBP2 expression, significant upregulation ofMICA (P¼ 0.006) and ULBP2 (P ¼ 0.011) on CEACAM1-3S transfectantswas detected comparedwith control cells (Fig. 6A andB). Remark-ably, expression of CEACAM1-4L resulted in damped cell surfaceexpression of MICA (P ¼ 0.020) without affecting ULBP2 levels(Fig. 6A and B).

A

CEACAM1

β-Actin

RT-PCR Western blot

Eve

nts

CEACAM1

C

B

800 bp

500 bp

170 kDa

100 kDa

100

80

60

40

20

0

638

100 101 102 103 104

100

80

60

40

20

0

615

100 101 102 103 104

410

100

80

6040

200100 101 102 103 104

19210080

6040

200100 101 102 103 104

71008060

40

200100 101 102 103 104

Control

CEACAM1-3S CEACAM1-3L


Figure 2.Transfectants of the human melanomacell line Ma-Mel-86a show specificexpression of CEACAM1 splice variants.A, expression of CEACAM1-4L,CEACAM1-4S, CEACAM1-3L, andCEACAM1-3S transfected into Ma-Mel-86a was analyzed by RT-PCR. MultiplePCR products result from the lack of theA2 domain and the different length ofthe cytoplasmic domains (CEACAM1-4L¼ 779 bp; CEACAM1-4S ¼ 726 bp;CEACAM1-3L ¼ 491 bp; CEACAM1-3S ¼438 bp). Cells transfected with emptyvector served as controls. B, Westernblot analyses of the CEACAM1transfectants. b-actin served as loadingcontrol. C, flow-cytometric analyses ofCEACAM1 cell surface expression. Cellswere stained for CEACAM1 (thick line)and for isotype-matched control (thinline). Numbers indicate meanfluorescence intensity values. For A–C,one representative experiment out ofthree is shown.






To verify whether CEACAM1-3S indeed affected cell surfaceexpression of NKG2D ligands, MICA and ULBP2 expression wasdetermined by confocal microscopy. Expression of CEACAM1-3Sresulted in enhanced recruitment of MICA and ULBP2 to the Ma-Mel-86a-cell surface (Fig. 6C). Parallel we analyzed ligand expres-sion levels in the CEACAM1 transfectants using Western blotanalysis. MICA and ULBP2 expression in whole-cell lysates ofCEACAM1-3L, CEACAM1-3S, and CEACAM1-4S transfectantswas consistent with the flow-cytometric data (Fig. 6D). Further-more, we confirmed that CEACAM1-3S–transfected cells upregu-lated MICA and ULBP2 expression (Fig. 6D). Unexpectedly,CEACAM1-4L expressiondid not alter the total protein expressionofMICA and enhanced ULBP2 protein expression compared withcontrol cells (Fig. 6D). These data did not correlate with our flow-cytometric findings, unless the reduced MICA and ULBP2 surfacelevels in CEACAM1-4L–positive cells would appear due to shed-ding. Accordingly, we examined the levels of soluble MICA(sMICA) and ULBP2 (sULBP2) in media conditioned by theCEACAM1 transfectants using Western blot analysis, and exclu-

sively detected enhanced amounts of sMICA and sULBP2 inconditioned medium obtained from CEACAM1-4L–expressingcells (Fig. 6E). Quantification of sMICA by ELISA revealed anapproximately 2-fold elevation (P¼ 0.033) in conditionedmedi-um from CEACAM1-4L–expressing cells compared with controlcells (Fig. 6F). Noteworthy, marimastat treatment reduced thelevel of soluble MICA (P¼ 0.0034) to the control level (Fig. 6G).

Next we asked, whether the approximately 2-fold induction ofMICAandULBP2on the surface ofCEACAM1-3S–expressing cellswould affect melanoma cell sensitivity to NK cell-mediated rec-ognition and cytolysis. Thus, the cytolytic activity of NK cells wasmeasured by a CFSE/7-AAD cytotoxicity assay. As shown in Fig.6H and I, the sensitivity of CEACAM1-3S–expressing melanomacells to NK cell-mediated cytolysis was proportional to MICA andULBP2 cell surface expression. Cytolysis of CEACAM1-3S cellswere significantly enhanced compared with control cells at aneffector to target ratio of 10:1 (P ¼ 0.031, Fig. 6I) with a corre-sponding outcome using a 5:1 ratio (data not shown). Blockingthe NKG2D receptor by antibody significantly reduced NKcell-mediated cytolysis (control, P ¼ 0.0113; CEACAM1-3S, P¼ 0.0085) compared with cultures to which isotype controlantibody was added (Fig. 6I).

To analyze whether the reduced MICA-surface level on CEA-CAM1-4L transfectants could dampen the NK cell-mediated kill-ing, we performed xCELLigence-based cytotoxicity assays. Expres-sion of CEACAM1-3S resulted in significant upregulation of NKcell-mediated cytolysis comparedwith control transfectants (Sup-plementary Fig. S1A, P¼0.0016). The specific lysis of control (P¼0.0257) and CEACAM1-3S transfectants (P ¼ 0.0137) wasreduced by blockage of NKG2D (Supplementary Fig. S1A). Theexpression of CEACAM1-4L revealed no significant modulatoryeffect althoughblockage ofNKG2Dsignificantly reduced cytolysis(P ¼ 0.0141, Supplementary Fig. S1A). Nevertheless, data pre-sented in Supplementary Fig. S1A are averaged values of fiveindependent experiments. Therefore, NK cells of five differentdonors have been used. Notably, two experiments showedreduced killing, two enhanced killing, and one no alteration ofthe NK cell-mediated killing if CEACAM1-4L was expressed.

Furthermore, we validated the expression of theNKG2D ligandULBP1 and ULBP3, the DNAM-1 ligands CD112, CD155 andligands for NKp30 and NKp46 receptors on CEACAM1 transfec-tants and control cells by flow cytometry (Supplementary Fig. S2).We found enhanced expression for CD155 in Ma-Mel-86a-CEA-CAM1-3S (P¼ 0.0124),whereas expression of CEACAM1-4S (P¼0.0442) and CEACAM1-4L (P ¼ 0.0117) resulted in decreasedexpression (Supplementary Fig. S2A, I). Also, CD112 was foundto be reduced inCEACAM1-4S (P¼ 0.0169) andCEACAM1-4L (P¼ 0.0424) transfectants (Supplementary Fig. S2A, II). NKp30 andNKp46 did not show altered expression (Supplementary Fig. S2A,III and IV). On the basis of these findings, we focused thecytotoxicity assays on CD155, blocking the NKG2D receptor,the DNAM-1 receptor, or both on NK cells before coculture withthe CEACAM1-3S transfectant and control cells, respectively(Supplementary Fig. S2B) . First, 2 hours after NK cell addition(E/T ¼ 1:1), anti-NKG2D decreased the lysis of CEACAM1-3Stransfectant (6.6%), and to lesser extend the vector control(1.7%). This confirmed our prior observation (Fig. 6H). Weobserved similar effects utilizing anti-DNAM-1, resulting inreduced cytolysis of CEACAM1-3S (15%) and control (5.1%)transfectants under corresponding conditions. Nevertheless,simultaneous blockage of NKG2D and DNAM-1 resulted in

Phalloidin CEACAM1

DAPI Merge

I II

III IV


CEACAM1-4LCEACAM1-4S

A

B

Figure 3.CEACAM1 splice variants localize to different cellular compartments. A,immunofluorescence-based assessment for cellular localization of CEACAM1(green) in the indicated CEACAM1 isoform transfectants. Nuclei werecounterstained using DAPI (blue). White arrowheads, CEACAM1 staining atcell–cell contact sites; white arrow, CEACAM1-positive vesicles arranged in aline toward cell–cell contacts. B, double staining for total CEACAM1 (mAb 4/3/17, green) and F-actin (phalloidin, red) in CEACAM1-3S transfectants; nucleiwere counterstained using DAPI (blue). Similar results were observed inCEACAM1-3L cells (data not shown). Representative images show maximumz-projection of confocal optical sections. Analyses were repeated three timesby using different passages of each transfectants. Scale bar, 10 mm.

Ullrich et al.





synergistic effects and reduced the NK cell-mediated lysis of bothtarget cell lines to a comparable level (CEACAM1-3S: 26%;control: 21.6%, Supplementary Fig. S2B, 2 hours after NK celladdition, E/T ¼ 1:1).

DiscussionTumor progression is a complexmultistep process orchestrated

by a variety of cellular factors, such as cell adhesionmolecules, butalso influenced by host-derivedmicroenvironmental cell popula-tions including cells of the immune system. Most recent studiesdiscussed CEACAM1 as a novel promising target for immuno-therapy of malignant melanoma patients (3, 45).

Here, we report CEACAM1-3S, CEACAM1-3L, CEACAM1-4S,and CEACAM1-4L expression in melanoma cell lines and biop-sies. The isoform expression pattern differed during malignantprogression. Early in tumor establishment (stage I/II), CEACAM1-4L and CEACAM1-4S were exclusively expressed in 50% ofmelanoma biopsies, whereas CEACAM1-4L was expressed toa higher extend in metastatic melanoma compared with pri-mary tumors. Interestingly, CEACAM1-3L was predominatelyand CEACAM1-3S was solely expressed in progressed

A B

C D

Control

CEACAM1-3L

CEACAM1-3S

CEACAM1-4LCEACAM1-4S

0 10 20 30 40 50 60 700

1

2

3

4

5

6

7

Cel

l ind

ex

Time (h)

Time (h)

Cel

l ind

ex

0 10 20 30 40 50 60 700

1

2

3

4

5

6

7

Inve

rted

CI 5

0(1

/hou

r)

0.00

0.01

0.02

0.03

0.04

*

****

0.00

0.04

0.08

0.12

Inve

rted

CI 5

0(1

/hou

r)

**

*

*

*

Figure 4.CEACAM1 splice variants differentiallyaffects cell motility. A–D, functionalanalysis of CEACAM1 transfectantsusing the Real-Time Cell AnalyzerSystem (xCELLigence). Verification ofmigration (A and B) and invasiveness(C and D) of each CEACAM1 variant.xCELLigence analyzes were repeatedthree times, each variant performed intriplicates per assay withcorresponding results. Arepresentative experiment is shown.B and D, statistical analyses wereperformed by comparing the halfmaximum CI values (CI50) in the rangefrom 0 to 70 hour of each transfectantwith control cells by Student t test.The inverted mean values of threeindependent experiments are shown.� , P < 0.05; �� , P < 0.01; error bars,mean � SEM.

Time (h)

Nor

mal

ized

cel

l ind

ex (

t = 4

h)

0 20 40 600

10

20

30

40

50

CEACAM1-4S + DMSO

CEACAM1-4S + marimastat

CEACAM1-4L + DMSO

CEACAM1-4L + marimastat

Control + DMSO

Figure 5.Enhanced invasiveness of CEACAM1-4S and CEACAM1-4L is mediated byMMPs. The xCELLigence system was used to analyze invasive potential ofCEACAM1-4L, CEACAM1-4S transfectants in the presence or absence of theMMP inhibitor marimastat compared with CEACAM1-negative control cells inreal time. Monitoring was conducted in triplicates and experiments wererepeated three times. One representative experiment out of three is shown.






MICA ULBP2

Eve

nts

A

CEACAM1-3L

CEACAM1-3S

CEACAM1-4L

CEACAM1-4S

B

D

sULBP2

E

MICA

70 kDa

55 kDa

70 kDa55 kDa

β-Actin40 kDa

F

HCEACAM1-3S + aNKG2DCEACAM1-3S + IgG

Control + aNKG2DControl + IgG

E:T

% o

f Spe

cific

lysi

s

0

20

40

60

sMICA

35 kDa

0.0

0.5

1.0

1.5

2.0

MF

I ULB

P2

*

0.0

0.5

1.0

1.5

2.0

MF

I MIC

A

*

**

Rel

ativ

e sM

ICA

0

1

2

3*

Specific lysis (normalized to control)

CEACAM1-3S + aNKG2D

CEACAM1-3S + IgG

Control + aNKG2D

Control + IgG*

*

**

I

CULBP2MICA

CE

AC

AM

1-3S

Con

trol

Rel

ativ

e sM

ICA

0

1

2

3

4 DMSOMarimastat

**

G

ULBP235 kDa

Figure 6.CEACAM1-3S enhances the immunogenicity ofmelanoma cells by upregulating cell surface expression of NKG2D ligands. A, MICA andULBP2 cell surface expressionof control cells compared with CEACAM1 transfectants by flow cytometry. Representative histograms are shown: background fluorescence of control cells(thin black line) or of indicated CEACAM1 transfectants (thin gray line), expression for NKG2D ligands of control cells (thick black line), or of CEACAM1isoform transfectants (dashed black line). (Continued on the following page.)

Ullrich et al.





melanoma (stage III/IV). Previously, Gambichler and collea-gues reported elevated CEACAM1 expression in melanomacompared with benign nevi (39). The data we present suggestthat the enhanced CEACAM1 expression levels observed byGambichler and colleagues in primary melanomas were due tochanges in the expression of CEACAM1-4S/L. Thus, theyseemed to have underestimated the impact of individual bio-logic functions of the four CEACAM1 isoforms on melanomaprogression and metastasis whose importance has been impli-cated by our study. The lack of CEACAM1-3S expression in earlymelanoma (stage I/II) followed by the dramatic increase in denovo expression for both CEACAM1-3 variants (86%) in laterstages (stage III/IV) identified CEACAM1-3 and, in particular,CEACAM1-3S as potential novel biomarker for disease progres-sion. This finding, however, requires a prospectively plannedlongitudinal evaluation.

Cellular functions of CEACAM1 require its localization at thecell surface and CEACAM1 has been shown to be recruited to sitesof cell–cell contacts (46). Interestingly, by selective expression ofindividual CEACAM1 isoforms in melanoma cells, we identifiedvariant-specific cellular localizations that are determined by theextracellular domain. However, CEACAM1-4 variants were pri-marily membrane associated, whereas CEACAM1-3 isoformswere predominantly localized in to vesicular-like structures thataccumulated around the nucleus. Schumann and colleagues (47)has previously reported thatCEACAM1-Sbinds to F-actin. Togeth-er with our data, this leads to the speculation that CEACAM1-3L-andCEACAM1-3S–positive vesicular structuresmove toward sitesof cell–cell contact via association with actin fibers. No evidencecurrently exists in the literature to identify which specific motorproteins would mediate this process. Sadekova and colleaguesdescribed localization of CEACAM1 in lamellipodia, suggesting apotential role in cell motility (48). However, it is still not under-stood if and how the extracellular domains of CEACAM1 affectdownstream effectors, which modulate subsequently the locali-zation of CEACAM1.

Early studies reported enhanced cell migration and invasionafter transfection of CEACAM1-L into human melanoma cells,whereas overexpression of the CEACAM1-S variant had no effect(13). Contrary to this report, our data precisely show that CEA-CAM1-4L and CEACAM1-4S trigger the migratory capacity andinvasive potential of melanoma cells, whereas CEACAM1-3Sdiminishes these cellular properties. In this context, Ebrahimne-jad and colleagues has proposed that Tyr-488 within the ITIMdomain of the CEACAM1 long cytoplasmic domain is essentialfor the invasive and migratory effect in CEACAM1-transfectedmelanoma cells (13). The fact that our study show that theexpression of CEACAM1-3L (also containing Tyr-488) influencedmigratory and invasive potential less strongly than CEACAM1-4Land CEACAM1-4S, argues against the hypothesis. Moreover, thehere shown contradictory effects of CEACAM1-4S andCEACAM1-

3S on cell motility, implicate that the modification of functionoccurs independent of the long cytoplasmic domain. To ourknowledge, this is the first report showing that the extracellulardomain of CEACAM1 has the potential to modulate cellularfunctions. The migratory behavior and invasive potential ofcancer cells are, of course, multifaceted and strongly influencedby the tumor microenvironment. In this context, our in vitroexperiments implicated the involvement of MMPs in regulatingCEACAM1-4L- and CEACAM1-4S–mediated invasion.

Recently, it was shown that CEACAM1-3L and CEACAM1-3Sdampen antitumor immunity by downregulating the surfaceexpression of ligands for the activating NK cell receptor NKG2Din colon cancer cells (22).Contrastingly, our data clearly point to adifferential isoform-specific function of CEACAM1 in melanomacells. In our experiments, CEACAM1-3S expression triggeredenhanced expression of the NKG2D ligands MICA, ULBP2, andDNAM-1 ligand CD155 on the melanoma cell surface, whichdrove these cells to NK cell-mediated cytolysis. In contrast,expression of CEACAM1-4L resulted in reduced cell surfaceexpression of MICA, ULBP2, CD155, and CD112, whereas thelower surface expression of NKG2DLs wasmediated by enhancedshedding of these ligands.

Our differing data on the impact of CEACAM1-4L for NK cell-mediated tumor cell cytolysis could be caused by donor-specificvariations. In line with this idea, Markel and colleagues presenteddata that melanoma patient-derived NK cells show an irregularphenotype and lower levels of NKp46, NKp30, and CD16 whilethe expression NKG2D was not altered (31). Furthermore, theyshowed that NK cells from healthy donors behaved differentlythan patient-derived NK cells. Thus, utilizing melanoma patient-derived NK cells expressing increased amounts of CEACAM1 ontheir cell surface, could result in stronger interaction with CEA-CAM1-4L on melanoma cells (Fig. 3A) and consequently showhigher susceptibility to CEACAM1-mediated inhibition ofNKG2D-triggered lysis.

Interestingly, treatment with marimastat reverted the solublelevel of MICA in medium conditioned by CEACAM1-4L–expres-singmelanoma cells almost back to the level of control cells. Thesedata implicate a possible involvement of MMPs in the regulationof NKG2D ligand shedding by CEACAM1-4L in melanoma cells.Data from Liu and colleagues strengthen our hypothesis. Theycould show that MMP mediates MICA shedding independent ofdisintegrin; otherMMPs and ADAMs (49). Our data implicate thepossibility that CEACAM1-4L–expressing melanoma cells reducethe tumor–immune response by MMP-mediated shedding ofNKG2DL but further studies are clearly needed to unravel theinvolvement of all the players in this process. Taken together, ourresults provide for the first time evidence that the expression ofCEACAM1-3S, CEACAM1-3L, CEACAM1-4S, and CEACAM1-4Ldifferentially affect disease progression in malignant melanoma.Importantly, we clearly demonstrated that CEACAM1 variants

(Continued.) B, normalized mean mean fluorescence intensity (MFI) values for MICA and ULBP2 in the indicated transfectants (N¼ 5). C, IHC for NKG2DL (green) incontrol cells and CEACAM1-3S transfectants. Nuclei were counterstained using Dapi (blue). White arrowheads, cell surface expression of MICA/ULBP2.Scale bar, 10 mm. D, immunoblot analysis for MICA and ULBP2 expression in CEACAM1 transfectants. One representative blot is shown (N¼ 4). E and F, soluble MICAand ULBP2 expression in conditioned media of indicated CEACAM1 transfectants analyzed by Western blot analysis (E) and MICA-specific sandwich ELISA(N ¼ 6; F). Sample loading was normalized to cell number. One representative blot is shown. G, MICA-specific ELISA of supernatants collected from CEACAM1-3S,CEACAM1-4L transfectants, and controls after treatment with marimastat or DMSO. Sample loading was normalized to cell number. Mean values of threeindependent experiments are presented. H, NK cell-mediated cytotoxicity at various E/T ratios was measured by CSFE/7AAD cytotoxicity assay. To blockNKG2D-dependent recognition of melanoma cells, NK cells were incubated with anti-NKG2DmAb. Controls were incubated with IgG antibody. One representativeexperiment is shown. I, cytolysis measured at an effector to target ratio of 10:1 (N¼ 4; � , P < 0.05; �� , P < 0.01). All statistical tests were run on un-normalized data.Mean � SEMs are shown.






differentially modulate attenuation of immune surveillance byregulating ligand expression for the NKG2D receptor on tumorcells inmalignantmelanoma. These findings are ofmajor interest,since CEACAM1 is discussed for use in clinical applications ofimmunotherapy. Caused by the fact that CEACAM1-3S poten-tially inhibits melanoma cell migration and invasion, and drivesthese cells through NK cell-mediated cytolysis, whereas CEA-CAM1-4L is acting in the opposite direction by supporting tumorprogression, it is indispensable to define the impact of differentCEACAM1 variants to secure therapeutic efficiency. Hence, deter-mination of all four CEACAM1 isoforms opens up new possibil-ities for diagnosis and prognosis of melanoma patients.

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

Authors' ContributionsConception and design: N. Ullrich, A. Heinemann, I. Scheffrahn, D. Schaden-dorf, B.B. Singer, I. HelfrichDevelopment of methodology: N. Ullrich, A. Heinemann, E. Nilewski,I. Scheffrahn, B.B. Singer, I. HelfrichAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): N. Ullrich, A. Heinemann, I. Scheffrahn, J. Klode,B.B. Singer, I. HelfrichAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): N. Ullrich, A. Heinemann, I. Scheffrahn, A. Scherag,B.B. Singer, I. Helfrich

Writing, review, and/or revision of themanuscript:N. Ullrich, A. Heinemann,I. Scheffrahn, J. Klode, A. Scherag, D. Schadendorf, B.B. Singer, I. HelfrichAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): N. Ullrich, I. Scheffrahn, J. Klode, B.B. Singer,I. HelfrichStudy supervision: N. Ullrich, B.B. Singer, I. Helfrich

AcknowledgmentsThe authors thank Christiane Breuer, Birgit Maranca-H€uwel, B€arbel

Gobs-Hevelke, Mohamed Benchellal, and Nadine Hochhard for excellenttechnical assistance; Antje Sucker for providing Skin Cancer Biobank informa-tion ofmelanoma samples and establishedmelanoma cell lines; Dr. Peter Horn(Institute for TransfusionMedicine) for providing PBMCs; andDr. Ingo Stoffelsfor providing freshmelanoma biopsies.Microscopic imaging and analyses weredone in the Imaging Center Essen (IMCES) in consultation with Dr. AnthonySquire and Prof. Matthias Gunzer and professionalmanuscript editingmade byDr. Kathy Astrahantseff.

Grant SupportThis work was supported by Hiege Stiftung gegen Hautkrebs (I. Helfrich) and

by the Federal Ministry of Education and Research (BMBF), Germany, FKZ:01EO1002 (A. Scherag).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received June11, 2014; revised January30, 2015; accepted February 17, 2015;published OnlineFirst March 5, 2015.

References1. Schadendorf D, Hauschild A. Melanoma in 2013: Melanoma-the run of

success continues. Nat Rev Clin Oncol 2014;11:75–6.2. Edward M. Integrins and other adhesion molecules involved in melano-

cytic tumor progression. Curr Opin Oncol 1995;7:185–91.3. Sapoznik S, Ortenberg R, Schachter J, Markel G. CEACAM1 in malignant

melanoma: a diagnostic and therapeutic target. Curr Top Med Chem2012;12:3–10.

4. Thompson JA, Grunert F, Zimmermann W. Carcinoembryonic antigengene family: molecular biology and clinical perspectives. J Clin Lab Anal1991;5:344–66.

5. Huang J, Simpson JF, Glackin C, Riethorf L, Wagener C, Shively JE.Expression of biliary glycoprotein (CD66a) in normal and malignantbreast epithelial cells. Anticancer Res 1998;18:3203–12.

6. Huang J, Hardy JD, Sun Y, Shively JE. Essential role of biliary glycoprotein(CD66a) in morphogenesis of the human mammary epithelial cell lineMCF10F. J Cell Sci 1999;112 (Pt 23):4193–205.

7. Muller MM, Singer BB, Klaile E, Obrink B, Lucka L. TransmembraneCEACAM1 affects integrin-dependent signaling and regulates extracellularmatrix protein-specific morphology and migration of endothelial cells.Blood 2005;105:3925–34.

8. Singer BB, Scheffrahn I, Heymann R, Sigmundsson K, Kammerer R, ObrinkB. Carcinoembryonic antigen-related cell adhesion molecule 1 expressionand signaling in human, mouse, and rat leukocytes: evidence for replace-ment of the short cytoplasmic domain isoform by glycosylphosphatidy-linositol-linked proteins in human leukocytes. J Immunol 2002;168:5139–46.

9. Singer BB, Klaile E, Scheffrahn I, Muller MM, Kammerer R, Reutter W, et al.CEACAM1 (CD66a) mediates delay of spontaneous and Fas ligand-induced apoptosis in granulocytes. Eur J Immunol 2005;35:1949–59.

10. Neumaier M, Paululat S, Chan A, Matthaes P, Wagener C. Biliary glyco-protein, a potential human cell adhesion molecule, is down-regulated incolorectal carcinomas. Proc Natl Acad Sci U S A 1993;90:10744–8.

11. Luo W, Tapolsky M, Earley K, Wood CG, Wilson DR, Logothetis CJ, et al.Tumor-suppressive activity of CD66a in prostate cancer. Cancer Gene Ther1999;6:313–21.

12. Riethdorf L, Lisboa BW, Henkel U, Naumann M, Wagener C, Loning T.Differential expression of CD66a (BGP), a cell adhesion molecule of the

carcinoembryonic antigen family, in benign, premalignant, andmalignantlesions of the human mammary gland. J Histochem Cytochem 1997;45:957–63.

13. Ebrahimnejad A, Streichert T, Nollau P, Horst AK, Wagener C, BambergerAM, et al. CEACAM1 enhances invasion andmigration of melanocytic andmelanoma cells. Am J Pathol 2004;165:1781–7.

14. Thies A, Berlin A, Brunner G, Schulze HJ, Moll I, Pfuller U, et al. Glyco-conjugate profiling of primarymelanoma and its sentinel node and distantmetastases: implications for diagnosis and pathophysiology of metastases.Cancer Lett 2007;248:68–80.

15. Thom I, Schult-Kronefeld O, Burkholder I, Schuch G, Andritzky B,Kastendieck H, et al. Expression of CEACAM-1 in pulmonaryadenocarcinomas and their metastases. Anticancer Res 2009;29:249–54.

16. Singer BB, Scheffrahn I, Obrink B. The tumor growth-inhibiting celladhesion molecule CEACAM1 (C-CAM) is differently expressed in prolif-erating and quiescent epithelial cells and regulates cell proliferation.Cancer Res 2000;60:1236–44.

17. Singer BB, Scheffrahn I, Kammerer R, Suttorp N, Ergun S, Slevogt H.Deregulation of the CEACAM expression pattern causes undifferentiatedcell growth in human lung adenocarcinoma cells. PLoS ONE 2010;5:e8747.

18. Stern N, Markel G, Arnon TI, Gruda R, Wong H, Gray-Owen SD, et al.Carcinoembryonic antigen (CEA) inhibits NK killing via interactionwith CEA-related cell adhesion molecule 1. J Immunol 2005;174:6692–701.

19. Markel G, Lieberman N, Katz G, Arnon TI, LotemM, Drize O, et al. CD66ainteractions between humanmelanoma andNK cells: a novel class IMHC-independent inhibitory mechanism of cytotoxicity. J Immunol 2002;168:2803–10.

20. MarkelG,WolfD,Hanna J, Gazit R,Goldman-WohlD, Lavy Y, et al. Pivotalrole of CEACAM1 protein in the inhibition of activated decidual lympho-cyte functions. J Clin Invest 2002;110:943–53.

21. Markel G, Seidman R, Stern N, Cohen-Sinai T, Izhaki O, Katz G, et al.Inhibition of human tumor-infiltrating lymphocyte effector functions bythe homophilic carcinoembryonic cell adhesion molecule 1 interactions.J Immunol 2006;177:6062–71.


Ullrich et al.




22. Chen Z, Chen L, Baker K, Olszak T, Zeissig S, Huang YH, et al. CEACAM1dampens antitumor immunity by down-regulating NKG2D ligand expres-sion on tumor cells. J Exp Med 2011;208:2633–40.

23. Champsaur M, Lanier LL. Effect of NKG2D ligand expression on hostimmune responses. Immunol Rev 2010;235:267–85.

24. Nausch N, Cerwenka A. NKG2D ligands in tumor immunity. Oncogene2008;27:5944–58.

25. Barnett TR, Drake L, Pickle W. Human biliary glycoprotein gene: charac-terization of a family of novel alternatively spliced RNAs and theirexpressed proteins. Mol Cell Biol 1993;13:1273–82.

26. Beauchemin N, Draber P, Dveksler G, Gold P, Gray-Owen S, Grunert F,et al. Redefined nomenclature for members of the carcinoembryonicantigen family. Exp Cell Res 1999;252:243–9.

27. Gray-Owen SD, Blumberg RS. CEACAM1: contact-dependent control ofimmunity. Nat Rev Immunol 2006;6:433–46.

28. Fiori V, Magnani M, Cianfriglia M. The expression and modulation ofCEACAM1 and tumor cell transformation. Ann Ist Super Sanita2012;48:161–71.

29. Obrink B. On the role of CEACAM1 in cancer. Lung Cancer 2008;60:309–12.

30. Thies A, Moll I, Berger J, Wagener C, Brummer J, Schulze HJ, et al.CEACAM1 expression in cutaneous malignant melanoma predicts thedevelopment of metastatic disease. J Clin Oncol 2002;20:2530–6.

31. Markel G, Ortenberg R, Seidman R, Sapoznik S, Koren-Morag N, Besser MJ,et al. Systemic dysregulation of CEACAM1 in melanoma patients. CancerImmunol Immunother 2010;59:215–30.

32. Sivan S, Suzan F, Rona O, Tamar H, Vivian B, Tamar P, et al. SerumCEACAM1 correlates with disease progression and survival in malignantmelanoma patients. Clin Dev Immunol 2012;2012:290536.

33. Balch CM, Gershenwald JE, Soong SJ, Thompson JF, Atkins MB, Byrd DR,et al. Final version of 2009 AJCC melanoma staging and classification.J Clin Oncol 2009;27:6199–206.

34. Ugurel S, Thirumaran RK, Bloethner S, Gast A, Sucker A, Mueller-Berghaus J, et al. B-RAF and N-RAS mutations are preserved duringshort time in vitro propagation and differentially impact prognosis.PLoS ONE 2007;2:e236.

35. Gast A, Scherer D, Chen B, Bloethner S, Melchert S, Sucker A, et al. Somaticalterations in the melanoma genome: a high-resolution array-based com-parative genomic hybridization study. Genes Chromosomes Cancer2010;49:733–45.

36. Hong J, Kandasamy K, Marimuthu M, Choi CS, Kim S. Electrical cell-substrate impedance sensing as a non-invasive tool for cancer cell study.Analyst 2011;136:237–45.

37. Fregni G, Perier A, Pittari G, Jacobelli S, Sastre X, Gervois N, et al. Uniquefunctional status of natural killer cells in metastatic stage IV melanoma

patients and its modulation by chemotherapy. Clin Cancer Res2011;17:2628–37.

38. Heinemann A, Zhao F, Pechlivanis S, Eberle J, Steinle A, Diederichs S, et al.Tumor suppressivemicroRNAsmiR-34a/c control cancer cell expression ofULBP2, a stress-induced ligand of the natural killer cell receptor NKG2D.Cancer Res 2012;72:460–71.

39. Gambichler T, Grothe S, Rotterdam S, Altmeyer P, Kreuter A. Proteinexpression of carcinoembryonic antigen cell adhesionmolecules in benignand malignant melanocytic skin lesions. Am J Clin Pathol 2009;131:782–7.

40. Airola K, Karonen T, Vaalamo M, Lehti K, Lohi J, Kariniemi AL, et al.Expression of collagenases-1 and -3 and their inhibitors TIMP-1 and -3correlates with the level of invasion in malignant melanomas. Br J Cancer1999;80:733–43.

41. ShumanMoss LA, Jensen-Taubman S, Stetler-StevensonWG.Matrixmetal-loproteinases: changing roles in tumor progression and metastasis. Am JPathol 2012;181:1895–9.

42. Bourd-Boittin K, Fridman R, Fanchon S, Septier D, GoldbergM,Menashi S.Matrixmetalloproteinase inhibition impairs the processing, formationandmineralization of dental tissues during mouse molar development.Exp Cell Res 2005;304:493–505.

43. Maekawa K, SatoH, FurukawaM, Yoshizaki T. Inhibition of cervical lymphnodemetastasis bymarimastat (BB-2516) in an orthotopic oral squamouscell carcinoma implantation model. Clin Exp Metastasis 2002;19:513–8.

44. Xin ZF, Kim YK, Jung ST. Risedronate inhibits human osteosarcoma cellinvasion. J Exp Clin Cancer Res 2009;28:105.

45. Ortenberg R, Sapir Y, Raz L, Hershkovitz L, Ben AA, Sapoznik S, et al. Novelimmunotherapy for malignant melanoma with a monoclonal antibodythat blocks CEACAM1 homophilic interactions. Mol Cancer Ther2012;11:1300–10.

46. Fournes B, Farrah J, Olson M, Lamarche-Vane N, Beauchemin N. DistinctRho GTPase activities regulate epithelial cell localization of the adhesionmolecule CEACAM1: involvement of the CEACAM1 transmembranedomain. Mol Cell Biol 2003;23:7291–304.

47. Schumann D, Chen CJ, Kaplan B, Shively JE. Carcinoembryonic antigencell adhesion molecule 1 directly associates with cytoskeleton proteinsactin and tropomyosin. J Biol Chem 2001;276:47421–33.

48. Sadekova S, Lamarche-Vane N, Li X, Beauchemin N. The CEACAM1-Lglycoprotein associates with the actin cytoskeleton and localizes to cell-cellcontact through activation of Rho-like GTPases. Mol Biol Cell 2000;11:65–77.

49. Liu G, Atteridge CL, Wang X, Lundgren AD, Wu JD. The membrane typematrix metalloproteinaseMMP14mediates constitutive shedding ofMHCclass I chain-related molecule A independent of A disintegrin and metal-loproteinases. J Immunol 2010;184:3346–50.






2015;75:1897-1907. Published OnlineFirst March 5, 2015.Cancer Res Nico Ullrich, Anja Heinemann, Elena Nilewski, et al. Cytolysis and Enhances Patient SurvivalCEACAM1-3S Drives Melanoma Cells into NK Cell-Mediated

Updated version

10.1158/0008-5472.CAN-14-1752doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2015/03/07/0008-5472.CAN-14-1752.DC1

http://cancerres.aacrjournals.org/content/suppl/2015/04/17/0008-5472.CAN-14-1752.DC2Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/75/9/1897.full#ref-list-1

This article cites 49 articles, 20 of which you can access for free 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://cancerres.aacrjournals.org/content/75/9/1897To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-14-1752

http://cancerres.aacrjournals.org/content/suppl/2015/04/17/0008-5472.CAN-14-1752.DC2 http://cancerres.aacrjournals.org/content/suppl/2015/03/07/0008-5472.CAN-14-1752.DC1

http://cancerres.aacrjournals.org/content/suppl/2015/04/17/0008-5472.CAN-14-1752.DC2 http://cancerres.aacrjournals.org/content/suppl/2015/03/07/0008-5472.CAN-14-1752.DC1

http://cancerres.aacrjournals.org/content/75/9/1897.full#ref-list-1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/75/9/1897


ceacam1-3s drives melanoma cells into nk cell-mediated ...tumor and stem cell biology ceacam1-3s...

Documents