The macrophage tetraspan MS4A4A enhances dectin-1-dependent NK cell–mediated resistance to metastasis

The plasma membrane tetraspan molecule MS4A4A is selectively expressed by macrophage-lineage cells, but its function is unknown. Here we report that MS4A4A was restricted to murine and human

mononuclear phagocytes and was induced during monocyte-to-macrophage differentiation in the presence of interleukin 4 or dexamethasone. Human MS4A4A was co-expressed with M2/M2-like

molecules in subsets of normal tissue-resident macrophages, infiltrating macrophages from inflamed synovium and tumor-associated macrophages. MS4A4A interacted and colocalized with the

β-glucan receptor dectin-1 in lipid rafts. In response to dectin-1 ligands, Ms4a4a-deficient macrophages showed defective signaling and defective production of effector molecules. In

experimental models of tumor progression and metastasis, Ms4a4a deficiency in macrophages had no impact on primary tumor growth, but was essential for dectin-1-mediated activation of

macrophages and natural killer (NK) cell–mediated metastasis control. Thus, MS4A4A is a tetraspan molecule selectively expressed in macrophages during differentiation and polarization,

essential for dectin-1-dependent activation of NK cell–mediated resistance to metastasis.

The data that support the findings of this study are available from the corresponding author upon reasonable request. Ms4a4afl/fl animals have been developed by the research team and are

available under a material transfer agreement after evaluation of potential conflict of interest.

The authors thank C. Garlanda and M. Kallikourdis (Humanitas Clinical and Research Center, Scientific Institute for Research and Healthcare (IRCCS)) for providing cDNAs of murine leukocyte

subsets, and T. Irimura (Juntendo University School of Medicine, Tokyo, Japan) and R. Giavazzi (Mario Negri Institute, Milan, Italy) for providing MC38 and SL4 cells. Technical assistance

from A. Fontanini, C. Perrucchini, T. Schorn, R. Porte and F. Pasqualini is acknowledged. A. Inforzato (Humanitas Clinical and Research Center, IRCCS), A. Diefenbach (Charité –

Universitätsmedizin Berlin, Germany) and L. Florin (University Medical Centre of the Johannes Gutenberg University, Mainz, Germany) are gratefully acknowledged for their support and

discussion. Financial support came from Fondazione Cariplo (grant no. 2015–0564 to A.M.), Cluster Alisei (grant no. MEDINTECH CTN01_00177_962865 to A.M.), the European Research Council

(grant no. 669415-PHII to A.M.), the Italian Association for Cancer Research (AIRC IG-2016 grant no. 19014 to A.M.; AIRC 5 × 1000 grant no. 21147 to A.M.; AIRC IG-2016 grant no. 19213 to

M.L.), Medical Research Council (Pathobiology of Early Arthritis Cohort grant no. 36661 to C.P.) and Arthritis Research UK Experimental Treatment Centre (grant no. 20022 to C.P.). I.M. was

supported by a Mario and Valeria Rindi fellowship and a Fellowship for abroad from the Italian Foundation for Cancer Research, and by a European Federation of Immunological Societies-IL

short-term fellowship. B.S. was supported by Ministero della Salute (progetto Finalizzata GR-2013-02356522). R.S.G. was supported by a PhD studentship (PD/BD/114138/2016) from Fundação para

a Ciência e Tecnologia, Portugal. S.K.B. was supported by core funding from Singapore Immunology Network—Agency for Science, Technology & Research (A*STAR), Singapore. F.T. was supported by

a fellowship from the A*STAR Research Attachment Program, Singapore. S.L. was supported by Fondazione Beretta, Italy.

Present address: Medical Faculty, Department of Physiology and Immunology, University of Rijeka, Rijeka, Croatia

Present address: Department of Pathology and Immunology, Washington University School of Medicine, St Louis, MO, USA

Present address: Faculty of Biomedical Sciences, Institute for Research in Biomedicine, Università della Svizzera italiana, Bellinzona, Switzerland

These authors contributed equally: Irene Mattiola, Federica Tomay, Maria De Pizzol.

Humanitas Clinical and Research Center, Scientific Institute for Research and Healthcare, Rozzano, Italy

Irene Mattiola, Federica Tomay, Maria De Pizzol, Rita Silva-Gomes, Benedetta Savino, Tamara Gulic, Andrea Doni, Roberta Carriero, Diego Morone, Marina Sironi, Barbara Bottazzi, Alberto

Mantovani & Massimo Locati

Laboratory of Innate Immunity, Department of Microbiology, Infectious Diseases and Immunology, Charité–Universitätsmedizin Berlin, Berlin, Germany

Mucosal and Developmental Immunology, Deutsches Rheuma-Forschungszentrum, Berlin, Germany

Department of Medical Biotechnologies and Translational Medicine, University of Milan, Milan, Italy

Irene Mattiola, Federica Tomay, Maria De Pizzol, Benedetta Savino, Tamara Gulic & Massimo Locati

Graduate Program in Areas of Basic and Applied Biology, Instituto de Ciências Biomédicas Abel Salazar, University of Porto, Porto, Portugal

Department of Molecular and Translational Medicine, School of Medicine, University of Brescia, Brescia, Italy

Centre for Experimental Medicine and Rheumatology, William Harvey Research Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, UK

Marie Astrid Boutet, Alessandra Nerviani, Costantino Pitzalis & Alberto Mantovani

Singapore Immunology Network, Agency for Science, Technology & Research, Singapore, Singapore

Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Endocrinology Unit, Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy

Department of Pathology and Immunology, Washington University School of Medicine, St Louis, MO, USA

I.M., F.T. and M.D.P. performed in vitro experiments. I.M., B.S. and F.T. performed in vivo experiments. R.S.-G., M.M. and M. Sironi provided support for in vivo and in vitro experiments.

A.D. conducted the imaging analysis. D.M. and M. Stravalaci provided support for the analysis of molecular interactions. R.C. performed bioinformatics analysis. I.N.S. and Y.L. performed

gene expression experiments. S.L., W.V., M.A.B., A.N. and T.G. performed histology. G.M. and C.P. provided access to patients’ samples. S.K.B., B.B., C.P., A.M. and M.L. contributed to the

experimental design and the supervision of the study.

A.M. is a recipient of commercial research grants from Novartis; a consultant/advisory board member for Novartis, Roche, Ventana, Pierre Fabre, Verily, AbbVie, Compugen, Macrophage

Therapeutics, AstraZeneca, Biovelocita, BG Fund, Third Rock and Verseau; an inventor of patents related to PTX3 and other innate immunity molecules; and also receives royalties for reagents

related to innate immunity.

Peer Review Information: Laurie Dempsey was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Representative dot plots and Mean Fluorescence Intensity (MFI) of the expression of MS4A4A (a, b), MS4A6A (c, d), and MS4A7 (e, f) in monocyte-derived macrophages (Mɸ) and dendritic cells

(DC) treated (Dex, open symbol) or not (NT, closed symbol) with 10−6 M Dex for 24 h. Ctrl represents isotype control staining. Results are shown as mean ± SEM. Three independent experiments

were performed for panels a to d (n = 5 donors) and four independent experiments were performed for panels e to h (n = 6 donors). Statistical analysis by one-way ANOVA.

MS4A4A transcript levels in human tissues (a) and cell types (b) based on the Laboratory for Systems Biology and Medicine database RefExA. EnC, endothelial cells, EpC, epithelial cells, SMC,

smooth muscle cells.

a) Boxplot of MS4A4A mRNA expression in human tumors (closed boxes) and corresponding normal tissues (open boxes) among different TCGA cancer datasets. b) Heatmap correlation between mRNAs

encoding MS4A4A and TAM markers in tumor tissues among different TCGA cancer datasets. Genes are represented according to their average Pearson correlation with MS4A4A among all cancer

datasets. Number of samples included in the analysis of both panels (normal tissue, corresponding tumor tissue) are: bladder urothelial cancer: 19, 408; breast cancer: 112, 1100; cervical

adenocarcinoma: 3, 306; cholangiocarcinoma: 9, 36; colon adenocarcinoma: 41, 287; colorectal adenocarcinoma: 51, 382; esophageal cancer: 11, 185; esophageal carcinoma: 46, 600; glioblastoma:

5, 166; glioma: 5, 696; head and neck cancer: 44, 522; kidney cancer: 25, 66; kidney clear cell cancer: 72, 534; kidney renal papillary cell cancer: 32, 291; liver cancer: 50, 373; lung

cancer: 59, 517; lung squamous cell cancer: 51, 501; pan-kidney cohort: 129, 891; pancreatic cancer: 4, 179; paraganglioma: 3, 184; prostate cancer: 52, 498; rectum cancer: 10, 95; sarcoma:

2, 263; stomach adenocarcinoma: 35, 415; thymoma: 2, 120; thyroid cancer: 59, 509; uterine cancer: 24, 177.

a) Ms4a4a mRNA levels in murine leukocyte subsets sorted form the spleen of wild-type animals (NK cells: CD45+/CD3-/NK1.1+; CD4+ T cells: CD45+/CD3+/CD4+; CD8+ T cells: CD45+/CD3+/CD8+; B

cells: CD45+/B220+; PMN: CD45+/CD11bhi/Gr1hi; mDC: CD45+/CD11b+/CD11c+; pDC: CD45+/CD11b-/CD11c+; macrophages: CD45+/CD11b+/F4/80hi). Results are shown as mean ± SEM of Ms4a4a relative

expression normalized to Gapdh (n = 3 mice). Statistical analysis by one-way ANOVA. b) MS4A4A mRNA levels in wild-type BMDMs exposed or not (M0) to 20 ng/ml IFN-γ plus 100 ng/ml LPS, 20 

ng/ml IL-4, 10−6 M Dex, or 20 ng/ml IL-4 plus 10−6 M Dex for 18 h. Results are shown as fold change over resting macrophages (M0; n = 5). Statistical analysis by one-way ANOVA. c, Ms4a4a

mRNA levels in BMDMs from wild-type (closed symbols) or Ms4a4a-/-(open symbols) mice treated (Dex) or not (NT) with 10−6 M Dex for 24 h. Results are shown as mean ± SEM of Ms4a4a relative

expression normalized to Gapdh (n = 3 mice). Statistical analysis by two-tailed unpaired (Mann-Whitney) Student’s t test and one-way ANOVA. d) Percentage of NK cells (CD45+/CD3-/NK1.1+), T

cells (CD45+/CD3+), B cells (CD45+/B220+), PMN (CD45+/CD11bhi/Gr1hi), mDC (CD45+/CD11b+/CD11c+), pDC (CD45+/CD11b-/CD11c+), and macrophages (CD45+/CD11b+/F4/80hi) in the spleen of wild-type

(closed symbols) or Ms4a4a-/- (open symbols) mice. Results are shown as mean ± SEM (n = 4). No statistical difference by two-tailed unpaired Student’s t test.

a, b) Flow cytometry analysis of Dectin-1 expression in monocyte-derived macrophages (Mɸ) and dendritic cells (DC) treated (Dex, open symbol in b) or not (NT, closed symbol in b) with 10−6 M

Dex for 24 h. Ctrl represents isotype control staining. Results are shown as mean ± SEM. Three independent experiments were performed (n = 5 donors). Statistical analysis by one-way ANOVA.

c, d) Flow cytometry analysis of Dectin-1 and MS4A4A expression in monocyte-derived macrophages (Mɸ) and dendritic cells (DC) treated (Dex, open symbol in d) or not (NT, closed symbol in d)

with 10−6 M Dex for 24 h. Ctrl represents isotype control staining. Percentage of double-positive cells is shown in d. Results are shown as mean ± SEM. Three independent experiments were

performed (n = 5 donors). Statistical analysis by one-way ANOVA.

a–c) Syk phosphorylation in BMDMs from Ms4a4a-/- (open symbol) and wild-type (closed symbol) mice primed for 18 h with 10 ng/ml GM-CSF and stimulated or not with 100 μg/ml depleted zymosan

(a), 100 μg/ml curdlan (b), or 100 ng/ml PMA (d). Results are shown as mean ± SEM of relative MFI (fold on untreated). Three independent experiments were performed. a) time 5: 4 wild-type

and 4 Ms4a4a-/- mice; time 15: 5 wild-type and 8 Ms4a4a-/- mice. b) time 5: 6 wild-type and 5 Ms4a4a-/- mice; time 15: 5 wild-type and 6 Ms4a4a-/- mice. c) time 5: 4 wild-type and 5

Ms4a4a-/- mice; time 15: 4 wild-type and 4 Ms4a4a-/- mice. Statistical analysis by two-tailed unpaired (Mann-Whitney) Student’s t test. d, e) Phosphorylation of ERK (d) and p38 (e) in BMDMs

from Ms4a4a-/- (open symbols) and wild-type (closed symbols) mice primed for 18 h with 10 ng/ml GM-CSF and stimulated or not (NT) with 100 μg/ml depleted zymosan, 100 μg/ml curdlan, 100 

ng/ml LPS, and 100 ng/ml PMA. Results are shown as mean ± SEM of relative MFI (fold on untreated). Two independent experiments were performed. One dot represents one mouse (2–4 for e, 2–6

for f). f–k) Secretion of IL-6 (f–h) and TNF (i–k) by BMDMs from Ms4a4a-/- (open symbol) and wild-type (closed symbol) mice primed for 18 h with 10 ng/ml GM-CSF and stimulated for 24 h (f, g

and i, j) or 6 h (h–k) with 100 μg/ml depleted zymosan (f–i), 100 μg/ml curdlan (g–j) or 100 ng/ml Pam3Cys (h–k). Cytokine levels in untreated cells were below detection limit. Results are

shown as mean ± SEM. Two independent experiments for f and i (9 wild-type and 11 Ms4a4a-/- mice for f; 8 wild-type and 12 Ms4a4a-/- mice for i); five independent experiments for g and j (14

wild-type and 16 Ms4a4a-/- mice for g; 13 wild-type and 16 Ms4a4a-/- mice for j), three independent experiments for h and k (7 wild-type and 10 Ms4a4a-/- mice). Statistical analysis by

two-tailed unpaired (Mann-Whitney) Student’s t test.

a) Representative images of lungs from Ms4a4a-/- and wild-type mice upon i.v. injection of B16F1 or B16F10 cells. b) Number of metastatic foci in lungs from Ms4a4a-/-Clec7a-/- (open symbol)

and wild-type (closed symbol) mice upon B16F1 i.v. injection. Results are shown as mean ± SEM. Four independent experiments were performed (11 wild-type and 19 Ms4a4a-/-Clec7a-/-mice).

Statistical analysis by two-tailed unpaired (Mann-Whitney) Student’s t test. c) Representative immunofluorescence images of the distribution of NK cells (NKp46+), macrophages (F4/80+), and

proliferating tumor cells (Ki-67+) in the lung of wild-type mice injected i.v. with B16F1 cells. Nuclei were counterstained with DAPI and visualized in blue. Images refer to one experiment

out of two analyzed. Scale bars: 150 μm in the confocal images (left panels), 15 μm in the 3D bland images of representative areas (right panels). d–g) Frequency (d, f) and numbers (e, g) of

interstitial (d, e) and alveolar (f, g) macrophages in the lungs of Ms4a4a-/- (open symbol) and wild-type (closed symbol) mice injected i.v. with B16F1 cells. Resuls are shown as percentage

or absolute numbers (n) of F4/80+/CD11c- (interstitial) or F4/80+/CD11c+ (alveolar) macrophages pregated on Live/CD45+/SSC-Ahi/CD11blo/hi cells that were Ly6C and Ly6G negative. Results are

expressed as mean ± SEM. Three independent experiments were performed (9 wild-type and 6 Ms4a4a-/- mice). Statistical analysis by two-tailed unpaired (Mann-Whitney) Student’s t test. h) NK

cell degranulation towards B16F10 cells. Results are shown as mean ± SEM of percentage of CD107a+ cells gated on Live/CD45+/CD3-/NK1.1+ splenocytes. Two independent experiments were

performed (5 wild-type and 4 Ms4a4a-/- mice). Statistical analysis by two tailed unpaired (Mann-Whitney), Student’s t test. i–k) Calhm6 mRNA levels in Ms4a4a-/- (open symbol) and wild-type

(closed symbol) BMDMs primed for 18 h with 10 ng/ml GM-CSF and stimulated for 24 h with zymosan (i), depleted zymosan (j), or LPS (k). Results are shown as mean ± SEM of fold of induction.

Two independent experiments were performed for i (3 wild-type and 5 Ms4a4a-/- mice), j (7 wild-type and 9 Ms4a4a-/- mice), and k (10 wild-type and 5 Ms4a4a-/- mice). Statistical analysis by

two-tailed unpaired (Mann-Whitney) Student’s t test. l) IL-18 secretion by Ms4a4a-/- (open symbol) and wild-type (closed symbol) BMDMs primed for 18 h with 10 ng/ml GM-CSF and stimulated for

24 h with zymosan. Resuts are shown as mean ± SEM. Two independent experiments were performed (6 wild-type and 5 Ms4a4a-/- mice). Statistical analysis by two-tailed unpaired (Mann-Whitney),

Student’s t test. IL-18 levels in untreated cells and IL-12p70 levels in untreated and stimulated cells were below detection limit and are not shown.

a) Representative dot plot (left) and Mean Fluorescence Intensity (right) of WGA binding on the surface of MC38 and SL4 cells assessed by flow cytometry. Results are shown as mean ± SEM of

relative MFI (fold on unstained). Three independent experiments were performed (5 independent samples for SL4, 7 independent samples for MC38). Statistical analysis by two-tailed unpaired

(Mann-Whitney) Student’s t test. b, c) Number of metastatic foci in liver (b) and lungs (c) of Clec7a-/- (open symbol) and wild-type (closed symbol) mice upon intrasplenic injection of MC38

(b) or i.v. injection of SL4 (c) cells. Results are shown as mean ± SEM. One experiment was performed (5 wild-type and 6 Ms4a4a-/- mice). Statistical analysis by two-tailed unpaired

(Mann-Whitney) Student’s t test. d) Number of metastatic foci in liver of Ms4a4a-/- (open symbol) and wild-type (closed symbol) mice injected intrasplenic with MC38 cells upon NK cell

depletion. Results are shown as mean ± SEM. Two independent experiments were performed (5 wild-type and 9 Ms4a4a-/- mice). Statistical analysis by two-tailed unpaired (Mann-Whitney)

Student’s t test and one-way ANOVA. e) Number of metastatic foci in lungs from Ms4a4a-/- (open symbol) and wild-type (closed symbol) mice upon i.v. injection of SL4 cells. Results are shown

as mean ± SEM. One experiment was performed (6 mice). Statistical analysis by two-tailed unpaired (Mann-Whitney) Student’s t test.

Gating strategies applied to Figures and Supplementary Figures are reported.

Anyone you share the following link with will be able to read this content: