S100A8/A9 Activate Key Genes and Pathways in Colon Tumor ... · kB–p65 binding activity using TransAM NF-kB assay kit (ActiveMotif) according to the manufacturer's instructions

Angiogenesis, Metastasis, and the Cellular Microenvironment

S100A8/A9 Activate Key Genes and Pathways in ColonTumor Progression

Mie Ichikawa1, Roy Williams2, Ling Wang3, Thomas Vogl4, and Geetha Srikrishna1

AbstractThe tumormicroenvironment plays an important role inmodulating tumor progression. Earlier, we showed that

S100A8/A9 proteins secreted by myeloid-derived suppressor cells (MDSC) present within tumors and metastaticsites promote an autocrine pathway for accumulation of MDSC. In a mouse model of colitis-associated coloncancer, we also showed that S100A8/A9-positive cells accumulate in all regions of dysplasia and adenoma. Here wepresent evidence that S100A8/A9 interact with RAGE and carboxylated glycans on colon tumor cells and promoteactivation of MAPK and NF-kB signaling pathways. Comparison of gene expression profiles of S100A8/A9-activated colon tumor cells versus unactivated cells led us to identify a small cohort of genes upregulated in activatedcells, including Cxcl1, Ccl5 and Ccl7, Slc39a10, Lcn2, Zc3h12a, Enpp2, and other genes, whose products promoteleukocyte recruitment, angiogenesis, tumor migration, wound healing, and formation of premetastatic niches indistalmetastatic organs. Consistentwith this observation, inmurine colon tumormodels we found that chemokineswere upregulated in tumors, and elevated in sera of tumor-bearing wild-type mice. Mice lacking S100A9 showedsignificantly reduced tumor incidence, growth andmetastasis, reduced chemokine levels, and reduced infiltration ofCD11bþGr1þ cells within tumors and premetastatic organs. Studies using bone marrow chimeric mice revealedthat S100A8/A9 expression on myeloid cells is essential for development of colon tumors. Our results thus reveal anovel role for myeloid-derived S100A8/A9 in activating specific downstream genes associated with tumorigenesisand in promoting tumor growth and metastasis. Mol Cancer Res; 9(2); 133–48. �2011 AACR.

Introduction

S100A8 and S100A9 belong to a family of more than 20low-molecular-weight intracellular EF-hand motif calcium-binding proteins found exclusively in vertebrates (1–3).They are expressed predominantly by myeloid cells, includ-ing granulocytes, monocytes, myeloid-derived suppressorcells (MDSC), and other immature cells of myeloid lineage(4–7). Although the proteins are products of distinct genes,they are often coexpressed and function mainly as hetero-dimer of S100A8/A9 (calprotectin). Expression is down-regulated during macrophage and dendritic celldifferentiation (6, 8, 9), but can be induced in epithelialcells, osteoclasts, and keratinocytes (10). When these intra-cellular proteins are released into the extracellular medium in

response to cell damage or activation they become dangersignals (damage-associated molecular pattern molecules orDAMP), which alert the host of danger by triggeringimmune responses and activating repair mechanismsthrough interaction with pattern recognition receptors(11–15). Elevated S100A8/A9 is the hallmark of inflamma-tory conditions such as rheumatoid arthritis, inflammatorybowel disease, multiple sclerosis, cystic fibrosis and psoriasis(4, 11, 16). Critical roles for these proteins in endotoxin-induced lethality and systemic autoimmunity have recentlybeen recognized (17, 18). In addition to expression withininflammatory milieu, strong upregulation of these proteinshas also observed in many tumors, including gastric, colon,pancreatic, bladder, ovarian, thyroid, breast, and skin cancers(10, 19), and it is becoming increasingly clear that S100A8/A9 not only serve as markers of immune cells within thetumor microenvironment, but that they may also haveindependent pathogenic roles in cancer progression.S100A8/A9 exhibit concentration-dependent dichotomy

of function in tumors. At high concentrations (80–100 mg/mL), S100A8/A9 exert apoptotic effects on tumor cells(20), whereas at low concentrations (<25 mg/mL) promotetumor cell growth (21, 22). S100A8/A9 also stimulatetumor cell migration at low concentrations (23–27). Morerecently, other pathophysiological roles for these proteins intumors have also been identified. Studies from our labora-tories and others show that S100A8/A9 regulate the accu-mulation of MDSC (6, 7). MDSC are immature myeloid

Authors' Affiliations: 1Sanford Children's Health Research Center, 2Bioin-formatics Shared Resource, and 3Transgenic Mouse Facility, Sanford-Burnham Medical Research Institute, La Jolla, California; and 4Institute ofImmunology, University of M€unster, M€unster, Germany.

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

Corresponding Author: Geetha Srikrishna, Sanford-Burnham MedicalResearch Institute, 10905 Road to the Cure, San Diego, CA 92121. Phone:858-795-5256; Fax: 858-713-6281. E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-10-0394

�2011 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 133

on November 16, 2020. © 2011 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst January 12, 2011; DOI: 10.1158/1541-7786.MCR-10-0394





http://mcr.aacrjournals.org/



cells that expand during inflammation and in tumors andare potent suppressors of T-cell–mediated immuneresponses (28–30). Cheng and colleagues showed thattumor-derived factors promote sustained STAT3-depen-dent upregulation of S100A9 in myeloid precursors, whichresults in inhibition of differentiation to DC and accumula-tion of MDSC (6). We showed that S100A8/A9 are notonly synthesized and secreted by MDSC, but they also havebinding sites for S100A8/A9, and activate intracellularsignaling that promote their migration (7). These findingsstrongly suggest that the S100A8/A9 proteins support anautocrine feedback loop that sustains accumulation ofMDSC in tumors (31). S100A8/A9 are also involvedin early metastatic processes. Expression of S100A8/A9in myeloid and endothelial cells in premetastatic organsin response to soluble factors such as VEGF, TGFb, andTNFa expressed by distal primary tumors promotes hom-ing of tumor cells to premetastatic niches (24).Recent studies argue for prominent roles for 2 pattern

recognition receptors, TLR4 and RAGE, in S100A8/A9-mediated pathologic effects. The interaction with TLR4promotes endotoxin-induced lethality and the develop-ment of systemic autoimmunity (17, 18). S100A8/A9-mediated signaling through TLR4 also promotes preme-tastatic niches in lungs (32). Interaction of S100A8/A9with RAGE has been shown to promote tumor growth,and MDSC migration (7, 21, 33). However, whichreceptor and signaling pathways are preferentially acti-vated is not understood and may depend on the pathologicsettings, cell types involved, ligand concentrations, andother factors. Distinct epitopes on RAGE and TLRsrecognized by the ligands may also impart specificity.We found that S100A8/A9 bind to a subpopulation ofRAGE expressing carboxylated glycans (22). These gly-cans show restricted expression on myeloid, endothelial,and tumor cells (34). Inhibiting carboxylated glycan–dependent interactions using mAbGB3.1, an anti-glycanantibody, blocked T-cell–mediated colitis (35), colitis-associated colon cancer (22), and accumulation of MDSCin a 4T1 model of metastatic mammary tumor (7).RAGE-deficient mice show reduced tumors in the colitisassociated cancer (CAC) model (22) and inflammation-induced skin cancer model (33), suggesting that RAGEand carboxylated glycans form important components oftumor and stromal cells promoting molecular commu-nications leading to myeloid accumulation and tumorgrowth.Our previous observation of the presence of S100A8- and

S100A9-positive myeloid cells in the microenvironment ofcolitis-induced colon tumors (22) prompted us to examinepossible interactions of these proteins with tumor cells. Weinvestigated S100A8/A9 binding to colon tumor cells andsubsequent activation of signaling pathways and geneexpression in vitro. The results led us to further investigatethe contribution of these proteins in vivo to colon tumorgrowth, establishment of premetastatic niches in distalorgans, and promotion of metastasis in tumor-bearingmice. Our findings uncovered several protumorigenic genes

activated in tumor cells by S100A8/A9, strongly supportinga novel role of S100A8/A9 and myeloid cells in tumorprogression.

Materials and Methods

Mouse and human S100A8 and S100A9 heterodimersand homodimers were purified as described (36) andrendered endotoxin-free. MC38 cells andMC38 cells stablyexpressing green fluorescent protein (GFP) were kindgifts from Drs. Ajit and Nissi Varki, University of Califor-nia, San Diego. Caco-2 cells were obtained from AmericanType Culture Collection (ATCC). Cells were maintained inDulbecco's modified Eagle's medium containing 100U/mLpenicillin, 100 mg/mL streptomycin, glutamine, 10% FBS,0.1 mmol/L nonessential amino acids, and 1 mmol/Lsodium pyruvate (and 1 mg/mL G418 for MC38 cellsexpressing GFP).

Flow cytometric analysisTo detect surface expression of RAGE or carboxylated

glycans, tumor cells were incubated with rabbit polyclonalanti-RAGE (raised against a peptide corresponding toamino acids 39–58 of human RAGE and which recognizeshuman, bovine, and mouse RAGE) or anti-carboxylatedglycan antibody mAbGB3.1 (34) in HBSS containing 1%BSA, followed by Phycoerythrin-conjugated secondary anti-bodies and analyzed by flow cytometry with a FACScan(Becton Dickinson), equipped with CellQuest software,and gated by the side scatter and forward scatter filters.

ImmunoprecipitationMC38 or Caco-2 cells were isolated using PBS, contain-

ing 5 mmol/L EDTA, washed with HBSS, and incubated inHBSS medium containing purified mouse S100A8,S100A9, or S100A8/A9 for MC38 cells (at 1 mg/millioncells in 100 mL final volume) or corresponding purifiedhuman proteins for Caco-2 cells for 1 hour at 4�C. Cellswere washed with cold PBS twice, lysed in 20 mmol/L Tris-HCl, pH 7.4, with 150 mmol/L sodium chloride, 0.5%NP-40, and protease inhibitors, and centrifuged at 10,000� g for 15 minutes to remove cell debris. Lysates wereprecleared with Protein G Sepharose beads for 1 hour at4�C, and S100 proteins was immunoprecipitated by usingrabbit polyclonal antibodies against the respective proteinsor an irrelevant control antibody overnight at 4�C andProtein G Sepharose beads. The beads were washed ofunbound proteins, and immunoprecipitated proteins wereanalyzed for RAGE or TLR4 by electrophoresis andWestern blots as described in the following text.

siRNA treatmentA target-specific 20 to 25 nucleotdes siRNA duplex

designed to knock down expression of mouse S100A9mRNA (calgranulin B siRNA; Santa Cruz BiotechnologyInc.) was used to transfect MC38 cells with transfectionreagents and protocol provided by the manufacturer. Genesilencing was confirmed by Western blots of whole-cell

Ichikawa et al.

Mol Cancer Res; 9(2) February 2011 Molecular Cancer Research134




lysates with anti-S100A9. Cells were used for immunopre-cipitation 48 hours after transfection.

Signaling assaysMC38 or Caco-2 cells were treated with mouse or human

S100A8, S100A9, or S100A8/A9 (10 mg/mL) for 0, 15, 30,and 60 minutes, after overnight (16 hours) starvation inmedium containing 0.1% serum. After indicated periods ofincubation, cells were washed with cold PBS, harvested andlysed at 4�C, and lysates were analyzed by Western blot,using respective MAPK or IkB antibodies as described inthe following text. For studies using mAbGB3.1 or anti-RAGE, following starvation, cells were preincubated for2 hours with 20 mg/mL of mAbGB3.1 or rabbit polyclonalaRAGE prior to activation.

Electrophoresis and Western blotsTumor cell lysates, immunoprecipitated proteins from

tumor cells, or cell lysates from signaling assays were elec-trophoresed on denaturing and reducing 10% polyacryla-mide gels and transferred to nitrocellulose membranes. Theblots were blocked with 10% dry skimmed milk. To detectRAGE or TLR4 in lysates or immunoprecipitates, blots wereincubated with goat polyclonal anti-mouse S100A9 or anti-mouse RAGE (R&D Systems), rat monoclonal anti-humanRAGE (kind gift from Novartis foundation), or rabbitpolyclonal anti-TLR4 antibody (Imgenex Corporation) fol-lowed by respective peroxidase-conjugated secondary anti-body. Phosphorylation of ERK1/ERK2, p38, SAPK/JNK,and IkB in activated cell lysate proteins was detected usingrespective rabbit polyclonal or mouse monoclonal phospho-specific antibodies (Cell Signaling Technology) followed byperoxidase-conjugated secondary antibodies. As loadingcontrols, separate lanes with lysate proteins were incubatedwith rabbit polyclonal antibodies for total ERK1/ERK2,p38, SAPK/JNK, IkB, or b-actin (Cell Signaling Technol-ogy) followed by peroxidase-conjugated secondary antibody.Bands were visualized using ECL detection system(GE Healthcare).

Measurement of NF-kB bindingNuclear extracts isolated fromMC38 cells or Caco-2 cells

treated with respective S100 proteins were assayed for NF-kB–p65 binding activity using TransAM NF-kB assay kit(ActiveMotif) according to the manufacturer's instructions.

Isolation of total RNA and gene expression profilingSubconfluent cultures of MC38 cells were serum-starved

for 16 hours and activated with 10 mg/mL S100A8/A9 for6 hours. Total RNA was extracted from unactivated oractivated cells, using an RNeasy kit (Qiagen), and biotiny-lated cRNA was prepared using the Illumina RNA Ampli-fication Kit (Applied Biosystems/Ambion). Hybridizationto the Sentrix Mouse-6 Expression BeadChip containingmore than 45,000 transcript-specific probe sequences/array(Illumina Incorporated), followed by washing and scanning,was performed according to manufacturer's instructions.The resulting images were analyzed using GenomeStudio

(Illumina Incorporated) and GeneSpringGX11 (AgilentTechnologies) image processing software. Experiments wereperformed in duplicates.

Real-time quantitative PCR analysis of chemokinegenesSYBR Green oligonucleotide primers for the real-time

quantitative PCR (qRT-PCR) analyses were designedusing Primer 3 software. Mouse glyceraldehyde-3-phos-phate dehydrogenase (GAPDH) was used as a control.Following RT using Roche Transcriptor First Strand cDNAsynthesis kit (Roche Applied Science) PCR (45 cycles) wasperformed using Roche Lightcycler 480 as follows: prein-cubation at 95�C for 5 minutes, denaturation at 95�C for10 seconds, annealing at 60�C for 10 seconds, and elonga-tion at 72�C for 10 seconds.

Quantitation of CXCL1CXCL1 in culture supernatants and mouse sera were

measured using a commercial ELISA kit (R&D Systems).

Mouse tumor modelsS100A9 null mice were generated as described (37). They

were backcrossed to C57BL/6 mice for more than 10generations. Six to 8-week-old S100A9 null mice, theirwild-type littermates, or age-matched wild-type mice wereused for experiments. All animal protocols were approved bythe Sanford-Burnham Medical Research Institute AnimalCare and Use Committee and were in compliance withNIH policies.CAC model. CAC was induced in separate groups of

wild-type or S100A9 null mice, using AOM and DSSessentially as described (38) except that mice were subjectedto 2 cycles of DSS 2 weeks apart. Animals were constantlymonitored for clinical signs of illness and were sacrificed atthe end of 2, 6, 12, or 20 weeks after DSS. Blood sampleswere collected by retro-orbital bleeding prior to induction ofdisease and at aforementioned time points. At each timepoint, colons were removed and fixed as "Swiss-rolls" in 4%buffered formalin. Stepwise sections were cut and stainedwith hematoxylin and eosin (H&E). Colonic inflammation,dysplasia, and neoplasms were graded on the basis ofdescribed criteria (38).CAC in bone marrow chimeric mice. Bone marrow

cells were aseptically isolated from the femur and tibia ofwild-type or S100A9 null mice and each injected intrave-nously into either wild-type or S100A9 null mice at6 million cells per mouse. Recipient mice were lethallyirradiated using a Gammacell 40 Exactor (9 Gy from a 137Cssource) before injection. Reconstitution of leukocyte popu-lations was comparable in these groups. We confirmedsuccessful engraftment by measurement of S100A8/A9 inserum. Mice were subjected to the AOM/DSS protocol4 weeks later, sacrificed 12 weeks after DSS, and blood andtissues were collected.MC38 ectopic tumor model. To generate primary

tumors, single-cell suspensions of 1 � 106 MC38 cells inlogarithmic phase of growth were injected subcutaneously

S100A8/A9 Activate Key Genes

www.aacrjournals.org Mol Cancer Res; 9(2) February 2011 135




into the flank of wild-type or S100A9 null mice and all-owed to grow for 10 to 20 days. To evaluate the role ofcarboxylated glycans, separate groups of wild-type micewere treated with 10 mg/gm of mAbGB3.1 weekly startingfrom 2 days prior to injection of tumor cells. Tumor growthwas measured using calipers over the experimental period,and tumor volume was estimated. Lungs, liver (primarymetastatic organs), and tumors were frozen for furtheranalysis. Bone marrow responses to tumor growth wereevaluated as follows: Bone marrow cells were isolated fromfemur and tibia of mice and RBCs lysed according tostandard protocols. Myeloid cells were stained with differ-entiation marker CD11b and costained with mAbGB3.1,anti-RAGE, or anti-Gr-1 (Ly6C and Ly6G; BD-Pharmin-gen). 7-Aminoactinomycin D (7-AAD; Invitrogen) or pro-pidium iodide (BD-Pharmingen) was included to identifydead cells and analyzed by flow cytometry. Peripheral bloodhematology profile was obtained on EDTA samples, using aVetScan HMII hematology system (Abaxis).Liver metastasis model. S100A9 null mice and age-

matched wild-type C57BL/6 mice were anesthetized usingi.p. injection of avertin. Under aseptic conditions, a smalllongitudinal incision was made in the left upper flank tovisualize the spleen, and 1 � 106 MC38 cells in 50 mL ofserum-free medium were injected under the spleen capsulewith a 27-gauge needle. The spleen was then inserted backinto the abdominal cavity and the peritoneum and abdom-inal walls were sutured with silk. Animals were sacrificed2 weeks later, and livers were isolated, fixed in bufferedformalin, and paraffin-embedded. Tumors were enumer-ated by visual inspection and by examining liver sectionsstained by H&E. Slides were scanned using an automatichigh-throughput ScanScope, viewed, and tumor areas weremeasured using Aperio software (Aperio).

Immunosuppression assaySpleen CD4þ T cells isolated from OTII transgenic mice

(kindly provided by the Rickert and Bradley labs, Sanford-Burnham Medical Research Institute) were cocultured in48-well plates at 37�C in RPMI-1640 medium (containing10% FBS, penicillin, streptomycin, and 2-ME) withCD11bþGr1þ cells isolated from the spleens of tumor-bearing mice by MACS, at increasing ratios of MDSC:Tcells in the presence of 10 mg/mL OVA peptide (OVA323-339). Cells were pulsed with 1 mCi [3H]thymidine/well onday 3, and 18 hours later, the cells were harvested andcounted. Proliferation in the absence of MDSC was con-sidered 100%.

Immunochemical analysisSwiss rolls of colons from the AOM/DSS model were

deparafinized and endogenous peroxidase blocked by incu-bating with 0.36% beta-glucose, 0.01% glucose oxidase,and 0.013% sodium azide in PBS for 60 minutes at 37�C.The sections were stained with 1:50 dilution of anti-mouseCXCL1, or CCL7 (Santa Cruz Biotechnology), followed byanti-goat peroxidase (1:100) and developed using DABsubstrate. To characterize macrophage populations, frozen

sections of tumors and premetastatic livers and lungs fromthe MC38 tumor model were stained with 1:50 dilution ofanti-mouse CD11b and anti-mouse Gr-1 (BD-Pharmin-gen) followed by Alexa-488- and Alexa-594–conjugatedsecondary antibodies (Invitrogen), and cover-slipped withVectaShield 40,6-diamidino-2-phenylindole (DAPI)mounting medium (Vector Laboratories). MC38 tumorsections were also separately blocked with glucose oxidase asbefore and stained with anti-mouse S100A9 (R&D Sys-tems), anti-mouse CD31 (BD-Pharmingen), or F4/80(Invitrogen), followed by 1:100 dilution of respective per-oxidase-conjugated secondary antibodies and developedwith DAB substrate. Slides were scanned using an auto-matic high-throughput ScanScope and viewed using Aperiosoftware. They were also examined using an InvertedTE300 Nikon Wide Field and Fluorescence Microscopeand images were acquired with a CCD SPOT RT Camera(Diagnostic Instruments Inc.), using SPOT advancedsoftware.

StatisticsStatistical comparisons were performed using paired t

test, and P values were calculated using GraphPad Prism.Differences were considered statistically significant whenP < 0.05.

Results

RAGE is the receptor for S100A8/A9 on colon tumorcellsWe earlier found S100A8/A9þCD11bþGr1þ myeloid

progenitor cells in the tumor microenvironment of colitis-induced colon tumors, whereas they were absent from thenormal adjacent colon tissue (22). Our subsequent study ina 4T1 model of mammary carcinoma showed that thesewere MDSC (7). Given that activated myeloid cells andMDSC can secrete S100A8/A9, we postulated that theremight be cross-talk between myeloid cells and colon tumorcells and that S100A8/A9 might interact with the tumorepithelium. In support of this, we earlier found that CT-26mouse colon tumor cells expressed binding sites forS100A8/A9 and that a subpopulation of RAGE expressedon these cells is modified by carboxylated glycans (22). Theproteins have been earlier shown to bind to RAGE onhuman prostate and breast cancer cells (21, 23). To furtherstudy cell-signaling pathways mediated by S100A8/A9interactions, in this study, we used MC38 colon tumorcells that are syngeneic to C57BL/6 strain, as S100A9 nullmice and RAGE null mice have been backcrossed to thisstrain. We first confirmed cell surface expression of RAGEand carboxylated glycans on MC38 colon tumor cells byflow cytometry (Fig. 1A). Examination of whole-cell lysatesalso showed that MC38 cells constitutively express bothRAGE and TLR4 (Fig. 1B and C). To investigate whetherRAGE or TLR4 provided binding sites for S100A8/A9 onMC38 colon tumor cells, we performed coimmunopreci-pitation assays. Lysates from MC38 cells incubatedwith S100A8 or S100A9 homodimers or S100A8/A9

Ichikawa et al.





heterodimer were immunoprecipitated with anti-S100A8 oranti-S100A9 antibodies and the immunoprecipitated pro-teins were separated by electrophoresis and immunoblottedwith anti-RAGE or anti-TLR4.We could only detect RAGEin lysates of cells incubated with S100A8 or S100A9 homo-dimers or the heterodimer but did not detect TLR4 (Fig. 1Band C). RAGEwas also present in immunoprecipitates fromcells not exposed to S100A8 or S100A9, suggesting thatsomeRAGE on the surface of colon tumor cells could exist asa complex with S100A8/A9 possibly from endogenoussources. To confirm this, we undertook immunoprecipita-

tion of MC38 cells in which endogenous S100A9 wassilenced using specific siRNA (knockdown was confirmedby Western blots using anti-S100A9, SupplementaryFig. S1) and found that RAGE in the immunoprecipitatewas significantly reduced in S100A9-silenced cells comparedwith control cells (Fig. 1B). No RAGE was detected incontrol immunoprecipitates obtained using an irrelevantcontrol rabbit antibody. Subsequently, we also found RAGEin immunoprecipitates of Caco-2 human colon cancer cellsincubatedwith human S100A8, S100A9, or the heterodimerbut not TLR4 (not shown). These results suggested thatRAGE could be the predominant receptor for S100A8/A9on colon tumor cells.

RAGE and carboxylated glycan–dependent binding ofS100A8/A9 promotes MAPK and NF-kB signalingAs RAGE ligation activates all members of the MAPK

cascades, including the p38, ERK, and the JNK families, andpromotesNF-kB activation (13), we next examined S100A8/A9 activated signaling pathways in colon tumor cells. Weanalyzed lysates of MC38 cells stimulated with low concen-trations (10mg/mL) of S100A8/A9 for varying periods of timeby immunoblotting, using specific phospho-MAPK antibo-dies. This concentration was chosen for stimulation, as ourearlier studies and that of Ghavami and colleagues show thatS100A8/A9 at 1 to 10mg/mL induced tumor cell growth (21,22). Ghavami and colleagues also showed that S100A8/A9 at10 mg/mL stimulated intracellular signaling in human breastcancer cell lines. In tumor-bearing mice, we found serumlevels of S100A8/A9 in the order of approximately 500 ng/mL(7). However, local concentrations in inflamed and tumortissues could be several micrograms permilliliter, for example,as shown in exudates of carrageenan-induced inflammation,where S100A9 levels are reported to be approximately 1 to4 mg/mL (39).S100A8/A9 stimulated rapid phosphorylation of ERK1/

ERK2 and SAPK/JNK in MC38 cells within 15 minutes,with reduced but sustained phosphorylation up to 60minutes, whereas there was no detectable phosphorylationof p38 (Fig. 2A). This effect was different from RAGE-dependent, S100A8/A9-induced activation triggered inhuman prostate and breast cancer cells (21, 23), whereERK1/ERK2 and p38 are activated but not SAPK/JNK,suggesting that S100A8/A9 preferentially activate differentMAPK pathways depending upon tumor cell type. Phos-phorylation of ERK1/ERK2 was also seen in Caco-2 humancolon tumor cells (Fig. 2B). ERK1/ERK2 phosphorylationwas inhibited in both MC38 and Caco-2 cells when theywere preincubated with mAbGB3.1 or anti-RAGE, sug-gesting that the effects were mediated through RAGE andcarboxylated glycans (Fig. 2B). Because S100A8 andS100A9 homodimers also bind to RAGE, we investigatedwhether homodimers would individually stimulate activa-tion. We found that S100A8 and S100A9 promoted a moredelayed activation of ERK1/ERK2 in MC38 cells whencompared with the heterodimers (Fig. 2C).S100A8/A9 also induced phosphorylation of IkBa

in MC38 and Caco-2 cells within 30 minutes (Fig. 3A),

Figure 1. A, RAGE and mAbGB3.1 glycans are expressed on colon tumorcells. MC38 colon tumor cells in culture were analyzed for surfaceexpression of mAbGB3.1 glycans and RAGE by flow cytometry. Cells werestained with mAbGB3.1 or anti-RAGE followed by FITC-conjugated anti-mouse or anti-rabbit Ig. Unstained cells (filled) and cells stained withsecondary antibody alone (dark line) served as negative control. B and C,receptor on colon tumor cells for S100A8/A9 were identified bycoimmunoprecipitation. MC38 cells were incubated with purified mouseS100A8/A9, S100A8, or S100A9 and bound proteins wereimmunoprecipitated with anti-S100A8 and/or anti-S100A9, or an irrelevantcontrol rabbit IgG. To confirm potential interaction of RAGE withendogenous S100 proteins, immunoprecipitation was also performedusingMC38 cells in which endogenous S100A9was silenced using target-specific siRNA. Whole-cell lysates and immunoprecipitated proteins wereseparated on SDS-PAGE gels, transferred and immunoblotted withanti-RAGE (B) or anti-TLR4 (C).






suggesting activation of NF-kB pathway. This was accom-panied by IkBa degradation in Caco-2 cells, evident by 30 to60 minutes, but not in MC38 cells. Native colonic epithelialcells, and many intestinal epithelial cells, have delayed orincomplete IkBa degradation following stimulation, despiteevidence for concomitant IkBa phosphorylation andNF-kBactivation (40). This has been attributed to impaired stimula-tion of an upstream IKK activator, and an altered steady-statelevel of IkBa, which is dependent on rate of resynthesis andstrength of the inducer (41).Therefore, to further confirm S100A8/A9-induced acti-

vation of NF-kB pathway, we examined nuclear extractsisolated from MC38 and Caco-2 colon tumor cells andfound considerable NF-kB–p65 in the extracts fromS100A8/A9 stimulated cells compared with unstimulatedcells (Fig. 3B). This effect was significantly inhibited whencells were pretreated with mAbGB3.1 or anti-RAGE IgG(Fig. 3B). These results indicated that RAGE and carboxy-

lated glycan–dependent binding of S100A8/A9 to RAGEon tumor cells resulted in activation of MAPK signalingpathways and nuclear translocation of NF-kB within thecells.

Stimulation of colon tumor cells by S100A8/A9promotes protumorigenic gene expressionMAPK pathways link extracellular signals with intracel-

lular responses promoting cell growth, proliferation, differ-entiation, and migration (42). Activation of NF-kB–dependent genetic programs in tumor cells and macro-phages is critical for development of inflammation-basedtumors (43–45). We therefore reasoned that gene expres-sion studies of activated colon tumor cells might providevaluable insight into the consequences of S100A8/A9 acti-vation. To identify whether S100A8/A9-activated signalingpathways promoted gene transcription, we isolated totalRNA from S100A8/A9 stimulated and unstimulatedMC38colon tumor cells and performed global gene expressionanalysis. Surprisingly, we found only a small cohort of 28differentially expressed genes (P < 0.01), of which, 24 wereupregulated and 4 were downregulated in stimulated cellscompared with unstimulated cells (Fig. 4A). Of these, theexpression of 14 genes was 2-fold or more when comparedwith nonstimulated cells. Most of the upregulated genesencoded proteins with either well-known or more recentlyrecognized functions in leukocyte migration and recruit-ment, inflammation, proliferation of tumor cells, tumorinvasion, angiogenesis, and wound healing (Table 1). Thesegenes included 4 chemokines (Cxcl1, Ccl5, Ccl2, Ccl7), NF-kB family member (Nfkbiz), zinc transporter (Slc39a10 orZip10), lipocalin-2 (Lcn2), Ectonucleotide pyrophospha-tase/phosphodiesterase family member 2 (Enpp2), zincfinger protein (Zc3h12a), guanylate binding protein 4(Gbp4), anti-leukoproteinase (Slpi), proliferin-2 (plf2),and apoptotic gene Fas. Many are upregulated in humantumors, their expression correlating with poor prognosis(46–51). qRT-PCR analysis of chemokine genes Cxcl1,Ccl5, and Ccl7 further confirmed elevated transcript levelsin S100A8/A9-stimulated cells (�6-fold increase in Cxcl1,2-fold increase in Ccl7, and 2.5-fold increase in Ccl5)compared with unstimulated cells (Fig. 4B). We nextexamined whether upregulation of the Cxcl1 gene in cellcultures stimulated with S100A8/A9 led to correspondingsecretion of CXCL1 (GROa or KC) protein. We analyzedculture supernatants of MC38 cells stimulated withS100A8/A9 for varying periods of time for CXCL1 byELISA. We found that unstimulated MC38 cells constitu-tively secreted CXCL1 (Fig. 4C). In addition, in line withthe gene expression data, S100A8/A9 activation led to a 3-fold increase in secretion of CXCL1 compared with unsti-mulated cells within 6 hours of stimulation (Fig. 4C). ThisS100A8/A9-induced secretion was significantly diminishedwhen cells were pretreated with anti-RAGE or mAbGB3.1.

Mouse models of colon tumorsOur earlier studies and those of Cheng and colleagues

have shown that S100A8 and S100A9 play a critical role in

Figure 2. S100A8/A9 activates MAPK signaling pathways in colon tumorcells. A, MC38 cells were incubated with purified endotoxin-free S100A8/A9 for different periods of time, and cell lysates were analyzed by Westernblotting, using respective antibodies against phosphorylated MAPK. Asloading controls, separate lanes with lysate proteins were incubated withrabbit polyclonal antibodies for total ERK1/ERK2, p38, SAPK/JNK, orb-actin B, MC38 or Caco-2 cells were incubated with purified endotoxin-free mouse or human S100A8/A9 for 15 minutes in the presence orabsence of mAbGB3.1 or RAGE, and cell lysates were analyzed byWestern blotting using respective antibodies against phosphorylated ortotal ERK1/ERK2 or b-actin. C, MC38 cells were incubated with purifiedmouse S100A8 or S100A9 homodimers for different periods of time, andcell lysates were analyzed by Western blotting using phosphorylated ortotal ERK1/ERK2 or b-actin.

Ichikawa et al.





tumor growth and metastasis through increased accumula-tionMDSC (6, 7). A role in growth and migration of tumorcells has also been described by many studies (21–27).However, molecular signature of S100A8/A9-activated cellsrevealed by gene expression analysis as shown previouslystrongly implied that RAGE and carboxylated glycan–dependent activation of tumor cells by S100A8/A9 differ-entially altered expression of genes whose products couldmediate many protumorigenic effects, predicting thatS100A8/A9 could have other novel roles in tumor progres-sion. To further elucidate protumorigenic and prometa-static roles of S100A8/A9 in vivo, we subjected S100A9 nullmice to different colon tumor models and comparedresponses with those observed in wild-type mice. Deletionof S100A9 in mice leads to a complete lack of S100A8 and afunctional S100A8/A9 complex in cells of peripheral bloodand bone marrow, despite normal mRNA levels of S100A8,

suggesting that S100A9 expression is important for thestability of S100A8 protein (37, 52). Induction of tumors inS100A9 null mice thus provided us an excellent opportu-nity to test the importance of both proteins in tumorigenesisand malignancy.

Reduced tumor incidence and chemokine expression inS100A9 null mice in the CAC modelIn the CAC model, we induced chemokine expression in

S100A9 null mice and wild-type mice by azoxymethane(AOM) injection followed by 2 cycles of dextran sodiumsulfate (DSS) treatment as described earlier (22). DSS causesepithelial damage and triggers an innate immune responsethat recruits activated macrophages and induces an acutecolitis evident within 2 weeks after DSS. This initialresponse progresses to chronic inflammation by about6 weeks by activation of adaptive immune responses(53). Wild-type mice develop dysplasia, adenoma, andadenocarcinoma within 12 to 20 weeks of combinedadministration of AOM and DSS (43, 54–56) with100% penetrance. Both S100A9 null mice and age-matched C57BL/6 wild-type mice lost up to 10% of bodyweight after DSS treatment before recovery, and colonsshowed inflammation in both S100A9 null mice and wild-type mice (not shown), suggesting that S100A8/A9 do notcontribute to DSS-induced colon inflammation. However,there was a significant reduction in tumor incidence inS100A9 null mice at 12 and 20 weeks after AOM/DSS(Fig. 5A and B). In contrast, all the wild-type mice devel-oped adenomas (5–8 tumors per mouse) by 12 to 20 weeks,with a few adenocarcinomas by 20 weeks, suggesting thatS100A8/A9 could exert independent roles in the tumori-genic phase of CAC.We had earlier shown that S100A8/A9þ and CD11bþ/

Gr1þ myeloid cells infiltrate all regions of dysplasia andtumors in this model (22). Because chemokines are upre-gulated in colon tumor cells in vitro in response to S100A8/A9, we examined whether chemokines CXCL1 and CCL7were also induced in the colon tumors in this model. Wefound moderate to intense staining for CXCL1 and CCL7in most epithelial and some stromal cells in tumor regionsbut not in adjacent normal tissues. Staining was reduced intumor regions from S100A9 null mice (Fig. 5C). We alsomeasured serum CXCL1 as a marker of S100A8/A9-induced activation of tumor cells. Serum CXCL1 waselevated 2- to 3-fold compared with pretumor levels inall wild-type mice, at 12 weeks of disease initiation, whenthe tumors are not invasive, whereas CXCL1 levels wereminimally altered in tumor-bearing S100A9 null mice(Fig. 5D). This further substantiated our in vitro findingsthat S100A9/A9 promoted expression of protumorigenicdownstream effectors in early tumors.

Bone marrow–derived cells in the tumormicroenvironment contribute S100A8/A9Epithelial cells can express S100A8/A9. To investigate

whether S100A8/A9 expressed by tumor cells or infiltrat-ing bone marrow–derived myeloid cells within the tumor

Figure 3. S100A8/A9 activates NF-kB in colon tumor cells. A, MC38 orCaco-2 cells were incubated with purified endotoxin-free mouse or humanS100A8/A9 for different periods of time, and cell lysates were analyzed byWestern blotting, using respective antibodies against phosphorylated ortotal IkBa or b-actin. B, MC38 or Caco-2 cells were treated with respectivepurified S100A8/A9 for 6 hours in the presence or absence of mAbGB3.1or anti-RAGE, and nuclear extracts were analyzed for NF-kBp65, usingDNA-binding ELISA.






microenvironment is required for disease progression, weevaluated tumorigenesis in chimeric mice after bonemarrow transplantation. Bone marrow cells from wild-type or S100A9 null mice were injected into lethallyirradiated groups of recipient wild-type or S100A9 nullmice. Chimerism was confirmed by measurement of

S100A8/A9 in serum (not shown). Mice were subjectedto the AOM/DSS protocol 4 weeks later and sacrificed12 weeks after initiation of disease. Tumor incidence inwild-type mice reconstituted with wild-type bone marrowcells (WT ! WT) and S100A9 null mice reconstitutedwith S100A9 null bone marrow cells (S100A9 null !

P < 0.05

P < 0.05

P < 0.05

A

B

C

Figure 4. A, profile of differentiallyexpressed genes from S100A8/A9-activated MC38 cellscompared with unstimulated cellsobtained by global geneexpression analysis. Fold variationrepresent the mean of replicate(n ¼ 2) analysis. B, cellular mRNAlevels of chemokines. mRNAlevels of Cxcl1, Ccl7, and Ccl5were measured by qRT-PCRusing RNA samples isolated fromcontrol MC38 cells, MC38 cellsstarved overnight and eitheruntreated or stimulated withS100A8/A9. The expressionvalues were normalized relative toGAPDH. The levels of mRNA inunstimulated or stimulated cellsare shown relative to controlnonstarved MC38 cellsconsidered as 100%. Each valueis the mean level � SD in 2different samples for eachcondition, each sample assayed induplicate. C, CXCL1 is secretedinto medium from activated MC38cultures. MC38 cultures wereactivated with S100A8/A9 inpresence or absence ofmAbGB3.1 or anti-RAGE, andCXCL1 in culture supernatantsharvested at different time pointswas measured by ELISA.

Ichikawa et al.





S100A9 null) was similar to responses seen earlier in wild-type mice and S100A9 null mice (Fig. 5E). However,S100A9 null mice reconstituted with bone marrow cellsfrom wild-type mice (WT! S100A9 null) showed higherincidence of tumors than wild-type mice reconstitutedwith bone marrow cells from S100A9 null (S100A9 !WT) mice. This strongly indicated that S100A8/A9expressed by bone marrow–derived cells in the tumormicroenvironment is essential for the promotion oftumorigenesis.

Reduced MC38 colon tumor growth and metastasis inS100A9 null miceThe CAC model described previously allows us to under-

stand the role of S100A8/A9 in early events in coloncarcinogenesis under the setting of inflammation. However,the tumors rarely became invasive and malignant within theexperimental period of 20 weeks. Therefore, to define therole of S100A8/A9 and its downstream effectors in tumorinvasion, myeloid cell migration, and formation of preme-tastatic niches in distal organs, we used a primary ectopictumor model using MC38 colon tumor cells. We followedtumor growth in wild-type and S100A9 null mice injected s.c. with 1 � 106 MC38 cells. Tumors were evident in allwild-type mice (n ¼ 10) by 7 to 10 days after injection andcontinued to grow until 21 days when the mice were

sacrificed. When S100A9 null mice were challenged withMC38 cells, tumors were significantly smaller in 6 of 12S100A9 null mice at 21 days after injection (Fig. 6A). Inaddition, tumors were completely rejected in 2 of theremaining 6 S100A9 null mice. Collectively, 8 of 12S100A9 null mice (67%) examined showed minimal tumorgrowth or tumor rejection. Tumor growth in wild-typemice was accompanied by elevated serum CXCL1 levels butnot in S100A9 null mice (Fig. 6B).Because CXCL1 promotes MDSC and other myeloid cell

recruitment within tumors and premetastatic organs, wemeasured bone marrow responses to the ectopic MC38tumors and found significantly increased CD11bþ popula-tions coexpressing Gr1, carboxylated glycans (as stained bymAbGB3.1), or RAGE in all of the tumor-bearing wild-type mice at 18 to 21 days after transplantation comparedwith tumor-free control mice (Fig. 6C). CD11bþGr1þ cellswere also found within the tumors and substantially reducedin tumors from S100A9 null mice, whereas the levels of F4/80þ macrophages and CD31þ endothelial cells wereunchanged, suggesting that S100A8/A9 do not alter intra-tumoral infiltration of other tumor-associated, angiogenicmacrophages, while affecting infiltration of CD11bþGr1þ

cells (Fig. 6D). To confirm that these were in fact MDSC,we isolated CD11bþGr1þ cells from the spleens of MC38tumor-bearing mice and cocultured them at varying ratios

Table 1. Protumorigenic genes activated in colon tumor cells by S100A8/A9

Gene Accession number Known functions

Cxcll NM-008176.1 Chemokine, promote angiogenesis, mobilization ofleukocytes including MDSC

C3 NM.009778.1 Complement component 3, mobilization of HSC into tumor stromaSlc39al0

NfkbizNM_172653.2

NM_030612Zinc transporter, belongs to Zip family (ZiplO), upregulated in endometrialcarcinoma, promote migration of breast tumor cells IkB family, IL-6 activator,modulates NF-kB transcription

Lcn2 NM.008491.1 Lipocalin 2, upregulated in inflammation, has pro- and antitumor effectsFas l\IM_007987.1 Apoptosis, TNF family member with the death domainZc3hl2a NM_153159.1 Zinc finger family, RNase, controls stability of inflammatory genes, mediates

CCL2 induced angiogenesis, macrophage activationEnpp2 NM_015744.1 Ectonucleotide pyrophosphatase/phosphodiesterase family member

2, autotaxin, promotes tumor cell migration (invadopodia), angiogenesis,and its expression is upregulated in several kinds of carcinomas

Ccl5 NM_013653.1 Chemokine, stimulate angiogenesisCcl2 NM_013653.2 Chemokine, potent stimulator of angiogenesisCcl7 NM_013653.3 Chemokine, promote macrophage infiltration into tumorsGbp4 l\IM_018734.2 Guanylate binding protein 4; gene family induced during macrophage activation,

IFN gamma inducible, GTPaseSlpi NM.011414.1 Secretory leucocyte peptidase inhibitor, upregulated in tumors, secreted

inhibitor, which protects epithelial tissues from serine proteases, implicatedin wound healing

Plf2 NM.011118.1 Proliferin-2, HSC growth factor, associated with angiogenesis andwound healing

NOTE: From refs. 47–51, 66–78.






with CD4þ T cells from OTII transgenic mice and OVApeptide (OVA323-339) and measured T-cell proliferationby uptake of [3H]thymidine. With increasing ratios ofMDSC: T cells, splenic CD11bþGr1þ cells from tumor-

bearing mice progressively reduced T-cell proliferation(Supplementary Fig. S2).A more recent study by Connolly and colleagues shows

that the expansion of CD11bþGr1þ MDSC in premeta-

Figure 5. A, representative H&E-stained colon "Swiss rolls"obtained from wild-type andS100A9 null mice subjected to theAOM-DSS protocol 12weeks afterinitiation (10� magnification).Arrows indicate regions ofdysplasia and adenoma. B,colonic tumor incidence inS100A9 null and wild-type mice12 and 20 weeks after AOM-DSS(n ¼ 5 mice per group per timepoint). C, representative sectionsof tumor regions and normaladjacent tissue in colons of wild-type and S100A9 null micesubjected to the AOM-DSSprotocol 20 weeks after initiationstained for CXCL1 or CCL7(400�). D, CXCL1 in sera of wild-type and S100A9 null mice beforeand 12 weeks after AOM-DSS(n ¼ 5 mice). E, colonic tumorincidence in bone marrowchimeric mice 12 weeks afterAOM-DSS (n ¼ 4 recipient miceper group).

Ichikawa et al.





static sites in liver in response to intraabdominal tumors iscontingent upon the expression of CXCL1 (57). In keepingwith observation, and with elevated serum CXCL1 levels,we found markedly increased accumulation of CD11bþ/

Gr1þ cells in the premetastatic lungs and liver of tumor-bearing mice 21 days after tumor initiation, compared withtumor-free mice (Fig. 6E, shown for liver). To exclude thepossibility that any micrometastasis of MC38 cells in liver

Figure 6. A, tumor volumes of ectopic MC38 tumors in wild-type (n ¼ 10) and S100A9 null mice (n ¼ 12) 3 weeks after s.c. injection of 1 � 106 cells.Six of 12 S100A9 null mice showed significantly reduced tumor growth shown here. In addition, 2 of the remaining 6 S100A9 null mice completely rejected thetumors. B, CXCL1 in sera of wild-type and S100A9 null mice before and 3 weeks after MC38 tumor growth. C, quantitation of CD11bþ cells costaining with Gr1or GB3.1 glycans or RAGE from bone marrow of MC38 tumor-bearing wild-type mice D, tumors were examined for infiltrating macrophages and tumorendothelial cells by immunochemical staining for S100A9þ cells, CD11bþGr1þ cells (merged images of Alexa-488–stained CD11bþ cells and Alexa-594–stained Gr1þ cells showing double-positive yellow cells), and CD31� and F4/80� cells (200�). E, representative sections showing CD11bþGr1þ cells inpremetastatic livers of tumor-free and tumor-bearing mice. F, quantitation of average number of CD11bþGr1þ cells in premetastatic livers in 3 highpower fields.






and lungs induced the accumulation of CD11bþGr1þ cells,we injected separate groups of wild-type mice with MC38cells stably expressing GFP. No GFPþ cells were detected inlungs or livers at 21 days after tumor initiation (not shown).MC38 tumor challenge in mice lacking S100A8/A9, orwild-type mice treated with mAbGB3.1 significantly dimin-ished accumulation of CD11bþGr1þ-positive niches inliver (Fig. 6E and F). This finding is consistent with thestudies of Hiratsuka and colleagues, who showed S100A8/A9 promote the formation of premetastatic niches in distalorgans in response to primary tumors (24).The liver is the primary site for colorectal carcinoma

metastasis. Because the CAC model and ectopic MC38tumor models did not show any evidence of distalmetastasis, we chose a liver metastasis model to furtherunderstand the role of S100A8/A9- and S100A8/A9-induced proteins in promoting metastasis. We injectedS100A9 null mice and age-matched C57BL/6 wild-typemice with 1 � 106 MC38 cells by the intrasplenic route.

MC38 cells generated tumors within the spleen (pri-mary) and in the liver (metastasis). Multiple hepatictumor nodules, detectable by gross inspection, wereevident by 2 weeks. Livers were isolated and the inci-dence of hepatic metastases was evaluated. Livers fromS100A9 null mice showed significantly reduced numbersof metastatic tumors, smaller tumor foci, and decreasedtumor-occupied area compared with livers from tumor-bearing wild-type mice (Fig. 7). These results furtherindicate that S100A8/A9 play a critical role in promotingmetastasis.Taken together, our observations strongly support the

notion that S100A8/A9 activate signaling pathways thatpromote tumor growth and metastasis by inducing expres-sion of multiple downstream protumorigenic effector pro-teins and suggest that strategies that target S100A8/A9 inthe tumor microenvironment could provide effective ther-apeutic approaches to treating patients with colorectalcancer.

Figure 7. S100A9 null mice exhibitreduced metastatic tumors.A, representative livers from wild-type and S100A9 null mice2 weeks after intrasplenic injectionof MC38 cells. Arrows indicatevisible tumors. B, histology ofrepresentative livers stained byH&E (25� magnification).C, numbers of metastatic nodulesin the livers, and total tumor arearepresented as % of liver tissue,2 weeks after intrasplenic injectionof MC38 cells (wild-type, n ¼ 6)and S100A9 null mice (n ¼ 5).

Ichikawa et al.





Discussion

Cells of the tumormicroenvironment contribute to tumorgrowth and metastasis through complex interactions withtumor cells (58–60). The presence of S100A8/A9 in manyhuman tumors, alongwith recent recognition of their roles intumorigenesis and MDSC accumulation, warrants a moredetailed understanding of the molecular mechanismsinvolved in their interactions within the tumor microenvir-onment. Our earlier studies provided evidence that S100A8/A9 promote accumulation ofMDSC (7). Here we show thatS100A8/A9 expressed by myeloid cells interacts with RAGEand carboxylated glycans expressed on colon tumor cellspromoting intracellular signaling pathways and protumori-genic gene expression and that S100A9 null mice showreduced tumor growth and metastasis, thus defining yetanother novel role for S100A8/A9 in tumor progression.Although many studies implicate both TLR4 and RAGE

in S100A8/A9-mediated pathologic effects, the relativecontribution of each receptor to downstream effects isunknown. Based on our earlier studies and immunopreci-pitation results shown here, we surmise that RAGE is theprincipal receptor of S100A8/A9 on tumor cells, and this isconsistent with the finding that S100A8/A9-mediatedresponses in human tumor cells involves RAGE (21, 23).However, studies implicating S100A8/A9 in endotoxin-induced lethality and systemic autoimmunity show thatTLR4, rather than RAGE, could play a more prominentrole as receptor for these ligands on macrophages (17, 18).This suggests that cell types and pathologic settings coulddictate which receptor predominates. Besides cell types, thedifferential effects could also be mediated by carboxylatedglycans, which are expressed on RAGE and not on TLR4.Also, TLR4 is only functional active in the presence ofmyeloid differentiation factor-2 (MD2) protein for bothLPS and S100A9-mediated interactions on macrophages(17). TLR4 expressed on MC38 cells is functional, as it hasbeen shown to respond to LPS, as determined by LPS-induced expression of IL-6 byMC38 cells, which is reducedupon TLR4 gene silencing (61).The activation of RAGE-mediated signaling pathways

could also depend on the tumor cell involved. We foundthat S100A8/A9 induces RAGE and carboxylated glycan–dependent phosphorylation of ERK1/ERK2 and SAPK/JNK MAPK in colon tumor cells, but we did not observesignificant phosphorylation for p38. In contrast, in humanprostate and breast cancer cells, S100A8/A9 activate p38but not SAPK/JNK (21, 23). In this context, it is interestingthat S100A8/A9 activate SAPK/JNK in macrophagesthrough TLR-dependent pathway (62). In support of ourfinding, it was recently shown that treatment of tumor cellswith a JNK inhibitor blocked RAGE ligand-induced cel-lular invasion (63). p38 and SAPK/JNKMAPK proteins areknown to function in cell context and cell type–specificmanner to coordinate signaling pathways mediating tumorcell proliferation, survival, and migration and may evenexert antagonistic effects, depending on signal duration andcross-talk with other signaling pathways (64). Their expres-

sion is altered in many human tumors, and it is thereforeimportant to consider the tumor type before modulations ofthe pathways are attempted for therapeutics.S100A8/A9 binding to colon tumor cells stimulates

RAGE and carboxylated glycan–dependent activation ofNF-kB pathway. NF-kB provides a critical link betweeninflammation and cancer (43, 44). As the binding ofS100A8/A9 to cells stimulates NF-kB transcription, andproximal promoter regions of S100A8 and S100A9 havebinding sites for NF-kB (10), ligation of cell surfacereceptors by S100A8/A9 in inflammation could lead to apositive feedback loop and sustained cellular activationpromoting tumor development. In support of this,S100A8/A9 proteins have been identified as novel NF-kB target genes in hepatic carcinoma cells during inflam-mation-mediated liver carcinogenesis (65).Our gene expression analysis revealed for the first time the

molecular signature of S100A8/A9 activation in tumor cells.Some of the genes that are activated represent known NF-kB target genes and are directly associated with tumorigen-esis. Most notable are the chemokines CXCL1 (GROa orKC), CCL2 (MCP-1), CCL5 (RANTES), and CCL7(MCP-3). Although many previous studies have focusedon the role of chemokines in immune responses, recentstudies show that they also promote chemotaxis and leu-kocyte recruitment to tumors, angiogenesis, bidirectionalcross-talk between tumor cells and tumor-associated fibro-blasts, tumor invasion, and migration of tumor cells to distalorgans (47, 48, 66). CXCL1, CCL2, and CCL5 have beenalso identified in human colorectal tumors and expressioncorrelates with poor prognosis (46, 67, 68). CCL2 is acrucial mediator of CAC in mice (69) and CXCL1 mediatesproangiogenic effects of PGE2 in colorectal cancer (70).More recent studies show that the expansion ofCD11bþGr1þ MDSC in premetastatic sites in liver inresponse to intraabdominal tumors is contingent uponthe expression of CXCL1, and self-seeding of circulatingtumor cells promote tumor growth, angiogenesis, andtumor recruitment through mediators such as CXCL1(57, 71). We found that CXCL1 is secreted by colon tumorcells in response to interaction with S100A8/A9 and isupregulated in colon tumors. It is elevated in sera of notonly the MC38 tumor-bearing mice but also in mice withcolitis-induced tumors within 12 weeks of AOM-DSStreatment, at which time point the tumors are early,well-contained, and noninvasive, suggesting that CXCL1could provide an early marker of metastatic tumor progres-sion. We also found that tumor-bearing mice lackingS100A8/A9 show marginal or no elevation of CXCL1and significantly diminished CD11bþGr1þ cells in tumorsand premetastatic niches in liver and lungs.In addition, we found other new S100A8/A9-induced

genes that are implicated in tumor progression. Zc3h12a,which encodes a zinc finger protein, is an RNase and adownstream effector of CCL2-mediated angiogenesis(72). Enpp2 encodes autotaxin, a lysophospholipase Denzyme that hydrolyzes extracellular lysophospholipids toproduce lysophosphatidic acid (LPA). LPA receptors are






overexpressed in many tumors (50), and LPA2 receptorknockout mice show markedly reduced tumor incidenceand progression of colon adenocarcinomas associated withreduced tumor-infiltrating macrophages (73). Morerecently, Liu and colleagues showed a causal link betweenautotaxin-LPA receptor signaling and mammary tumorprogression (74). Autotaxin is also implicated in the for-mation of invadopodia by various human cancer cell types(75). Slpi encodes a secretory leukocyte peptidase inhibitorthat is upregulated in tumors. It protects epithelial tissuesfrom serine proteases and has been implicated in woundhealing (76). Neutrophil gelatinase–associated lipocalin-2encoded by NF-kB inducible Lcn2 gene has paradoxicallyboth pro- and antitumor effects (52). Plf2 encodes prolif-erin-2, a hematopoietic stem cell growth factor associatedwith angiogenesis and wound healing (77, 78).The induction of these downstream effector genes in

tumors would thus greatly amplify tumor growth, migrationand invasion, induction of myeloid cells, and metastaticprogression promoted by S100A8/A9. Consistent with this,we found that S100A9 null mice showed markedly reducedtumor incidence and progression of AOM-DSS inducedcolon adenomas and reduced ectopic MC38 tumor growthand tumor metastasis. Because CD11bþGr1þ cells in thetumor-bearing mice are MDSC, it is likely that reducedtumor growth and metastasis in S100A9 null mice are dueto combined effects of lack of immune suppression andreduced induction of protumorigenic genes. The effects ofS100A8/A9 in tumorigenesis in the AOM-DSS modelcould be mediated through RAGE and carboxylated gly-cans, as we earlier showed that RAGE null mice and wild-type mice receiving mAbGB3.1 treatment showed reducedAOM-DSS induced tumor incidence. However, the con-tribution of TLR4 in S100A8/A9-mediated effects intumors cannot be overlooked, because TLR4 null mice

are protected markedly from CAC (79) and TLR4 mediatesthe formation of premetastatic niches promoted byS100A8/A9 (32).Colorectal cancer is one of the most common malignan-

cies affecting both sexes and a common cause of mortalityworldwide. Each year about 50,000 people die from thedisease in the Unites States alone. Our present findings,along with earlier studies, show that S100A8/A9 function atmultiple stages in disease progression. S100A8/A9, theirreceptors and signaling pathways therefore provide impor-tant targets for development of pharmacological interven-tions and for the identification of early-stage diseasebiomarkers.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Dr. Hudson Freeze for his long-standing support and collaboration, andfor his critical reading of the manuscript. We also thank Drs. Dirk Foell and JohannesRoth at the University of Muenster, Germany, for their collaboration. We gratefullyacknowledge the invaluable help from Adriana Charbono, Buddy Charbono, andLarkin Slater with the animal experiments and colony maintenance; Robbin Newlinand Gia Garcia for histology expertise; Kang Liu and Jian Xing for microarray andqRT-PCR analysis; and Harish Khandrika for illustrations. Microarray data describedin this manuscript have been deposited at the NCBI Gene Expression Omnibus(GEO) repository. Accession number is GSE2635.

Grant Support

NIH grant R21-CA127780 (G. Srikrishna).The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

Received August 27, 2010; revised December 8, 2010; accepted December 20,2010; published OnlineFirst January 12, 2011.

References1. Donato R. S100: a multigenic family of calcium-modulated proteins of

the EF-hand type with intracellular and extracellular functional roles.Int J Biochem Cell Biol 2001;33:637–68.

2. Heizmann CW, Fritz G, Schafer BW. S100 proteins: structure, func-tions and pathology. Front Biosci 2002;7:d1356–68.

3. Marenholz I, Heizmann CW, Fritz G. S100 proteins in mouse andman: from evolution to function and pathology (including an updateof the nomenclature). Biochem Biophys Res Commun 2004;322:1111–22.

4. Roth J, Vogl T, Sorg C, Sunderk€otter C. Phagocyte-specific S100proteins: a novel group of proinflammatory molecules. Trends Immu-nol 2003;24:155–8.

5. Goyette J, Geczy CL, Inflammation-associated S100 proteins: newmechanisms that regulate function. Amino Acids 2010.

6. Cheng P, Corzo CA, Luetteke N, Yu B, Nagaraj S, Bui MM, et al.Inhibition of dendritic cell differentiation and accumulation of myeloid-derived suppressor cells in cancer is regulated by S100A9 protein.J Exp Med 2008;205:2235–49.

7. Sinha P, Okoro C, Foell D, Freeze HH, Ostrand-Rosenberg S, Srik-rishna G. Proinflammatory S100 proteins regulate the accumulation ofmyeloid-derived suppressor cells. J Immunol 2008;181:4666–75.

8. Lagasse E, Clerc RG, Cloning and expression of two human genesencoding calcium-binding proteins that are regulated during myeloiddifferentiation. Mol Cell Biol 1988;8:2402–10.

9. Odink K, Cerletti N, Br€uggen J, Clerc RG, Tarcsay L, Zwadlo G, et al.Two calcium-binding proteins in infiltrate macrophages of rheumatoidarthritis. Nature 1987;330:80–2.

10. Gebhardt C, N�emeth J, Angel P, Hess J. S100A8 and S100A9 ininflammation and cancer. Biochem Pharmacol 2006;72:1622–31.

11. Foell D, Wittkowski H, Vogl T, Roth J. S100 proteins expressed inphagocytes: a novel group of damage-associated molecular patternmolecules. J Leukoc Biol 2007;81:28–37.

12. Srikrishna G, Freeze HH. Endogenous Damage associated molecularpattern (DAMP) molecules at the crossroads of inflammation andcancer. Neoplasia 2009;11:615–28.

13. Donato R. RAGE: a single receptor for several ligands and differentcellular responses: the case of certain S100 proteins. Curr Mol Med2007;7:711–24.

14. Leclerc E, Fritz G, Vetter SW, Heizmann CW. Binding of S100 proteinsto RAGE: an update. Biochim Biophys Acta 2009;1793:993–1007.

15. Ehrchen JM, Sunderk€otter C, Foell D, Vogl T, Roth J. The endogenousToll-like receptor 4 agonist S100A8/S100A9 (calprotectin) as innateamplifier of infection, autoimmunity, and cancer. J Leukoc Biol2009;86:557–66.

16. Roth J, Goebeler M, Sorg C. S100A8 and S100A9 in inflammatorydiseases. Lancet 2001;357:1041.

17. Vogl T, Tenbrock K, Ludwig S, Leukert N, Ehrhardt C, van Zoelen MA,et al. Mrp8 and Mrp14 are endogenous activators of Toll-like receptor

Ichikawa et al.





4, promoting lethal, endotoxin-induced shock. Nat Med 2007;13:1042–9.

18. Loser K, Vogl T, Voskort M, Lueken A, Kupas V, Nacken W, et al. TheToll-like receptor 4 ligands Mrp8 and Mrp14 are crucial in the devel-opment of autoreactive CD8þ T cells. Nat Med 2010;16:713–7.

19. Salama I, Malone PS, Mihaimeed F, Jones JL. A review of the S100proteins in cancer. Eur J Surg Oncol 2008;34:357–64.

20. Ghavami S, Kerkhoff C, LosM, Hashemi M, Sorg C, Karami-Tehrani F.Mechanism of apoptosis induced by S100A8/A9 in colon cancer celllines: the role of ROS and the effect of metal ions. J Leukoc Biol2004;76:169–75.

21. Ghavami S, Rashedi I, Dattilo BM, Eshraghi M, Chazin WJ, HashemiM, et al. S100A8/A9 at low concentration promotes tumor cell growthvia RAGE ligation andMAP kinase-dependent pathway. J Leukoc Biol2008;83:1484–92.

22. TurovskayaO, Foell D, Sinha P, Vogl T, Newlin R, Nayak J, et al. RAGE,carboxylated glycans and S100A8/A9 play essential roles in colitis-associated carcinogenesis. Carcinogenesis 2008;29:2035–43.

23. Hermani A, De Servi B, Medunjanin S, Tessier PA, Mayer D. S100A8and S100A9 activate MAP kinase and NF-kappaB signaling pathwaysand trigger translocation of RAGE in human prostate cancer cells. ExpCell Res 2006;312:184–97.

24. Hiratsuka S, Watanabe A, Aburatani H, Maru Y. Tumour-mediatedupregulation of chemoattractants and recruitment of myeloid cellspredetermines lung metastasis. Nat Cell Biol 2006;8:1369–75.

25. Moon A, Yong HY, Song JI, Cukovic D, Salagrama S, Kaplan D, et al.Global gene expression profiling unveils S100A8/A9 as candidatemarkers in H-ras-mediated human breast epithelial cell invasion. MolCancer Res 2008;6:1544–53.

26. Saha A, Lee YC, Zhang Z, Chandra G, Su SB, Mukherjee AB. Lack ofan endogenous anti-inflammatory protein in mice enhances coloniza-tion of B16F10 melanoma cells in the lungs. J Biol Chem 2010;285:10822–31.

27. Ang CW, Nedjadi T, Sheikh AA, Tweedle EM, Tonack S, Honap S, et al.Smad4 loss is associated with fewer S100A8-positive monocytes incolorectal tumors and attenuated response to S100A8 in colorectaland pancreatic cancer cells. Carcinogenesis 2010;31:1541–51.

28. Dolcetti L, Marigo I, Mantelli B, Peranzoni E, Zanovello P, Bronte V.Myeloid-derived suppressor cell role in tumor-related inflammation.Cancer Lett 2008;267:216–25.

29. Gabrilovich DI, Nagaraj S. Myeloid-derived suppressor cells as reg-ulators of the immune system. Nat Rev Immunol 2009;9:162–74.

30. Ostrand-Rosenberg S, Sinha P. Myeloid-derived suppressor cells:linking inflammation and cancer. J Immunol 2009;182:4499–506.

31. Ostrand-Rosenberg S. Cancer and complement. Nat Biotechnol2008;26:1348–9.

32. Hiratsuka S, Watanabe A, Sakurai Y, Akashi-Takamura S, Ishibashi S,Miyake K, et al. The S100A8-serum amyloid A3-TLR4 paracrinecascade establishes a pre-metastatic phase. Nat Cell Biol2008;10:1349–55.

33. Gebhardt C, Riehl A, Durchdewald M, N�emeth J, F€urstenberger G,M€uller-Decker K, et al. RAGE signaling sustains inflammation andpromotes tumor development. J Exp Med 2008;205:275–85.

34. Srikrishna G, Toomre DK, Manzi A, Panneerselvam K, Freeze HH,Varki A, et al. A novel anionicmodification of N-glycans onmammalianendothelial cells is recognized by activated neutrophils andmodulatesacute inflammatory responses. J Immunol 2001;166:624–32.

35. Srikrishna G, Turovskaya O, Shaikh R, Newlin R, Foell D, Murch S,et al. Carboxylated Glycans Mediate Colitis through Activation of NF-{kappa}B. J Immunol 2005;175:5412–22.

36. Vogl T, Leukert N, Barczyk K, Strupat K, Roth J. Biophysical char-acterization of S100A8 and S100A9 in the absence and presence ofbivalent cations. Biochim Biophys Acta 2006;1763:1298–306.

37. Manitz MP, Horst B, Seeliger S, Strey A, Skryabin BV, Gunzer M, et al.Loss of S100A9 (MRP14) results in reduced interleukin-8-inducedCD11b surface expression, a polarized microfilament system, anddiminished responsiveness to chemoattractants in vitro. Mol Cell Biol2003;23:1034–43.

38. Suzuki R, Kohno H, Sugie S, Tanaka T. Sequential observations on theoccurrence of preneoplastic and neoplastic lesions in mouse colon

treated with azoxymethane and dextran sodium sulfate. Cancer Sci2004;95:721–7.

39. Shibata F, Miyama K, Shinoda F, Mizumoto J, Takano K, NakagawaH.Fibroblast growth-stimulating activity of S100A9 (MRP-14). Eur JBiochem 2004;271:2137–43.

40. Jobin C, Haskill S, Mayer L, Panja A, Sartor RB. Evidence for alteredregulation of I kappa B alpha degradation in human colonic epithelialcells. J Immunol 1997;158:226–34.

41. Russo MP, Schwabe RF, Sartor RB, Jobin C. NF-kappaB-inducingkinase restores defective IkappaB kinase activity and NF-kappaBsignaling in intestinal epithelial cells. Cell Signal 2004;16:741–50.

42. Kim EK, Choi EJ. Pathological roles of MAPK signaling pathways inhuman diseases. Biochim Biophys Acta 2010;1802:396–405.

43. Greten FR, Eckmann L, Greten TF, Park JM, Li ZW, Egan LJ, et al.IKKbeta links inflammation and tumorigenesis in a mouse model ofcolitis-associated cancer. Cell 2004;118:285–96.

44. Karin M, Greten FR. NF-kappaB: linking inflammation and immunity tocancer development and progression. Nat Rev Immunol 2005;5:749–59.

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

46. Erreni M, Bianchi P, Laghi L, Mirolo M, Fabbri M, Locati M, et al.Expression of chemokines and chemokine receptors in human coloncancer. Methods Enzymol 2009;460:105–21.

47. Mantovani A. Chemokines in neoplastic progression. Semin CancerBiol 2004;14:147–8.

48. Hembruff SL, Cheng N. Chemokine signaling in cancer: Implicationson the tumor microenvironment and therapeutic targeting. CancerTher 2009;7:254–267.

49. Mishra P, Banerjee D, Ben-Baruch A. Chemokines at the crossroadsof tumor-fibroblast interactions that promote malignancy. J LeukocBiol 2010;89:31–9.

50. Liu S, Murph M, Panupinthu N, Mills GB. ATX-LPA receptor axis ininflammation and cancer. Cell Cycle 2009;8:3695–701.

51. Bolignano D, Donato V, Lacquaniti A, FazioMR, Bono C, Coppolino G,et al. Neutrophil gelatinase-associated lipocalin (NGAL) in humanneoplasias: a new protein enters the scene. Cancer Lett 2010;288:10–6.

52. Hobbs JA, May R, Tanousis K, McNeill E, Mathies M, Gebhardt C,et al. Myeloid cell function in MRP-14 (S100A9) null mice. Mol Cell Biol2003;23:2564–76.

53. Egger B, Procaccino F, Lakshmanan J, Reinshagen M, Hoffmann P,Patel A, et al. Mice lacking transforming growth factor alpha have anincreased susceptibility to dextran sulfate-induced colitis. Gastroen-terology 1997;113:825–32.

54. Tanaka T, Kohno H, Suzuki R, Yamada Y, Sugie S, Mori H. A novelinflammation-related mouse colon carcinogenesis model induced byazoxymethane and dextran sodium sulfate. Cancer Sci 2003;94:965–73.

55. Dieleman LA, Palmen MJ, Akol H, Bloemena E, Peña AS, MeuwissenSG, et al. Chronic experimental colitis induced by dextran sulphatesodium (DSS) is characterized by Th1 and Th2 cytokines. Clin ExpImmunol 1998;114:385–91.

56. Melgar S, Karlsson A, Michaelsson E. Acute colitis induced by dextransulphate sodium progresses into chronicity in C57BL/6 but not inBALB/c mice – correlation between symptoms and inflammation. AmJ Physiol Gastrointest Liver Physiol 2005;288:G1328–38.

57. Connolly MK, Mallen-St Clair J, Bedrosian AS, Malhotra A, Vera V,Ibrahim J, et al. Distinct populations of metastases-enabling myeloidcells expand in the liver of mice harboring invasive and preinvasiveintra-abdominal tumor. J Leukoc Biol 2010;87:713–25.

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

59. Mantovani A, Garlanda C, Allavena P. Cancer-related inflammation.Nature 2008;454:436–44.

60. Qian BZ, Pollard JW. Macrophage diversity enhances tumor progres-sion and metastasis. Cell 2010;141:39–51.

61. Earl TM, Nicoud IB, Pierce JM, Wright JP, Majoras NE, Rubin JE, et al.Silencing of TLR4 decreases liver tumor burden in a murine model of






colorectal metastasis and hepatic steatosis. Ann Surg Oncol2009;16:1043–50.

62. Pouliot P, Plante I, Raquil MA, Tessier PA, Olivier M. Myeloid-relatedproteins rapidly modulate macrophage nitric oxide production duringinnate immune response. J Immunol 2008;181:3595–601.

63. Kalea AZ, See F, Harja E, Arriero M, Schmidt AM, Hudson BI. Alter-natively spliced RAGEv1 inhibits tumorigenesis through suppressionof JNK signaling. Cancer Res 2010;70:5628–38.

64. Wagner EF, Nebreda AR. Signal integration by JNK and p38 MAPKpathways in cancer development. Nat Rev Cancer 2009;9:537–49.

65. Nemeth J, Stein I, Haag D, Riehl A, Longerich T, Horwitz E, et al.S100A8 and S100A9 are novel nuclear factor kappa B target genesduring malignant progression of murine and human liver carcinogen-esis. Hepatology 2009;50:1251–62.

66. Mantovani A, Allavena P, Sozzani S, Vecchi A, Locati M, Sica A.Chemokines in the recruitment and shaping of the leukocyte infiltrateof tumors. Semin Cancer Biol 2004;14:155–60.

67. Danese S, Mantovani A. Inflammatory bowel disease and intestinalcancer: a paradigm of the Yin-Yang interplay between inflammationand cancer. Oncogene 2010;29:3313–23.

68. Horst D, Budczies J, Brabletz T, Kirchner T, Hlubek F. Invasionassociated up-regulation of nuclear factor kappaB target genes incolorectal cancer. Cancer 2009;115:4946–58.

69. Popivanova BK, Kostadinova FI, Furuichi K, Shamekh MM, Kondo T,Wada T, et al. Blockade of a chemokine, CCL2, reduces chroniccolitis-associated carcinogenesis in mice. Cancer Res 2009;69:7884–92.

70. Wang D, Wang H, Brown J, Daikoku T, Ning W, Shi Q, et al. CXCL1induced by prostaglandin E2 promotes angiogenesis in colorectalcancer. J Exp Med 2006;203:941–51.

71. Kim MY, Oskarsson T, Acharyya S, Nguyen DX, Zhang XH, Norton L,et al. Tumor self-seeding by circulating cancer cells. Cell2009;139:1315–26.

72. Matsushita K, Takeuchi O, Standley DM, Kumagai Y, Kawagoe T,Miyake T, et al. Zc3h12a is an RNase essential for controlling immuneresponses by regulating mRNA decay. Nature 2009;458:1185–90.

73. Lin S, Wang D, Iyer S, Ghaleb AM, Shim H, Yang VW, et al. Theabsence of LPA2 attenuates tumor formation in an experimentalmodel of colitis-associated cancer. Gastroenterology 2009;136:1711–20.

74. Liu S, Umezu-Goto M, Murph M, Lu Y, Liu W, Zhang F, et al.Expression of autotaxin and lysophosphatidic acid receptorsincreases mammary tumorigenesis, invasion, and metastases. Can-cer Cell 2009;15:539–50.

75. Harper K, Arsenault D, Boulay-Jean S, Lauzier A, Lucien F, DuboisCM. Autotaxin promotes cancer invasion via the lysophosphatidicacid receptor 4: participation of the cyclic AMP/EPAC/Rac1 signalingpathway in invadopodia formation. Cancer Res 2010;70:4634–43.

76. Devoogdt N, Revets H, Ghassabeh GH, De Baetselier P. Secretoryleukocyte protease inhibitor in cancer development. Ann N Y Acad Sci2004;1028:380–9.

77. Choong ML, Tan AC, Luo B, Lodish HF. A novel role for proliferin-2 inthe ex vivo expansion of hematopoietic stem cells. FEBS Lett2003;550:155–62.

78. Fassett JT, Nilsen-Hamilton M. Mrp3, a mitogen-regulated protein/proliferin gene expressed in wound healing and in hair follicles.Endocrinology 2001;142:2129–37.

79. Fukata M, Chen A, Vamadevan AS, Cohen J, Breglio K, KrishnareddyS, et al. Toll-like receptor-4 promotes the development of colitis-associated colorectal tumors. Gastroenterology 2007;133:1869–81.

Ichikawa et al.





Correction

Correction: S100A8/A9 Activate Key Genes andPathways in Colon Tumor Progression

In this article (Mol Cancer Res 2011;9:133–48), which was published in the February,2011 issue of Molecular Cancer Research (1), an incorrect microarray data accessionnumber was published. The correct accession number is GSE26359.

Reference1. Ichikawa M, Williams R, Wang L, Vogl T, Srikrishna G. S100A8/A9 activate key genes and pathways

in colon tumor progression. Mol Cancer Res 2011;9:133–48.

Published OnlineFirst August 9, 2011.�2011 American Association for Cancer Research.doi: 10.1158/1541-7786.MCR-11-0345

MolecularCancer

Research

Mol Cancer Res; 9(9) September 20111266

2011;9:133-148. Published OnlineFirst January 12, 2011.Mol Cancer Res Mie Ichikawa, Roy Williams, Ling Wang, et al. ProgressionS100A8/A9 Activate Key Genes and Pathways in Colon Tumor

Updated version

10.1158/1541-7786.MCR-10-0394doi:

Access the most recent version of this article at:

Cited articles

http://mcr.aacrjournals.org/content/9/2/133.full#ref-list-1

This article cites 76 articles, 17 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/9/2/133.full#related-urls

This article has been cited by 19 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/9/2/133To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-10-0394

http://mcr.aacrjournals.org/content/9/2/133.full#ref-list-1

http://mcr.aacrjournals.org/content/9/2/133.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/9/2/133


Documents

S100A8/A9 Activate Key Genes and Pathways in Colon Tumor ... · kB–p65 binding activity using TransAM NF-kB assay kit (ActiveMotif) according to the manufacturer's instructions