Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

The tumour microenvironment creates a niche for the self-renewal of tumour-promoting macrophages in colon adenoma

The tumour microenvironment creates a niche for the self-renewal of tumour-promoting macrophages... ARTICLE DOI: 10.1038/s41467-018-02834-8 OPEN The tumour microenvironment creates a niche for the self-renewal of tumour-promoting macrophages in colon adenoma 1 1 1 2 2 2 2 Irene Soncin , Jianpeng Sheng , Qi Chen , Shihui Foo , Kaibo Duan , Josephine Lum , Michael Poidinger , 2,3 1 1 Francesca Zolezzi , Klaus Karjalainen & Christiane Ruedl Circulating CCR2 monocytes are crucial for maintaining the adult tissue-resident hi hi F4/80 MHCII macrophage pool in the intestinal lamina propria. Here we show that hi low a subpopulation of CCR2-independent F4/80 MHCII macrophages, which are the hi most abundant F4/80 cells in neonates, gradually decline in number in adulthood; these hi macrophages likely represent the fetal contribution to F4/80 cells. In colon adenomas of Min/+ hi low Apc mice, F4/80 MHCII macrophages are not only preserved, but become the dominant subpopulation among tumour-resident macrophages during tumour progression. hi low hi hi Furthermore, these pro-tumoural F4/80 MHCII and F4/80 MHCII macrophages can self-renew in the tumour and maintain their numbers mostly independent from bone marrow contribution. Analyses of colon adenomas indicate that CSF1 may be a key facilitator of macrophage self-renewal. In summary, the tumour microenvironment creates an isolated niche for tissue-resident macrophages that favours macrophage survival and self-renewal. 1 2 School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore. Singapore Immunology Network, Agency for Science, Technology and Research (A*STAR), 8A Biomedical Grove, Singapore 138648, Singapore. Present address: GALDERMA R&D, 06902 Sophia Antipolis Cedex France. Klaus Karjalainen and Christiane Ruedl jointly supervised this work. Correspondence and requests for materials should be addressed to K.K. (email: Klaus@ntu.edu.sg) or to C.R. (email: Ruedl@ntu.edu.sg) NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 Min/+ 5 acrophages comprise a heterogeneous population of treated Apc mice (Fig. 1). Similar as previously reported , tissue-resident immune cells that contribute to tissue colon LP CD45 cells comprised three major myeloid cell hi hi Mhomeostasis, support the host defence system and can fractions (I–III) after excluding CD11c MHCII dendritic hi + impact on the initiation and propagation of several diseases, cells: F4/80 CD11b tissue-resident (tumour-resident in 1,2 int + including cancer (reviewed in refs. ). The majority of tissue- polyps) macrophages (fraction I), F4/80 CD11b (fraction II) 3–5 low + resident macrophages are established prenatally and are mainly and F4/80 CD11b neutrophils (fraction III) (Fig. 1a). The independent from any further haematopoietic input due to their majority (>85%) of fraction I cells expressed high levels of 6–10 capacity to self-renew in situ to maintain population size . For major histocompatibility complex (MHC) class II and only a example, microglia, the resident macrophage population of the small proportion (~5%) of cells were low for MHC class II central nervous system, develop early during embryonic devel- expression. Fraction II could be further subdivided into four opment from yolk sac precursors and effectively self-maintain distinct subpopulations in a waterfall-shaped distribution on a 5,6,11 in situ throughout adulthood . Although the majority of Ly6C vs. MHCII dot plot consisting of three distinct monocyte- hi − hi + the tissue-resident macrophages are maintained independently differentiation stages (P1: Ly6C MHCII ;P2: Ly6C MHCII ; − + − from bone marrow (BM)-derived monocytes, numerous studies P3: Ly6C MHCII ) and one eosinophil fraction (Ly6C MHCII have demonstrated that macrophages of certain tissues, including )(Fig. 1a). We then analysed the presence of these sub- 5,12 13 14 15 Min/+ the dermis , mammary gland , heart , pancreas and populations in DSS-accelerated colon tumours from Apc 5,16,17 intestine , require the input of monocytes to retain their pool mice. The same LP populations were also identified in colon during adulthood. In particular, the intestinal macrophages polyps but with markedly different frequencies (Fig. 1b). In depend on the constant and fast replenishment of circulating these tumours, neutrophils (fraction III) were the most abun- hi blood Ly6C monocytes, not only during inflammation but also dant myeloid cell type (>40%) followed by monocyte-derived 16–18 under normal healthy conditions . In a healthy unperturbed cells (fraction II) and tumour-resident macrophages (fraction colon, monocytes attracted to the intestine gradually differentiate I). Interestingly, eosinophils in tumours represented only a into tissue-resident macrophages by losing Ly6C expression, up- minor fraction (~3%) of the total CD11b cells. Furthermore, regulating the expression of macrophage markers, such as tumour-resident macrophages (fraction I) were clearly enriched 16,19 low CX3CR1, F4/80, CD64 and CD11c , and secreting or in MHCII cells (Fig. 1a,b), which were almost neglectable in 20,21 responding to the anti-inflammatory cytokine interleukin-10 . normal adult colon LP. Mice lacking either the chemokine receptor CCR2 or its ligand Characterization of myeloid subpopulations in the colon LP of CCL2 have reduced numbers of intestinal macrophages , thus untreated C57BL/6J mice and in “spontaneously” formed tumours Min/+ implying that the homeostatic recruitment of macrophage pre- obtained from 5-month-old Apc mice confirmed the cursors to the gut lamina propria (LP) may be dependent on a presence of all myeloid subpopulations described in Fig. 1, CCR2–CCL2 axis. Upon intestinal inflammation, tissue-resident although with a clear increase in the proportion of the tumoural + − + macrophages are still derived from circulating monocytes, but CD11b F4/80 neutrophil fraction (up to 70% of total CD11b convert from being anti-inflammatory macrophages to highly cells) (Supplementary Fig. 1). Toll-like receptor-responsive inflammatory cells . Tissue-resident macrophages are also naturally present in low various tumours, including lung and mammary tumours where MHCII macrophages accumulate during tumour progres- 22,23 24 hi macrophages are monocyte-derived or even self-renewing . sion. A previous study reported that F4/80 tissue-resident The developmental origins and the maintenance kinetics of macrophages in the colon LP progressively disappear with resident macrophages in intestinal tumours, however, have not increasing age . We therefore analysed the myeloid cell subsets been studied thus far. present in the fetal colon (embryonic day 19.5) and compared Here we show that the colon LP of normal adult mice contains them to those present in the colon LP of young and old mice hi hi 17 a major population of F4/80 MHCII macrophages and a (from 1 week to 12 months). In contrast to Bain et al., our hi low population of gradually disappearing F4/80 MHCII cells. Both analysis clearly found that all three macrophage fractions (I–III) hi F4/80 macrophage populations are relatively slowly replenished were maintained across all age groups with only minor fluctua- low by BM-derived cells, particularly MHCII cells, compared to tions detected in their frequency (Fig. 2a). The predominant int F4/80 monocyte/macrophage cells that have a fast turnover. tissue-resident macrophage subset in the fetal colon was the F4/ hi low The tumour microenvironment, however, is enriched with F4/ 80 MHCII fraction. However, as a function of age this mac- hi low hi hi 80 MHCII macrophages. Furthermore, while F4/80 MH- rophage subset almost disappeared to leave mainly F4/80 MH- low + hi CII macrophages do not require CCR2 monocytes for their CII cells to represent the colon tissue-resident macrophages low maintenance in the healthy gut LP or in tumours, we found that (fraction I) (Fig. 2b). This decrease in MHCII cell representa- hi hi + F4/80 MHCII macrophages become independent from CCR2 tion suggests that the LP environment gradually changes with age, monocytes in tumours only, potentially due to tumour-derived perhaps due to the establishment of a mature microbiota and/or a colony stimulating factor-1 (CSF1) that supports macrophage tonic increase in inflammatory mediators . hi low self-renewal. Moreover, we demonstrate that CCR2-independent Given the high frequency of F4/80 MHCII cells in colon hi intratumoural F4/80 macrophages, and not the CCR2- polyps (Fig. 1b), we assessed whether there was an association low dependent monocytes or monocyte-derived macrophages, sup- between tumour progression and the frequency of MHCII hi port cancer progression, suggesting this particular F4/80 cell macrophages. Therefore, for this analysis we opted for the Min/+ fraction as an attractive therapeutic target. spontaneous tumour model of Apc mice, which generate progressively adenomas up to 5–6 mm in size that is rarely achieved in the DSS-accelerated tumour model. Indeed, we Results observed an increasing abundance of this cell fraction in low Skewed representation of myeloid cells in colon adenomas.We conjunction with tumour progression until MHCII macro- first profiled colon LP myeloid subpopulations from wild-type phages became a dominant cell type among tumour-resident (WT) C57BL/6J mice, previously treated with dextran sodium macrophages (Fig. 2c). This observation clearly suggests that the sulphate (DSS), and compared them with their tumoural tumour microenvironment favours the differentiation or main- low counterparts obtained from the polyps collected from DSS- tenance of MHCII macrophages. 2 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE ab c Fraction Fraction Fraction Colon LP 4 I II III 3 685 5 Colon LP Fr I 4 30 Fr II 3 4 5 0 10 10 10 P1 3 P2 Fr III 10 5 2 eos 3 4 5 0 10 10 10 P3 low CD11b MHCII 3 4 5 high 0 10 10 10 MHCII MHCII P1 P2 P3 10 Fraction Fraction Fraction Eosinophils I II III Neutrophils Colon tumour Fr I Colon tumour Fr II 3 4 5 0 10 10 10 10 10 P1 P2 Fr III P3 2 3 4 5 2 eos 0 10 10 10 10 CD11b 3 4 5 0 10 10 10 MHCII Min/+ Fig. 1 Myeloid cell heterogeneity in colon lamina propria and colon tumours. WT C57BL/6J and Apc mice aged 6 weeks old were administered drinking water supplemented with dextran sodium sulphate for 7 days, and analysed 4 weeks later. a Flow cytometry representative dot plots of colon lamina propria (LP) (upper panel) and tumour cell subpopulations (lower panel). Three different myeloid fractions (I–III) are defined by the differential hi hi low expression of F4/80. Fraction I represents F4/80 tissue-resident macrophages, which can be further subdivided into MHCII and MHCII . Fraction II contains monocytes (P1), two monocyte-derived macrophage subpopulations (P2 and P3) and eosinophils (eos), based on differential expression of MHCII and Ly6C. Fr III consists of neutrophils. Gating strategy is shown in Supplementary Fig. 4. b Bar charts of the distinct myeloid cell subpopulations obtained from the colon LP (n = 8) (upper bar chart) and 2–3 mm colon tumours (n = 20) (lower bar chart). White bars: fraction I; light blue bars: fraction II and hi hi low black bars: fraction III. Error bars represent the s.e.m. c Pie charts show the proportions of F4/80 tissue-resident macrophages (MHCII and MHCII ), monocytes (P1), monocyte-derived macrophages (P2–P3), neutrophils and eosinophils across colon LP and tumours Intratumoural macrophages are independent from CCR2 Taken together, these data assert that the tumour microenvir- hi hi monocytes. Given that circulating Ly6C monocytes express onment can modify the properties of MHCII tissue-resident high levels of CCR2 receptor , we decided to examine the pre- macrophages such that they also become CCR2 independent like low sence of this receptor on tissue-resident and monocyte-derived MHCII cells in the LP and hence are presumably able to macrophages in colon LP and adenoma polyps. The aim of this maintain themselves without monocyte replenishment. experiment was to determine the potential origins of these cells. In the LP, consistent with being descendants of classical mono- hi − hi hi cytes, Ly6C MHCII (P1), Ly6C MHCII (P2) and Ly6C Turnover of colon lamina propria myeloid cells. We utilized the − + MerCreMer/R26 5 MHCII (P3) cells in fraction II subset had the highest and most Kit fate mapping mouse —where yellow florescence homogeneous expression of CCR2 (Fig. 3a) , whereas tissue- protein (YFP) expression can be induced in early BM progenitors hi hi hi low resident F4/80 MHCII and F4/80 MHCII macrophages —to monitor the cell population turnover rates driven by the BM were highly heterogeneous in their CCR2 expression levels, from input in colon LP myeloid cell populations. After injecting adult very high to very low levels (Fig. 3a and Supplementary Fig. 2). As mice with tamoxifen, the labelling index of tissue-resident mac- hi low hi expected, numbers of monocytes and monocyte-derived cells (P1, rophages (F4/80 MHCII and MHCII ), monocytes (P1), −/− P2 and P3) in fraction II were severely diminished in Ccr2 monocyte-derived macrophages (P2–P3 fractions), eosinophils hi mice (Fig. 3b). In addition, MHCII tissue-resident macrophages and neutrophils was analysed at different time points over a low were also mostly CCR2 dependent, whereas MHCII cells were period of 5 months. As anticipated, all tested myeloid popula- not, which is likely a reflection of their markedly lower CCR2 tions, with the exception of tissue-resident macrophages, were expression level (Fig. 3b). rapidly labelled with YFP and reached a plateau by 1–2 weeks hi Interestingly, we found that tumour-resident MHCII and after the final tamoxifen injection. These data suggest rapid low MHCII macrophages both expressed relatively low levels of replacement of myeloid cell populations by BM-derived cells CCR2 compared to their LP counterparts and, consistently, their (Fig. 4). −/− hi hi numbers did not change in Ccr2 mice. Conversely, fraction II Although it is widely considered that gut F4/80 MHCII cells cells (P1, P2 and P3) in polyps behaved exactly like their WT lose their prenatal origin and are replaced by CCR2 monocytes counterparts (Fig. 3a, c). in adulthood , our data clearly demonstrated that they are in fact NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 3 | | | low MHCII high MHCII P1 P2 P3 Eosinophils Neutrophils low MHCII high MHCII P1 P2 P3 Eosinophils Neutrophils F4/80 F4/80 Ly6C Ly6C % Of CD11b myeloid cells + % Of CD11b myeloid cells ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 a Fraction I 3 Weeks 4 Weeks 6 Weeks 8 Weeks 12 months 5 5 5 5 5 10 10 10 10 10 I 2 2 1.5 3 4 4 4 4 4 10 II 10 10 10 4 7 10 11 3 5 3 3 3 12 10 10 10 10 10 0.5 0.3 0 0.5 0 0.3 0 0 0.3 III 3 4 5 3 4 5 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 CD11b E19.5 2 Days 1 Week 3 Weeks 5 5 5 5 10 10 10 10 4 4 4 4 10 10 10 10 low 80 MHCII 3 3 3 3 10 10 49 50 10 10 93 6 47 51 high MHCII 2 2 2 10 10 10 0 0 3 4 5 3 4 5 3 4 5 3 4 5 40 10 10 10 01 10 10 10 0 10 100 0 10 10 10 0 4 Weeks 6 Weeks 8 Weeks 12 Months 5 5 5 10 10 10 4 4 4 10 10 10 2 96 3 18 79 3 3 3 2 10 10 13 10 10 0 0 0 0 Age 3 4 5 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 1001 10 10 10 0 10 100 MHCII 0.5 mm 1 mm 2 mm 3–4 mm 5–6 mm low MHCII 5 5 5 5 60 5 high 10 10 10 10 MHCII 4 4 4 4 4 40 10 10 10 10 73 63 62 36 21 32 45 19 70 49 3 3 3 3 10 10 3 10 10 10 20 2 2 2 2 10 10 10 10 0 0 0 0 0 3 4 5 3 4 5 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 MHCII Tumour diameter hi hi hi low hi Fig. 2 Ageing and tumour progression influence the ratio between F4/80 MHCII and F4/80 MHCII subpopulations. a F4/80 tissue-resident macrophages persist in the colon LP after birth into adulthood, as shown by the representative flow cytometry dot plots and bar chart. Error bars represent hi low hi hi hi low the s.e.m. b The frequency of colon F4/80 MHCII cells rapidly declines after birth. Representative dot plots of F4/80 MHCII and F4/80 MHCII subpopulations obtained from fetal colon (E19.5) and from colons of mice aged 2 days, 1, 3, 4, 6 and 8 weeks and 12 months. The bar chart represents the hi hi hi low age-dependent ratio between F4/80 MHCII (blue) and F4/80 MHCII (red) subpopulations in the colon LP. E19.5: n = 1 (pool of 15 embryos obtained from 3 different pregnant mice); day 2: n = 2 (each group pool of 5 mice); 1 week: n = 2 (each group pool of 5 mice); 3 weeks: n = 6 mice; 4 weeks: n = 5 mice; 6 weeks: n = 4; 8 weeks: n = 7 and 12 months: n = 4 mice. Error bars represent the s.e.m. c Representative flow cytometry analysis (left panel) and hi hi hi low mean percentage of F4/80 MHCII and F4/80 MHCII subpopulations (right panel) in tumours of different sizes (0.5–6.0 mm in diameter) obtained Min/+ from Apc mice. 0.5 mm: n = 3; 1 mm: n = 3; 2 mm: n = 4; 3–4 mm: n = 3; 5–6 mm: n = 4. Error bars represent the s.e.m. Scale bars: 2.5 mm. Gating strategy is shown in Supplementary Fig. 4 replaced by BM-derived monocytes only very slowly, if at all. Our “frozen” to the level corresponding to the time for their entrap- fate mapping data indicated that it took several months to fully ment in this niche that became independent from new inputs hi hi low replace F4/80 MHCII cells, whereas MHCII macrophages (Fig. 4 and Fig. 5a). exhibited minimal labelling (Fig. 4), suggesting that these tissue- Of note, when tamoxifen was administered during embryogen- resident macrophages retained their original fetal seed population esis (Fig. 5b, upper panel), all myeloid cell subpopulations— for a long period. including tissue-resident macrophages in the healthy LP and colon tumours—exhibited similar YFP tagging patterns (Fig. 5b, lower panel). This result was expected, as all of these cells are the progeny of classical haematopoietic stem cells, as previously Intratumoural resident macrophages expand by self-renewal. hi low described . We noted that F4/80 MHCII tumour-resident macrophages We found clear evidence, however, for increased self-renewal in particular seemed to expand in polyps and thus wanted to hi of F4/80 macrophages that could explain their capacity to determine if this effect was due to increased self-renewal or the expand in polyps. Firstly, tumour-resident macrophages were recruitment of new cells. Again, our adult fate mapping analyses MerCreMer/R26 Min/+ clearly in an active phase of the cell cycle unlike those in the LP, in Kit Apc mice showed that the recruitment of as indicated by the expression of the Ki-67 cellproliferation new cells to DSS-accelerated adenomas was not increased and hi low marker (Fig. 6a). Secondly, RNA-sequencing (RNA-seq) that the population turnover rate of F4/80 MHCII cells was hi analyses of cell-sorted, purified tumour-resident F4/80 (both comparable to that of the LP (Fig. 5a). A decrease in YFP labelling hi low hi hi MHCII and MHCII ) cells showed a clear upregulation of was observed in tumour-resident F4/80 MHCII cells when several cell cycle regulators, including cyclin-dependent kinase compared to their LP counterparts. This is probably due to the Cdk1, cyclin-A2 and -B2, E2f2, NEK family of serine/threonine fact that the labelling index of macrophages in polyps was 4 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | 3 Weeks 4 Weeks 6 Weeks 8 Weeks 12 Months E19.5 2 Days 1 Week 3 Weeks 4 Weeks 6 Weeks 8 Weeks 12 Months 0.5 mm 1 mm 2 mm 3–4 mm 5–6 mm Ly6C F4/80 Ly6C Subset (%) Subset (%) % Of CD11b myeloid cells NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE Lamina propria Tumour low MHCII high MHCII P1 P2 P3 Eosinophils 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 CCR2 b Fraction II Fraction I low high Eosinophils MHCII MHCII 5 5 5 0.6 10 10 10 ns ns 6000 *** 4 4 300 4 20 10 10 10 10,000 II 3 3 200 10 10 10 2 2 0.5 100 10 10 0 0 0 0 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 P1 P2 P3 5 5 5 10 0.2 10 *** I *** 2500 *** 3000 1.8 4 4 4 6000 10 0.8 8 2000 22 2000 II 3 3 1500 3 10 0.4 0 28 500 0 0 0 0 0 3 4 5 3 4 5 3 4 5 0 10 10 1001 10 10 10 0 10 100 CD11b MHCII low high Eosinophils MHCII MHCII 5 5 10 10 10 *** ns ns 8000 4000 4 4 4 I 17 10 10 10 9000 11 6000 3000 II 3 3 3 10 10 10 4000 2000 2 2 9 10 10 3000 2000 1000 0 0 0 0 0 3 4 5 3 4 5 3 4 5 01 10 10 10 01 10 100 0 10 100 P1 P2 P3 5 5 5 10 10 5 ** 25,000 *** *** 4 4 4 1 20,000 10 10 10 II 23 73 15,000 6000 3 3 3 10 10 10,000 48 1000 2 2 3000 10 10 10 0.5 0 76 0 0 0 0 0 2 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 10 MHCII CD11b hi + Fig. 3 Lamina propria and intratumoural colon F4/80 CD11b macrophage subsets show differential CCR2-driven monocyte dependence. a Fluorescence- activated cell sorting histograms showing CCR2 expression profiles on distinct colon LP and intratumoural myeloid cell subpopulations: fraction I consists low hi of MHCII and MHCII cells, and fraction II consists of monocytes (P1) and monocyte-derived macrophages (P2 and P3). b Myeloid cell profiling in WT −/− and Ccr2 colon LP. Representative flow cytometry analysis with F4/80 and CD11b-expressing myeloid subpopulations, obtained from the colon LP of −/− WT and Ccr2 mice. Fractions I and II were further dissected for MHCII and Ly6C expression (left panel). The absolute numbers of eosinophils, F4/ hi hi hi low 80 MHCII and F4/80 MHCII tissue-resident macrophages and P1–P3 subpopulations are shown (right panel). The bar chart represents the mean −/− −/− number of mice in each group and the error bars represent the s.e.m. (WT; n = 7 and Ccr2 ; n = 14). c Myeloid cell profiling in WT and Ccr2 colon hi hi hi low tumours as described above for LP cells. Bar charts show the mean±s.e.m. of absolute numbers of eosinophils, F4/80 MHCII and F4/80 MHCII Min/+ Min/+ −/− tissue-resident macrophages and P1–P3 subpopulations obtained from Apc (n = 8) and Apc Ccr2 (n = 9) mice. Statistical significance was determined using an unpaired Student's t-test. **P<0.001; ***P < 0.0001; ns, not significant. Gating strategy is shown in Supplementary Fig. 4 hi low kinases and various cell division cycle and mini-chromosome because F4/80 MHCII macrophages gradually disappear in maintenance proteins family members, in comparison to their the adult LP and probably do not vigorously self-renew, and hi hi LP counterparts (Fig. 6b). Interestingly, we could not find any that F4/80 MHCII cells are predominantly maintained in the hi hi evidence for active cell cycling of F4/80 MHCII and F4/ LP by recruited BM-derived cells and, therefore, do not depend hi low 80 MHCII cells in the LP (Fig. 6a, b). This effect could be on in situ self-renewal. NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 5 | | | WT –/– Ccr2 WT –/– Ccr2 WT –/– Ccr2 Min/+ Apc –/– Ccr2 Min/+ Apc Min/+ Apc –/– Ccr2 Min/+ Apc Min/+ Apc –/– Ccr2 Min/+ Apc –/– Min/+ Min/+ Ccr2 Apc Apc –/– Ccr2 WT F4/80 F4/80 Fraction II Fraction I Ly6C Ly6C Absolute no. Absolute no. Absolute no. Absolute no. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 1 Week 2 Weeks 3 Weeks Analysis of YFP cells 4 Weeks Tamoxifen MerCreMer/R26 7 Weeks Kit 20 Weeks Tracing of BM-precursors b low high Neutrophils MHCII MHCII P1 P2 P3 Eosinophils 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 4 10 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 22.9 0.3 10 0 4.5 0 0 0 0 0 0 24.9 21.9 12.6 11.7 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 4 10 10 10 10 10 10 10 3 10 3 3 3 3 10 10 10 10 10 10 14.3 0 0.8 5.5 0 0 0 0 0 13.5 12.8 14.9 3 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 4 5 –10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 0 10 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 4 10 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 3.3 2 0.4 7.6 0 0 0 0 0 0 6.5 7.1 8.4 0 6.5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 1.3 4.1 5.7 0 0 0 0 0 0 5.3 5.2 4.8 3 3 3 3 3 3 5.8 –10 –10 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 10 10 10 10 10 10 3 3 3 3 3 3 3 10 10 10 10 10 10 10 1.9 7.8 12.3 0 0 0 0 0 0 0 10.7 11.8 11.3 10.9 3 3 3 3 3 3 3 –10 –10 –10 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 2.5 2 13.8 10 0 0 12.9 0 0 0 0 0 13.6 13.7 3 3 3 3 3 13 3 14.1 –10 –10 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 YFP YFP 1 Week 2 Weeks 3 Weeks 4 Weeks 7 Weeks 20 Weeks low high Neutrophils MHCII MHCII P1 P2 P3 Eosinophils Fig. 4 Intestinal tissue-resident macrophages exhibit a slower, gradual replacement by bone marrow-derived cells compared to other gut myeloid cells. a MerCreMer/R26 Schematic representation of the adult fate mapping protocol using Kit mice. Mice aged 6 weeks were injected with tamoxifen five times and groups of 4–8 animals were sacrificed 1, 2, 3, 4, 7 and 20 weeks later. b Representative flow cytometry analysis indicating the labelling efficiency of distinct low hi colon myeloid cell populations as defined in Fig. 1: MHCII -expressing and MHCII -expressing tissue-resident macrophages, monocytes and monocytes- derived macrophages (P1–P3), eosinophils and neutrophils. Neutrophils acted as internal controls for labelling efficiency and the tracings are from the same mouse. Gating strategy is shown in Supplementary Fig. 4. c The bar chart represents the mean percentage of yellow florescence protein-positive (YFP ) cells after normalization to the percentage of YFP neutrophils. The error bars represent the s.e.m. 26,27 low Tumour microenvironment alters resident macrophage phe- tumour-associated macrophages (TAMs) , with MHCII notype. Besides the upregulation of genes involved in cell pro- macrophages being the major producers of ARG1 within tumours liferation, macrophages in polyps clearly adopted a distinct (Fig. 6c–e). The expression of other urea cell cycle genes, such as metabolic signature. Similar to cancer cells, which can alter their Arginase-2 (ARG2) and Arginosuccinate synthetase 1 (AAS1), metabolism in aerobic glycolysis and production of lactate (so- was increased in tumour-resident macrophages as well (Fig. 6c). called Warburg effect) macrophages in polyps increased the Moreover, several transcripts of matrix metalloproteases (MMPs), transcript levels of key glycolytic genes, such as ENO1, GAPDH, such as MMP2, MMP9 and MMP12, involved in the degradation hi low TPI1, PGAM and LDHA, with the F4/80 MHCII macro- of extracellular matrix and promotion of metastasis were upre- phages expressing the highest levels (Fig. 6c). In addition, our gulated in both subpopulations of tissue-resident macrophages RNA-seq found that both tissue-resident subsets displayed when compared to their LP counterparts (Fig. 6f), suggesting a enhanced transcripts of ARG1, typical for M2-polarized and possible contribution of these cells in tumour progression. 6 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | YFP after normalization (%) 5 Months 7 Weeks 4 Weeks 3 Weeks 2 Weeks 1 Week CD11b F4/80 NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE ab MerCreMer/R26 Kit Tamoxifen E10.5 DSS Tamoxifen DSS MerCreMer/R26 Min/+ Kit 8 Weeks old Apc 1 Week 4 Weeks 6–7 Weeks 1 week 4 Weeks Min/+ 10 Weeks Apc Analysis of Analysis of Tumour formation Tracing of BM-precursors Tracing of BM-precursors Tumour formation + YFP cells YFP cells high low Neutrophils MHCII MHCII high low Neutrophils MHCII MHCII 5 5 5 10 10 10 5 5 5 10 10 10 4 4 4 10 10 10 4 4 4 10 10 10 3 3 3 2 10 10 7 10 3 3 3 10 10 69 10 65 0 0 0 65 0 0 3 3 3 –10 –10 –10 3 3 3 4 5 3 4 5 3 4 5 –10 –10 –10 0 10 10 10 0 10 10 10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 Eosinophils –10 10 10 10 P1 P2 P3 –10 0 10 10 10 –10 0 10 10 10 0 Eosinophils P1 P2 P3 5 5 5 5 10 10 10 10 5 5 5 5 10 10 10 10 4 4 4 4 10 10 10 10 4 4 4 4 10 10 10 10 3 3 3 3 10 10 10 10 10 10 3 3 3 3 11 64 69 10 63 10 10 10 0 0 0 0 0 0 0 3 3 3 3 –10 –10 –10 –10 3 3 3 3 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 –10 –10 –10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 YFP YFP high low high low Neutrophils MHCII MHCII Neutrophils MHCII MHCII 5 5 10 10 5 5 5 10 10 10 4 4 4 10 10 10 4 4 4 10 10 10 3 3 3 3 3 10 10 4 10 10 10 61 62 0 0 0 0 0 3 3 0 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 Eosinophils P1 P2 P3 Eosinophils P1 P2 P3 5 5 5 10 10 10 5 5 5 5 10 10 10 10 4 4 4 4 4 4 4 4 10 10 10 10 10 10 10 10 3 3 3 10 3 3 3 3 10 10 10 10 10 10 10 14 15 62 15 62 62 0 0 0 0 0 0 0 3 3 3 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 YFP YFP *** 80 *** 100 LP *** Tumour 0 0 Fig. 5 Identification of self-renewing colon macrophage subsets in healthy and tumour tissue. a Dextran sodium sulphate (DSS)-induced intestinal MerCreMer/R26 MerCreMer/R26 Min/+ adenoma formation was combined with adult fate mapping using the Kit and Kit Apc mice (schematic representation shown in upper panel). Representative flow cytometry analysis illustrates the colon myeloid cell subset labelling efficiency. Gating strategy is shown in Supplementary Fig. 4. The bar chart represents the mean percentage of yellow fluorescence protein-positive (YFP ) cells after normalization to the + MerCreMer/R26 Min/+ MerCreMer/R26 percentage of YFP neutrophils±s.e.m. A total of 15 Kit Apc and 20 Kit 6–8-week-old mice were subjected to DSS treatment. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; ***P < 0.001. For reasons of clarity, the non-significant differences between groups were not indicated. b Embryonic fate mapping confirms the prenatal origins of intestinal LP and tumoural tissue-resident F4/ hi hi hi low MerCreMer/R26 MerCreMer/R26 Min/+ 80 MHCII and F4/80 MHCII macrophages. Pregnant Kit and Kit Apc mice (E10.5) were administered tamoxifen by intraperitoneal injection, as schematically illustrated in the upper panel. The offspring were then subjected to DSS treatment at 8 weeks old and sacrificed 4 weeks later for tissue and cell isolation. Representative flow cytometry analysis shows the labelling efficiency of each myeloid cell subset analyzed. Gating strategy is shown in Supplementary Fig. 4. The bar chart represents the mean percentage of YFP cells after normalization to the + MerCreMer/R26 Min/+ MerCreMer/R26 percentage of YFP neutrophils. The error bars represent the s.e.m. A total of 4 Kit Apc and 6 Kit mice were analysed. E, embryonic day; P1, monocytes; P2 and P3, monocyte-derived macrophages. Statistical significance was determined by two-way ANOVA followed by Bonferroni test. For reasons of clarity, the non-significant differences between groups were not indicated Of note, although hierarchical clustering between distinct colon undergo a robust metabolic reprogramming during tumour hi hi hi myeloid cells clearly identified F4/80 MHCII and F4/80 MH- progression. low CII as demarcated populations, “high/low” pairs always formed more closely related couples in tumours or in the LP than “low/ low” or “high/high” comparisons between samples. Although CSF1 maintains and expands pro-tumoural resident macro- hi hi these data suggested a close relationship between F4/80 MHCII phages. CSF1 is a crucial growth factor for macrophage pro- hi low and F4/80 MHCII cells (Fig. 6g), ingenuity pathway analysis liferation and survival, and supports the self-maintenance of 26–29 (IPA ) revealed that multiple pathways involved in glycolysis, tissue-resident macrophages . Here, we compared the levels gluconeogenesis, urea cycle, hypoxia-inducible factor-1 and of CSF1 in the colon LP and colon adenomas and detected a colorectal cancer metastasis signalling were significantly upregu- progressive increase in the amounts of CSF1 as a function of hi low lated in F4/80 MHCII cells (P < 0.05 by right-tailed Fisher’s polyp size (Fig. 7a). In addition, when the effects of CSF1 were exact test, Supplementary Fig. 3, red arrows). Clearly, tumour- neutralized upon exposure to an anti-CSF1 receptor neutralizing hi low Min/+ resident F4/80 MHCII cells not only accumulate, but also antibody in tumour-bearing Apc mice, we observed a strong NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 7 | | | Neutrophils high MHCII low MHCII P1 P2 P3 Eosinophils Neutrophils high MHCII low MHCII P1 P2 P3 Eosinophils Tumour Lamina propria CD45 CD45 YFP after normalization (%) YFP after normalization (%) CD45 CD45 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 hi low reduction in the number of both F4/80 MHCII and F4/ Student's t-test) and the tumours formed were of smaller size hi hi 80 MHCII tumour-resident macrophages, whereas the num- (P < 0.001 by unpaired Student's t-test) suggesting their pro- bers of monocytes and monocyte-derived macrophages were tumoural role in established tumours (Fig. 7c, lower panel). unperturbed (Fig. 7b). All together, these data strongly suggest hi that F4/80 -resident macrophages exposed to high levels of CSF1 Discussion (and likely other tumour niche factors) in the tumour micro- The macrophages found resident in the majority of tissues are environment gain the ability to self-renew within the tumour. established during fetal development and are able to locally self- To evaluate the pro-tumoural potential of different types of maintain in adulthood without any further input from BM hae- tumour-resident macrophages, polyp numbers and their sizes matopoietic progenitors . In tissues exposed to a low-grade −/− Min/+ were assessed in Ccr2 Apc and anti-CSF1R injected inflammation, however, such as the intestine and skin, or to Min/+ Apc mice. Clearly, CCR2 deficiency, although strongly mechanical stress, such as the heart, the maintenance of these reduced monocytes and monocyte-derived macrophage P2 and cells relies not on self-renewal, but on a continuous replenish- hi P3 populations, had no effect on colonic polyp numbers in Ccr2 ment by infiltrating BM-derived CCR2-dependent Ly6C circu- −/− Min/+ Min/+ 12,14,17 Apc mice compared to their Apc littermates in the lating monocytes . DSS-accelerated tumour model (Fig. 7c, upper panel). In contrast, Here we confirmed that in adult mice the vast majority of hi depletion of tumour-resident F4/80 macrophages lowered colon LP macrophages are derived from CCR2-dependent BM hi hi significantly the polyp counts in the colon (P < 0.05 by unpaired progenies. These macrophages included F4/80 MHCII LP- a b –2.0 0.0 2.0 Tumour LP LP Tumour ** low high low high *** MHCII MHCII hi lo hi lo MHCII MHCII MHCII MHCII MHCII MHCII ns 5 5 5 5 10 10 10 10 Nek3 Nek2 4 4 4 4 10 20 10 10 10 Cdc25c Ccna2 3 3 3 3 10 10 1 10 27 10 6 10 ns 23 Cdc20 0 0 0 0 Cks1b 3 3 3 3 –10 –10 –10 –10 Ccnb2 4 5 4 5 4 5 4 5 Cdca3 0 10 10 010 10 0 10 10 0 10 10 Low High Low High Cdca71 Ki67 Prc1 LP Tumour Nek4 Cdk1 E2f2 Rrm1 Mcm5 Sample Cdca8 Arg1 *** Mcm2 *** *** Mcm3 Mcm7 15 *** Mki67 10 Mcm4 LP Tumour lo hi hi lo MHCII MHCII MHCII MHCII 0 MMP2 Low High Low High Pdk3 4 *** *** ns LP Tumour Ganc Pdha1 LP Tumour 2 Dlat * Pdk2 Arg2 0 Aldoc MMP9 *** Ass1 2 3 4 5 2 3 4 5 ns 01 10 0 10 10 01 10 0 10 10 *** Ldhb H6pd Arg1 Pdk1 2 3 4 5 2 3 4 5 ns Dcxr 0 10 10 10 10 01 10 0 10 10 Arg1 Pgam1 0 Pgm2 ns MMP12 Eno1b 5000 *** ** 8 *** * Eno1 Tpi1 Ldha 4 ns Pkm ns ns Mdh1 Pgk1 0 Low High Low High Low High Low High Gapdh LP Tumour LP Tumour LP neutrophils 1 LP neutrophils 2 LP neutrophils 3 Tumour neutrophils 1 Tumour neutrophils Tumour neutrophils 2 Tumour neutrophils 3 LP neutrophils LP monocytes 1 LP monocytes 2 Tumour monocytes LP monocytes 3 Tumour monocytes 1 LP monocytes Tumour monocytes 2 hi lo Tumour F4/80 MHCII Tumour monocytes 3 hi hi Tumour F4/80 MHCII 1 hi hi 200 hi lo Tumour F4/80 MHCII 0 Tumour F4/80 MHCII 1 hi lo hi hi Tumour F4/80 MHCII 2 LP F4/80 MHCII hi lo –50 Tumour F4/80 MHCII 3 hi lo hi hi Tumour F4/80 MHCII 2 LP F4/80 MHCII –100 –50 hi hi –100 Tumour F4/80 MHCII 3 hi hi –150 –150 LP F4/80 MHCII 1 hi hi –200 –100 0 100 200 300 LP F4/80 MHCII 2 hi hi LP F4/80 MHCII 3 PC1 hi lo LP F4/80 MHCII 1 hi lo LP F4/80 MHCII 2 hi lo LP F4/80 MHCII 3 8 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | 0.05 0.10 0.15 0.20 0.25 0.30 0.35 F4/80 PC3 high low MHCII MHCII Mean fluorescence intensity (MFI) of Arg1 Relative expression (fold change) % Of Ki67 PC2 Relative expression (fold-change) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE resident macrophages (fraction I) in addition to a classical tumourigenesis , and focused on both subsets of tissue-resident int + hi monocyte-derived series of F4/80 CD11b cells (fraction II). We F4/80 macrophages in the colon LP. found one exception to this rule, whereby LP-resident F4/ The lymphoid and myeloid composition of colon adenomas of hi low Min/+ 80 MHCII macrophages that could be detected in high Apc mice differed from that of the colon LP. In the LP, we numbers only in young mice were also clearly found in adult Ccr2 found that lymphoid cells (mainly B cells) were the predominant −/− mice. Interestingly, our fate mapping experiments showed cell type, whereas in adenomas the majority consisted of myeloid hi that LP-resident F4/80 macrophages exhibited a very slow cells and were almost devoid of lymphocytes and, notably, eosi- turnover rate compared to other myeloid populations in the LP. nophils that were the most common LP myeloid cells. Neutrophils hi low In particular, F4/80 MHCII macrophages showed minimal and monocytes were strongly increased in adenomas, but interest- hi low low labelling after adult mouse tamoxifen administration, which ingly also F4/80 MHCII macrophages. These MHCII cells are suggests that they are a fetal-derived population that gradually usually a minor population in LP, but clearly “flourished” in the disappear with age. This phenomenon is perhaps expected due to tumour environment in our model to such an extent that in large hi increased competition to occupy the niche by monocyte-derived polyps they became the dominant population among F4/80 cells. hi hi −/− cells, which can then efficiently differentiate into F4/80 MHCII Our fate mapping studies and analyses of Ccr2 mice showed that 17 low macrophages . MHCII macrophages expanded in situ in adenomas by self- Similarly to the gut, an analogous tissue-resident macrophage renewal, without any recruitment of new cells. This finding was also subpopulation that expresses low levels of MHCII was recently supported by our RNA-seq and Ki-67 analyses showing that low characterized in the skin dermis—another tissue exposed to MHCII macrophages in polyps were actively cycling, unlike commensal microbes and under tonic, low-level inflammation . macrophages found in the colon LP. Similarly, the tumour micro- Consistent with the colon tissue-resident macrophages, the der- environment was also conducive for the self-renewal of F4/ hi hi hi hi mal F4/80 MHCII fraction is clearly BM derived, whereas the 80 MHCII macrophages. This subpopulation of macrophages hi low recruitment of the F4/80 MHCII subset in the dermis occurs was also actively cycling and had become independent from CCR2- mainly during embryogenesis and rapidly declines during dependent inputs from the BM. The tumour milieu, however, did 5,12 adulthood . not favour monocyte differentiation to macrophages. In adenomas, To date, the developmental origins of intestinal TAMs and monocytes (P1) got “stuck” at the beginning of the “waterfall” their maintenance in colon adenomas have not been extensively maturation pathway to macrophages (P3), which resulted in the F4/ hi studied. As macrophages have become potential targets for cancer 80 cells becoming the dominant macrophage population. This 31 13 therapy , we therefore decided to study the dynamics of myeloid effect has also been reported in the context of mammary cancers . 13 32 populations in colon adenomas. In breast cancer , glioma and Our data clearly suggest that unique characteristics of the tumour 33 hi hi lung metastases TAMs can arise from Ly6C circulating niche can trigger self-renewal of F4/80 macrophages. CSF1 is monocytes. Their influx and positioning in the tumour is medi- certainly a key component of the tumour niche to manipulate ated by chemokines (e.g., CCL2), growth factors (e.g., CSF1) and macrophage self-renewal, but other cues can also contribute to this 31,34 complement components . In addition, a recent study process and, therefore, reconstruction of the tumour niche would be demonstrated that local proliferation of TAMs contributes to very illuminating. their accumulation in spontaneous mammary tumours . Many According to the “M1/M2 paradigm”, mainly based on in vitro have speculated that in cancers exposed to microbes, such as polarization experiments, pro-inflammatory M1macrophages colon cancer, the recruitment of macrophages and their main- display anti-tumoural characteristics, whereas alternatively acti- tenance might differ from cancers located in “sterile” sites, such as vated M2-macrophages are generally considered pro-tumoural. breast cancers . Both populations can co-exist in the tumour, although M2-like low Here, we have analysed in detail the origins and maintenance MHCII TAMs are claimed to enrich in hypoxic and M1-like hi 22,36 of macrophage subpopulations in colon tumours obtained from MHCII TAMs in normoxic areas . Due to the complex Min/+ Apc mice, which is a widely used animal model for intestinal environment of the gut, intestinal tissue-resident macrophages do Fig. 6 Tissue-resident macrophages in polyps expand by self-renewal and undergo metabolic reprogramming. a Proliferative activity of the LP and tumoural hi hi hi low tissue-resident macrophage subpopulations. Intracellular expression levels of Ki-67 in adult F4/80 MHCII and F4/80 MHCII cells, obtained from the Min/+ colon LP of wild-type mice and colon tumours from dextran sodium sulphate-treated Apc mice. Representative dot plot and bar chart representing data obtain for colon LP (n = 5 mice) and tumours (n = 10 mice). Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; **P < 0.01; ***P < 0.001; ns, not significant. b, c Myeloid subpopulations were sorted from pooled colon LP and hi − hi − + tumours of 13-15 mice in three independent experiments in an RNA-seq study: Ly6C MHCII monocytes, Ly6G F4/80 CD11b neutrophils, F4/ hi hi hi low 80 MHCII and F4/80 MHCII tissue-resident macrophage subpopulations. Gating strategy is shown in Supplementary Fig. 4. b Heat map and hi hi hi low clustering of cell cycle-associated transcripts enriched in intratumoural F4/80 MHCII and F4/80 MHCII subsets. The heat map was generated with log2 transformed RPKM values and with the row/gene median subtracted. c Heat map and clustering of glycolysis- and urea cycle-associated transcripts hi hi hi low enriched in intratumoural F4/80 MHCII and F4/80 MHCII subsets. The heat map was generated with log2 transformed FPKM values and with the hi hi hi low row/gene median subtracted. d Quantitative PCR of Arg-1 expression levels in the LP and intratumoural F4/80 MHCII and F4/80 MHCII subpopulations. The qPCR results shown are representative of sorted macrophage subpopulations obtained from pooled LP and tumours of 13-15 mice. Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; ***P<0.001 e Representative flow cytometry histograms (upper panels) and bar charts (lower panel) showing the intracellular mean of fluorescence intensity (MFI) of ARG1 expression in hi hi hi low colon LP and intratumoural F4/80 MHCII and F4/80 MHCII cells. Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; **P<0.01; ***P < 0.001; ns, not significant. f Differential expression of distinct metalloproteinases (Mmp2, Mmp9 and hi hi hi low Mmp12) in LP and intratumoural F4/80 MHCII and F4/80 MHCII subsets measured by qPCR. Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; *P<0.05; ***P < 0.001; ns, not significant. g Transcriptome analysis of distinct colon LP and tumoural myeloid cell populations. Hierarchical clustering of the LP and intratumoural monocytes (P1), neutrophils and tissue-resident macrophage hi hi low (F4/80 MHCII and MHCII ) subsets NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 9 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 CSF1 d–4 d–1 Anti-CSF1R b Ab or Min/+ DSS isotype Ab Apc 1 Week 4 Weeks Tumour formation Analysis of colon tumours Fraction I Fraction II low high Eosinophils MHCII MHCII 5 5 5 10 10 *** *** 12,000 *** 4 4 4 9000 10 10 10 3000 26 72 II 3 3 6000 3 2000 18 10 10 2 2 5 1000 50 10 10 0 0 0 0 3 4 5 3 4 5 3 4 5 0 10 10 10 01 10 10 10 0 10 100 P1 P2 P3 ns 5 0.5 5 5 ns 9000 ns 3000 10 10 I 20,000 4 4 4 6000 2000 II 10 10 10 15,000 14 84 10,000 3 3 3 10 10 10 3000 1000 2 2 10 10 8 0 0 0 0 0 28 Min/+ Min/+ Apc + isotype Ab Apc + anti-CSF1R Ab 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 CD11b MHCII c Total polyps Polyps < 2 mm Polyps > 2 mm Min/+ Apc –/– Min/+ DSS 20 10 Ccr2 Apc 1 Week 4 Weeks Tumour formation Polyps count 10 5 0 0 Min/+ –/– Min/+ –/– Min/+ –/– Apc Ccr2 Apc Ccr2 Apc Ccr2 Min/+ Min/+ Min/+ Apc Apc Apc Total polyps Polyps < 2 mm Polyps > 2 mm 2 weeks 40 40 Anti-CSF1R Ab Min/+ Apc DSS or isotype Ab 30 30 1 Week 4 Weeks ** Tumour formation Polyps count 6 10 10 0 0 0 Min/+ Min/+ Min/+ Min/+ Min/+ Min/+ Apc + Apc + Apc + Apc + Apc + Apc + isotype Ab anti-CSF1R Ab isotype Ab anti-CSF1R Ab isotype Ab anti-CSF1R Ab hi Fig. 7 Pro-tumoural F4/80 resident macrophages depend on CSF1. a Intra-tumoural CSF1 levels gradually increase with tumour progression. Colon LP and polyps of different sizes (ranging from 0.5 mm to 6 mm in diameter) were processed for quantitative PCR analysis of Csf1 expression. Data were normalized to β-actin and are expressed as fold change in Csf1 expression from 0.5 mm tumours. b Anti-CSF1 receptor (CSF1R) antibody depletes intratumoural F4/ hi low hi hi Min/+ 80 MHCII and F4/80 MHCII macrophages. The 12-week-old dextran sodium sulphate (DSS)-treated Apc mice were injected intraperitoneally with rat IGg2a or anti-CSF1R blocking antibody (400 μg/mouse) at days −4 and −1 before collection of the colons, as shown in the schematic representation (upper panel). Polyps were processed and cells were analysed by flow cytometry for the presence of different myeloid cell populations hi hi defined in Fig. 1. Gating strategy is shown in Supplementary Fig. 4. The bar charts show the means±s.e.m. of absolute numbers of F4/80 MHCII and F4/ hi low 80 MHCII tissue-resident macrophage subpopulations, monocytes (P1), monocyte-derived macrophages (P2–P3) and eosinophils (control mice: n = 6 and anti-CSF1R Ab injected mice: n = 5). Statistical significance was determined using an unpaired Student’s t-test. ***P < 0.0001; ns, not significant. c Min/+ Min/+ −/− Min/+ CSF1R blockade but not CCR2 deficiency reduces tumour burden in Apc mice. Polyp counts in the colon of Apc (n = 9) and Ccr2 Apc mice Min/+ Min/+ (n = 9) (upper panel) and anti-rat IgG2a injected Apc (n = 11) and anti-CSF1R Ab injected Apc mice (n = 9) (lower panel). Statistical significance was determined using an unpaired Student’s t-test. *P < 0.05 and **P < 0.001 not follow the rigid M1–M2 classification and “classic” M1 and resident macrophage subset shows always the highest transcript 20,38 22,36,39 hi M2 features are often shared by the same cells . Similar mixed levels. As observed in other tumour models , F4/80 MH- low M1/M2 phenotype is observed in macrophages of colon tumours, CII cells, with their augmented metabolic switch together with hi hi low since both F4/80 MHCII and MHCII populations express their progressive predominance during tumour development, the classical M2 marker ARG1, but also many glycolytic genes represent a pro-tumoural macrophage subpopulation also in hi low typical for M1 macrophages, although the F4/80 MHCII - intestinal adenomas. 10 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | LP 0.5 mm 1–1.5 mm 2 mm 3–4 mm 5–6 mm Min/+ Min/+ Apc Apc + anti-CSF1R Ab + isotype Ab Relative expression F4/80 (fold change) Ly6C Polyps count/colon Polyps count/colon Polyps count/colon Polyps count/colon Absolute no. Absolute no. Polyps count/colon Polyps count/colon NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE Tumours were manually cut, minced and digested in 2% IMDM containing 1 Macrophages are not only crucial mediators of many pathol- mg/ml Collagenase D and 20 U/ml DNase I (Life Technologies, Carlsbad, CA, ogies (ranging from infectious to metabolic diseases and cancers) USA) under shaking conditions at 37 °C for 1 h. Digested tissue was subsequently but also influence the outcome of anticancer therapeutics, such as mashed through a cell strainer to obtain a cell suspension. The leukocyte 40–42 chemotherapy, radiotherapy and even immunotherapy .As population was enriched by 35% Percoll and further processed for flow cytometry and, if required, cell sorting using a FACSAria cell sorter (BD Biosciences, San Jose, such, these cells are becoming attractive candidates for ther- 43 CA, USA). Gating strategy used for sorting is shown in Supplementary Fig. 4. apeutic intervention. Manipulation of macrophage numbers and/or reprogramming their phenotype are techniques cur- rently used to target macrophages in a therapeutic context. A Transcriptomics analysis by RNA-sequencing. Total RNA was extracted using TM TM the Arcturus PicoPure RNA Isolation Kit (Cat. No. KIT0214, Thermo Fisher better understanding of their ontogeny in different tumours, such Scientific, Waltham, MA, USA) according to the manufacturer’s protocol. All as the data presented here for colon cancer, will facilitate the mouse RNAs were analysed using an Agilent Bioanalyser (Agilent, Santa Clara, CA, development of successful innovative macrophage-based ther- USA) for quality assessment; the RNA Integrity Number range was 6.5–10, with a apeutic approaches. median of 9.2. Complementary DNA (cDNA) libraries were prepared from 2 ng total RNA starting material and 1 μl of a 1:50,000 dilution of ERCC RNA Spike in Controls (Cat. No. 4456740, Ambion® Thermo Fisher Scientific) using the Methods SMARTSeq v2 protocol with the following modifications: (1) addition of 20 µM MerCreMer/R26 Mice. Kit mice were generated in our laboratory as previously TSO; and (2) use of 250 pg cDNA with 1/5 reaction of Illumina Nextera XT kit 5 −/− tm1lfc 45 Min/+ described . Ccr2 (B6.129S4-Ccr2 /J) , C57BL/6J and Apc (C57BL/6J- (Cat. No. FC-131–1024, Illumina, San Diego, CA, USA). The length distribution of 46 Min/+ ApcMin/J) mice were obtained from The Jackson Laboratory (USA). Apc the cDNA libraries was monitored using a DNA High Sensitivity Reagent Kit on MerCreMer/R26 −/− mice were backcrossed with Kit mice or Ccr2 mice to obtain the Perkin Elmer Labchip (Cat. No. CLS760672, Perkin Elmer, Waltham, MA, Min/+ MerCreMer/R26 −/− Min/+ Apc Kit and Ccr2 Apc mice, respectively. Mice were USA). All eight samples were subjected to an indexed paired-end sequencing run of specific-pathogen-free (SPF) and maintained under a 12 h light/dark cycle with 2 × 51 cycles on an Illumina HiSeq 2500 system (Illumina) under rapid run mode standard chow diet provided ad libitum. Both male and female mice (6–8 weeks (17 samples/lane). old) were used for all experiments and were equally distributed within experi- The paired-end reads were mapped to the Mouse GRCm38/mm10 reference mental and control groups. 49 genome using the STAR alignment tool . Mapped reads were summarized to the All transgenic mice were bred and housed under SPF conditions in the Nanyang gene level using featureCounts (V1.5.0-P1) software and with GENCODE gene Technological University animal facility. This study was carried out in strict 51 annotation . Genes with an average number of reads per sample <10 in all cell accordance with the recommendations of the NACLAR (National Advisory subpopulations were filtered out from subsequent analyses. For differentially Committee for Laboratory Animal Research) guidelines under the Animal & Birds expressed gene (DEG) analysis, the limma/voom pipeline was used as (Care and Use of Animals for Scientific Purposes) Rules of Singapore. The protocol 52 recommended by the MicroArray Quality Control (MAQC) project as one of the was approved by the Institutional Animal Care and Use Committee of the Nanyang best performing RNA-seq data analysis pipelines. Comparisons between different Technological University of Singapore. cell populations were performed using limma and DEGs were selected with Benjamini–Hochberg adjusted P-values < 0.05. Hierarchical clustering and Min/+ principal component analysis were performed with Log2 transformed value of Spontaneous and chemical-induced intestinal tumourigenesis. Apc mice RPKM (Reads Per Kilobase of transcript per Million mapped reads). All analyses spontaneously develop macroscopically detectable adenomas mainly in the small were carried out in R version 3.1.2 (URL http://www.R-project.org/). intestine and only few in the colon in aged mice . Therefore, for all our experi- ments (with exception of the experiments shown in Fig. 2 and Supplementary Fig. 1) we opted for a chemically induced colitis model which selectively enhances Flow cytometry. Single-cell suspensions were then stained and subsequently the development of adenomas in the colon in a relatively short time of 5 weeks. analysed by a BD Fortessa 5 laser flow cytometer (BD Bioscience). Data were Mice were terminated when showing symptoms of anaemia in combination with analysed using a FlowJo software (TreeStar, Ashland, OR, USA). The following weight loss and/or other signs of physical discomfort. antibodies were used: APC/Cy7-labelled anti-CD11b (clone: M1/70, 1:600) and In the “spontaneous” tumour model, mice were analysed at the age of BV605-labelled anti-Ly6C (HK1.4, 1:600) were purchased from Biolegend (San 6–7 months, whereas in case of chemically induced tumourigenesis, 6–8-week-old Min/+ Diego, CA, USA). PE-labelled anti-EMR1 (also known as F4/80) (clone: BM8, female and male Apc mice (and correspondent control C57BL/6J mice) were 1:600), PE/Cy7-labelled anti-CD45.2 (clone: 104, 1:600), APC-labelled anti-CD11c treated with 1.5 or 1.25% (w/v) DSS (50,000 Da; MP Biomedicals, Santa Ana, CA, (clone: N418, 1:600) and eFluor 450-labelled anti-MHCII (clone:M5/114.15.2, USA), respectively, in the drinking water for 1 week and sacrificed 4 weeks later at 1:800) were obtained from eBioscience (San Diego, CA, USA). FITC-labelled an age of 3–4 months. polyclonal sheep anti-human/mouse ARG1 (1:20) and APC-labelled anti-CCR2 (Cat. No. FAB5538A) (1:20) were obtained from R&D System (Minneapolis, MN, Colorectal polyp counts. Polyps in each colon were macroscopically counted and USA), FITC-labelled anti-Mouse/Rat Ki-67 (clone SolA15, 1:40) from Miltenyi categorized as >2 mm (large) and <2 mm (small). Colon tumours were measured Biotech (Bergisch-Gladbach, Germany) and BUV395-labelled anti-Ly6G (clone: ex vivo with the help of a sliding caliper. 1A8, 1:600) from BD (San Diego, CA, USA) MerCreMer/R26 Min/ Tamoxifen-inducible fate mapping mouse model. Kit and Apc Min/+ + MerCreMer/R26 Anti-CSF1R antibody treatment. Twelve-week-old DSS-treated Apc mice Kit mice were used for cell fate mapping to delineate the ontogeny of were injected intraperitoneally with rat Ig2a isotype control (Biolegend) or anti- intratumoural macrophages. Then, 80 mg tamoxifen (Sigma, T5648; Sigma- CSF1R blocking antibody (Clone AFS98, BioXCell, West Lebanon, NH, USA) (400 Aldrich, St. Louis, MO, USA) was dissolved in 6 ml corn oil (C8267; Sigma) and a μg/mouse) at days −4 and −1 before collection of the colons. total of 4 mg tamoxifen per mouse was administered for five consecutive days by gavage for adult labelling or by intraperitoneal injection in pregnant mothers (embryonic day 10.5) for embryonic labelling. After tamoxifen administration, c- Quantitative PCR (qPCR) analysis. cDNA was generated using SuperScript III + Min/+ MerCreMer/R26 kit BM-derived haematopoietic stem cells (HSCs) in Apc Kit + Reverse Transcriptase (Cat. No. 18080093, Invitrogen, Carlsbad, CA, USA) mice were labelled with yellow fluorescent protein (YFP ) and all cells deriving according to the manufacturer’s instructions. Quantitative real-time PCR was then from YFP BM-HSCs would maintain YFP expression. This effect enables to performed using the FAST 2× qPCR Master mix (PrecisionFAST-SY, Primerdesign distinguish whether adult intratumoural macrophages are derived from adult BM- Ltd, Cambridge, UK). Reactions were run on a real-time qPCR system (Illumina, definitive haematopoiesis or from embryonic haematopoiesis. Adult and fetal San Diego, CA, USA). Samples were normalized to β-actin, and data represent the labelling experimental protocols are shown in Fig. 5. hi hi mean of triplicate analyses and are displayed as a fold change from F4/80 MHCII LP macrophages unless otherwise stated. The primer sequences were as follows: Myeloid cell isolation. LP myeloid cells were isolated as previously described . Arg1; Fwd: gaatctgcatgggcaacc, Rev: gaatcctggtacatctgggaac; Mmp2; Fwd: taacctg- Briefly, after CO killing of the mice, the colon was opened and rinsed with gatgccgtcgt, Rev: ttcaggtaataagcacccttgaa; Mmp9; Fwd: catccagtatctgtatggtcgtg, Rev: phosphate-buffered saline (PBS) to remove the luminal contents. To remove the gctgtggttcagttgtggtg; Mmp12; Fwd: gctgctcccatgaatgaca, Rev: aagcattgcacacggttgt; epithelium, the colon was incubated in 25 ml PBS with 1.3 mM EDTA under Csf1; Fwd: ggtggaactgccagtatagaaag, Rev: tcccatatgtctccttccataaa; β-actin Fwd: shaking conditions at 37 °C for 1 h. After incubation, the colon was washed in 2% aaggccaaccgtgaaaagat, Rev: gtggtacgaccagaggcatac. Iscove’s modified Dulbecco’s medium (IMDM) to remove the EDTA solution, and then minced. The colon pieces were digested in 2% IMDM containing 1 mg/ml Collagenase D (Roche, Switzerland) under shaking conditions at 37 °C for 1.5 h. Statistical analysis. Statistical analysis was performed using GraphPad Prism The digested tissue was then gently mashed through a 150 μm cell strainer. The 6 software (GraphPad Software, La Jolla, CA, USA). All values are expressed as the leukocyte population was enriched using a 70/40% Percoll gradient (GE Healthcare mean±s.e.m as indicated in the legend. Samples were analysed by unpaired Stu- Life Science, Chicago, IL, USA). Low-density cells at the interface were harvested dent’s t-test (two-tailed) or Bonferroni two-way analysis of variance (ANOVA). A and processed further for flow cytometry. P-value of <0.05 was considered to be statistically significant. NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 11 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 Data availability. The accession number for the RNA-seq data reported in this 27. Tagliani, E. et al. Coordinate regulation of tissue macrophage and dendritic paper is GEO: GSE90153. Original flow cytometry data are deposited in the NTU cell population dynamics by CSF-1. J. Exp. Med. 208, 1901–1916 (2011). Open Access Data Repository DR-NTU (https://researchdata.ntu.edu.sg/dataset. 28. Hume, D. A., Pavli, P., Donahue, R. E. & Fidler, I. J. The effect of human xhtml?persistentId=doi:10.21979/N9/EQXCRF). All other data are available from recombinant macrophage colony-stimulating factor (CSF-1) on the murine the authors upon reasonable request. mononuclear phagocyte system in vivo. J. Immunol. 141, 3405–3409 (1988). 29. Stanley, E. R. & Chitu, V. CSF-1 receptor signaling in myeloid cells. Cold Spring Harb. Perspect. Biol. 6, a021857 (2014). Received: 11 July 2017 Accepted: 2 January 2018 30. Belkaid, Y. & Tamoutounour, S. The influence of skin microorganisms on cutaneous immunity. Nat. Rev. Immunol. 16, 353–366 (2016). 31. Noy, R. & Pollard, J. W. Tumor-associated macrophages: from mechanisms to therapy. Immunity 41,49–61 (2014). References 32. Pyonteck, S. M. et al. CSF-1R inhibition alters macrophage polarization and 1. Varol, C., Mildner, A. & Jung, S. Macrophages: development and tissue blocks glioma progression. Nat. Med. 19, 1264–1272 (2013). specialization. Annu. Rev. Immunol. 33, 643–675 (2015). 33. Qian, B. et al. A distinct macrophage population mediates metastatic breast 2. Gross, M., Salame, T. M. & Jung, S. Guardians of the gut - murine intestinal cancer cell extravasation, establishment and growth. PLoS ONE 4, e6562 macrophages and dendritic cells. Front. Immunol. 6, 254 (2015). (2009). 3. Ginhoux, F. & Guilliams, M. Tissue-resident macrophage ontogeny and 34. Bonavita, E. et al. PTX3 is an extrinsic oncosuppressor regulating homeostasis. Immunity 44, 439–449 (2016). complement-dependent inflammation in cancer. Cell 160, 700–714 (2015). 4. Perdiguero, E. G. & Geissmann, F. The development and maintenance of 35. Moser, A. R. et al. ApcMin: a mouse model for intestinal and mammary resident macrophages. Nat. Immunol. 17,2–8 (2016). tumorigenesis. Eur. J. Cancer 31A, 1061–1064 (1995). 5. Sheng, J., Ruedl, C. & Karjalainen, K. Most tissue-resident macrophages except 36. Van Overmeire, E. et al. M-CSF and GM-CSF receptor signaling differentially microglia are derived from fetal hematopoietic stem cells. Immunity 43, regulate monocyte maturation and macrophage polarization in the tumor 382–393 (2015). microenvironment. Cancer Res. 76,35–42 (2016). 6. Ginhoux, F. et al. Fate mapping analysis reveals that adult microglia derive 37. Bain, C. C. & Mowat, A. M. Macrophages in intestinal homeostasis and from primitive macrophages. Science 330, 841–845 (2010). inflammation. Immunol. Rev. 260, 102–117 (2014). 7. Schulz, C. et al. A lineage of myeloid cells independent of Myb and 38. Weber, B., Saurer, L., Schenk, M., Dickgreber, N. & Mueller, C. CX3CR1 hematopoietic stem cells. Science 336,86–90 (2012). defines functionally distinct intestinal mononuclear phagocyte subsets which 8. Yona, S. et al. Fate mapping reveals origins and dynamics of monocytes maintain their respective functions during homeostatic and inflammatory and tissue macrophages under homeostasis. Immunity 38,79–91 conditions. Eur. J. Immunol. 41, 773–779 (2011). (2013). 39. Wang, B. et al. Transition of tumor-associated macrophages from MHC class 9. Hashimoto, D. et al. Tissue-resident macrophages self-maintain locally II(hi) to MHC class II(low) mediates tumor progression in mice. BMC throughout adult life with minimal contribution from circulating monocytes. Immunol. 12, 43 (2011). Immunity 38, 792–804 (2013). 40. Colegio, O. R. et al. Functional polarization of tumour-associated 10. Guilliams, M. et al. Alveolar macrophages develop from fetal monocytes that macrophages by tumour-derived lactic acid. Nature 513, 559–563 (2014). differentiate into long-lived cells in the first week of life via GM-CSF. J. Exp. 41. Ries, C. H. et al. Targeting tumor-associated macrophages with anti-CSF-1R Med. 210, 1977–1992 (2013). antibody reveals a strategy for cancer therapy. Cancer Cell 25, 846–859 (2014). 11. Ginhoux, F., Lim, S., Hoeffel, G., Low, D. & Huber, T. Origin and 42. Mantovani, A. & Allavena, P. The interaction of anticancer therapies with differentiation of microglia. Front Cell Neurosci. 7, 45 (2013). tumor-associated macrophages. J. Exp. Med. 212, 435–445 (2015). 12. Tamoutounour, S. et al. Origins and functional specialization of macrophages 43. Engblom, C., Pfirschke, C. & Pittet, M. J. The role of myeloid cells in cancer and of conventional and monocyte-derived dendritic cells in mouse skin. therapies. Nat. Rev. Cancer 16, 447–462 (2016). Immunity 39, 925–938 (2013). 44. Schultze, J. L. Reprogramming of macrophages–new opportunities for 13. Franklin, R. A. et al. The cellular and molecular origin of tumor-associated therapeutic targeting. Curr. Opin. Pharmacol. 26,10–15 (2016). macrophages. Science 344, 921–925 (2014). 45. Boring, L. et al. Impaired monocyte migration and reduced type 1 (Th1) 14. Lavine, K. J. et al. Distinct macrophage lineages contribute to disparate cytokine responses in C-C chemokine receptor 2 knockout mice. J. Clin. patterns of cardiac recovery and remodeling in the neonatal and adult heart. Invest. 100, 2552–2561 (1997). Proc. Natl. Acad. Sci. USA 111, 16029–16034 (2014). 46. Moser, A. R., Pitot, H. C. & Dove, W. F. A dominant mutation that 15. Calderon, B. et al. The pancreas anatomy conditions the origin and properties predisposes to multiple intestinal neoplasia in the mouse. Science 247, of resident macrophages. J. Exp. Med. 212, 1497–1512 (2015). 322–324 (1990). 16. Bain, C. C. et al. Resident and pro-inflammatory macrophages in the colon 47. Muzaki, A. R. et al. Intestinal CD103(+)CD11b(-) dendritic cells restrain represent alternative context-dependent fates of the same Ly6Chi monocyte colitis via IFN-gamma-induced anti-inflammatory response in epithelial cells. precursors. Mucosal Immunol. 6, 498–510 (2013). Mucosal Immunol. 9, 336–351 (2016). 17. Bain, C. C. et al. Constant replenishment from circulating monocytes 48. Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. maintains the macrophage pool in the intestine of adult mice. Nat. Immunol. Protoc. 9, 171–181 (2014). 15, 929–937 (2014). 49. Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 18. Smythies, L. E. et al. Mucosal IL-8 and TGF-beta recruit blood monocytes: 15–21 (2013). evidence for cross-talk between the lamina propria stroma and myeloid cells. J. 50. Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose Leukoc. Biol. 80, 492–499 (2006). program for assigning sequence reads to genomic features. Bioinformatics 30, 19. Tamoutounour, S. et al. CD64 distinguishes macrophages from dendritic 923–930 (2014). cells in the gut and reveals the Th1-inducing role of mesenteric lymph node 51. Harrow, J. et al. GENCODE: producing a reference annotation for ENCODE. macrophages during colitis. Eur. J. Immunol. 42, 3150–3166 (2012). Genome Biol. 7(Suppl. 1), S4.1–S4.9 (2006). 20. Zigmond, E. et al. Macrophage-restricted interleukin-10 receptor deficiency, 52. SEQC/MAQC-III Consortium. A comprehensive assessment of RNA-seq but not IL-10 deficiency, causes severe spontaneous colitis. Immunity 40, accuracy, reproducibility and information content by the Sequencing Quality 720–733 (2014). Control Consortium. Nat. Biotechnol. 32, 903–914 (2014). 21. Shouval, D. S. et al. Interleukin-10 receptor signaling in innate immune cells regulates mucosal immune tolerance and anti-inflammatory macrophage function. Immunity 40, 706–719 (2014). 22. Movahedi, K. et al. Different tumor microenvironments contain functionally Acknowledgements distinct subsets of macrophages derived from Ly6C(high) monocytes. Cancer The authors would like to thank Monika Tetlak for providing excellent mouse man- Res. 70, 5728–5739 (2010). agement. The authors would also like to thank Insight Editing London for proofreading 23. Franklin, R. A. & Li, M. O. Ontogeny of tumor-associated macrophages and the manuscript prior to submission. This work was supported by MOE2014-T2-1-011 its implication in cancer regulation. Trends Cancer 2,20–34 (2016). and MOE2016-T2-1-012 Ministry of Education Tier2 grants to C.R. 24. Tymoszuk, P. et al. In situ proliferation contributes to accumulation of tumor- associated macrophages in spontaneous mammary tumors. Eur. J. Immunol. 44, 2247–2262 (2014). Author contributions 25. Shi, C. & Pamer, E. G. Monocyte recruitment during infection and Conceptualization: C.R.; methodology: I.S., J.S., S.F., J.L. and F.Z.; investigation: I.S., J.S. inflammation. Nat. Rev. Immunol. 11, 762–774 (2011). and Q.C.; formal analysis: I.S.; bioinformatic analysis: K.D. and M.P.; writing (original 26. Tushinski, R. J. et al. Survival of mononuclear phagocytes depends on a draft): C.R.; writing (review and editing): C.R. and K.K.; visualization: I.S.; funding lineage-specific growth factor that the differentiated cells selectively destroy. acquisition: C.R.; supervision: J.S., K.K. and C.R. Cell 28,71–81 (1982). 12 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE Additional information Open Access This article is licensed under a Creative Commons Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- Attribution 4.0 International License, which permits use, sharing, 018-02834-8. adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Competing interests: The authors declare no competing financial interests. Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless Reprints and permission information is available online at http://npg.nature.com/ indicated otherwise in a credit line to the material. If material is not included in the reprintsandpermissions/ article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in the copyright holder. To view a copy of this license, visit http://creativecommons.org/ published maps and institutional affiliations. licenses/by/4.0/. © The Author(s) 2018 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 13 | | | http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

The tumour microenvironment creates a niche for the self-renewal of tumour-promoting macrophages in colon adenoma

Loading next page...
 
/lp/springer-journals/the-tumour-microenvironment-creates-a-niche-for-the-self-renewal-of-L6O49yB7x7

References (53)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/s41467-018-02834-8
Publisher site
See Article on Publisher Site

Abstract

ARTICLE DOI: 10.1038/s41467-018-02834-8 OPEN The tumour microenvironment creates a niche for the self-renewal of tumour-promoting macrophages in colon adenoma 1 1 1 2 2 2 2 Irene Soncin , Jianpeng Sheng , Qi Chen , Shihui Foo , Kaibo Duan , Josephine Lum , Michael Poidinger , 2,3 1 1 Francesca Zolezzi , Klaus Karjalainen & Christiane Ruedl Circulating CCR2 monocytes are crucial for maintaining the adult tissue-resident hi hi F4/80 MHCII macrophage pool in the intestinal lamina propria. Here we show that hi low a subpopulation of CCR2-independent F4/80 MHCII macrophages, which are the hi most abundant F4/80 cells in neonates, gradually decline in number in adulthood; these hi macrophages likely represent the fetal contribution to F4/80 cells. In colon adenomas of Min/+ hi low Apc mice, F4/80 MHCII macrophages are not only preserved, but become the dominant subpopulation among tumour-resident macrophages during tumour progression. hi low hi hi Furthermore, these pro-tumoural F4/80 MHCII and F4/80 MHCII macrophages can self-renew in the tumour and maintain their numbers mostly independent from bone marrow contribution. Analyses of colon adenomas indicate that CSF1 may be a key facilitator of macrophage self-renewal. In summary, the tumour microenvironment creates an isolated niche for tissue-resident macrophages that favours macrophage survival and self-renewal. 1 2 School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore. Singapore Immunology Network, Agency for Science, Technology and Research (A*STAR), 8A Biomedical Grove, Singapore 138648, Singapore. Present address: GALDERMA R&D, 06902 Sophia Antipolis Cedex France. Klaus Karjalainen and Christiane Ruedl jointly supervised this work. Correspondence and requests for materials should be addressed to K.K. (email: Klaus@ntu.edu.sg) or to C.R. (email: Ruedl@ntu.edu.sg) NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 Min/+ 5 acrophages comprise a heterogeneous population of treated Apc mice (Fig. 1). Similar as previously reported , tissue-resident immune cells that contribute to tissue colon LP CD45 cells comprised three major myeloid cell hi hi Mhomeostasis, support the host defence system and can fractions (I–III) after excluding CD11c MHCII dendritic hi + impact on the initiation and propagation of several diseases, cells: F4/80 CD11b tissue-resident (tumour-resident in 1,2 int + including cancer (reviewed in refs. ). The majority of tissue- polyps) macrophages (fraction I), F4/80 CD11b (fraction II) 3–5 low + resident macrophages are established prenatally and are mainly and F4/80 CD11b neutrophils (fraction III) (Fig. 1a). The independent from any further haematopoietic input due to their majority (>85%) of fraction I cells expressed high levels of 6–10 capacity to self-renew in situ to maintain population size . For major histocompatibility complex (MHC) class II and only a example, microglia, the resident macrophage population of the small proportion (~5%) of cells were low for MHC class II central nervous system, develop early during embryonic devel- expression. Fraction II could be further subdivided into four opment from yolk sac precursors and effectively self-maintain distinct subpopulations in a waterfall-shaped distribution on a 5,6,11 in situ throughout adulthood . Although the majority of Ly6C vs. MHCII dot plot consisting of three distinct monocyte- hi − hi + the tissue-resident macrophages are maintained independently differentiation stages (P1: Ly6C MHCII ;P2: Ly6C MHCII ; − + − from bone marrow (BM)-derived monocytes, numerous studies P3: Ly6C MHCII ) and one eosinophil fraction (Ly6C MHCII have demonstrated that macrophages of certain tissues, including )(Fig. 1a). We then analysed the presence of these sub- 5,12 13 14 15 Min/+ the dermis , mammary gland , heart , pancreas and populations in DSS-accelerated colon tumours from Apc 5,16,17 intestine , require the input of monocytes to retain their pool mice. The same LP populations were also identified in colon during adulthood. In particular, the intestinal macrophages polyps but with markedly different frequencies (Fig. 1b). In depend on the constant and fast replenishment of circulating these tumours, neutrophils (fraction III) were the most abun- hi blood Ly6C monocytes, not only during inflammation but also dant myeloid cell type (>40%) followed by monocyte-derived 16–18 under normal healthy conditions . In a healthy unperturbed cells (fraction II) and tumour-resident macrophages (fraction colon, monocytes attracted to the intestine gradually differentiate I). Interestingly, eosinophils in tumours represented only a into tissue-resident macrophages by losing Ly6C expression, up- minor fraction (~3%) of the total CD11b cells. Furthermore, regulating the expression of macrophage markers, such as tumour-resident macrophages (fraction I) were clearly enriched 16,19 low CX3CR1, F4/80, CD64 and CD11c , and secreting or in MHCII cells (Fig. 1a,b), which were almost neglectable in 20,21 responding to the anti-inflammatory cytokine interleukin-10 . normal adult colon LP. Mice lacking either the chemokine receptor CCR2 or its ligand Characterization of myeloid subpopulations in the colon LP of CCL2 have reduced numbers of intestinal macrophages , thus untreated C57BL/6J mice and in “spontaneously” formed tumours Min/+ implying that the homeostatic recruitment of macrophage pre- obtained from 5-month-old Apc mice confirmed the cursors to the gut lamina propria (LP) may be dependent on a presence of all myeloid subpopulations described in Fig. 1, CCR2–CCL2 axis. Upon intestinal inflammation, tissue-resident although with a clear increase in the proportion of the tumoural + − + macrophages are still derived from circulating monocytes, but CD11b F4/80 neutrophil fraction (up to 70% of total CD11b convert from being anti-inflammatory macrophages to highly cells) (Supplementary Fig. 1). Toll-like receptor-responsive inflammatory cells . Tissue-resident macrophages are also naturally present in low various tumours, including lung and mammary tumours where MHCII macrophages accumulate during tumour progres- 22,23 24 hi macrophages are monocyte-derived or even self-renewing . sion. A previous study reported that F4/80 tissue-resident The developmental origins and the maintenance kinetics of macrophages in the colon LP progressively disappear with resident macrophages in intestinal tumours, however, have not increasing age . We therefore analysed the myeloid cell subsets been studied thus far. present in the fetal colon (embryonic day 19.5) and compared Here we show that the colon LP of normal adult mice contains them to those present in the colon LP of young and old mice hi hi 17 a major population of F4/80 MHCII macrophages and a (from 1 week to 12 months). In contrast to Bain et al., our hi low population of gradually disappearing F4/80 MHCII cells. Both analysis clearly found that all three macrophage fractions (I–III) hi F4/80 macrophage populations are relatively slowly replenished were maintained across all age groups with only minor fluctua- low by BM-derived cells, particularly MHCII cells, compared to tions detected in their frequency (Fig. 2a). The predominant int F4/80 monocyte/macrophage cells that have a fast turnover. tissue-resident macrophage subset in the fetal colon was the F4/ hi low The tumour microenvironment, however, is enriched with F4/ 80 MHCII fraction. However, as a function of age this mac- hi low hi hi 80 MHCII macrophages. Furthermore, while F4/80 MH- rophage subset almost disappeared to leave mainly F4/80 MH- low + hi CII macrophages do not require CCR2 monocytes for their CII cells to represent the colon tissue-resident macrophages low maintenance in the healthy gut LP or in tumours, we found that (fraction I) (Fig. 2b). This decrease in MHCII cell representa- hi hi + F4/80 MHCII macrophages become independent from CCR2 tion suggests that the LP environment gradually changes with age, monocytes in tumours only, potentially due to tumour-derived perhaps due to the establishment of a mature microbiota and/or a colony stimulating factor-1 (CSF1) that supports macrophage tonic increase in inflammatory mediators . hi low self-renewal. Moreover, we demonstrate that CCR2-independent Given the high frequency of F4/80 MHCII cells in colon hi intratumoural F4/80 macrophages, and not the CCR2- polyps (Fig. 1b), we assessed whether there was an association low dependent monocytes or monocyte-derived macrophages, sup- between tumour progression and the frequency of MHCII hi port cancer progression, suggesting this particular F4/80 cell macrophages. Therefore, for this analysis we opted for the Min/+ fraction as an attractive therapeutic target. spontaneous tumour model of Apc mice, which generate progressively adenomas up to 5–6 mm in size that is rarely achieved in the DSS-accelerated tumour model. Indeed, we Results observed an increasing abundance of this cell fraction in low Skewed representation of myeloid cells in colon adenomas.We conjunction with tumour progression until MHCII macro- first profiled colon LP myeloid subpopulations from wild-type phages became a dominant cell type among tumour-resident (WT) C57BL/6J mice, previously treated with dextran sodium macrophages (Fig. 2c). This observation clearly suggests that the sulphate (DSS), and compared them with their tumoural tumour microenvironment favours the differentiation or main- low counterparts obtained from the polyps collected from DSS- tenance of MHCII macrophages. 2 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE ab c Fraction Fraction Fraction Colon LP 4 I II III 3 685 5 Colon LP Fr I 4 30 Fr II 3 4 5 0 10 10 10 P1 3 P2 Fr III 10 5 2 eos 3 4 5 0 10 10 10 P3 low CD11b MHCII 3 4 5 high 0 10 10 10 MHCII MHCII P1 P2 P3 10 Fraction Fraction Fraction Eosinophils I II III Neutrophils Colon tumour Fr I Colon tumour Fr II 3 4 5 0 10 10 10 10 10 P1 P2 Fr III P3 2 3 4 5 2 eos 0 10 10 10 10 CD11b 3 4 5 0 10 10 10 MHCII Min/+ Fig. 1 Myeloid cell heterogeneity in colon lamina propria and colon tumours. WT C57BL/6J and Apc mice aged 6 weeks old were administered drinking water supplemented with dextran sodium sulphate for 7 days, and analysed 4 weeks later. a Flow cytometry representative dot plots of colon lamina propria (LP) (upper panel) and tumour cell subpopulations (lower panel). Three different myeloid fractions (I–III) are defined by the differential hi hi low expression of F4/80. Fraction I represents F4/80 tissue-resident macrophages, which can be further subdivided into MHCII and MHCII . Fraction II contains monocytes (P1), two monocyte-derived macrophage subpopulations (P2 and P3) and eosinophils (eos), based on differential expression of MHCII and Ly6C. Fr III consists of neutrophils. Gating strategy is shown in Supplementary Fig. 4. b Bar charts of the distinct myeloid cell subpopulations obtained from the colon LP (n = 8) (upper bar chart) and 2–3 mm colon tumours (n = 20) (lower bar chart). White bars: fraction I; light blue bars: fraction II and hi hi low black bars: fraction III. Error bars represent the s.e.m. c Pie charts show the proportions of F4/80 tissue-resident macrophages (MHCII and MHCII ), monocytes (P1), monocyte-derived macrophages (P2–P3), neutrophils and eosinophils across colon LP and tumours Intratumoural macrophages are independent from CCR2 Taken together, these data assert that the tumour microenvir- hi hi monocytes. Given that circulating Ly6C monocytes express onment can modify the properties of MHCII tissue-resident high levels of CCR2 receptor , we decided to examine the pre- macrophages such that they also become CCR2 independent like low sence of this receptor on tissue-resident and monocyte-derived MHCII cells in the LP and hence are presumably able to macrophages in colon LP and adenoma polyps. The aim of this maintain themselves without monocyte replenishment. experiment was to determine the potential origins of these cells. In the LP, consistent with being descendants of classical mono- hi − hi hi cytes, Ly6C MHCII (P1), Ly6C MHCII (P2) and Ly6C Turnover of colon lamina propria myeloid cells. We utilized the − + MerCreMer/R26 5 MHCII (P3) cells in fraction II subset had the highest and most Kit fate mapping mouse —where yellow florescence homogeneous expression of CCR2 (Fig. 3a) , whereas tissue- protein (YFP) expression can be induced in early BM progenitors hi hi hi low resident F4/80 MHCII and F4/80 MHCII macrophages —to monitor the cell population turnover rates driven by the BM were highly heterogeneous in their CCR2 expression levels, from input in colon LP myeloid cell populations. After injecting adult very high to very low levels (Fig. 3a and Supplementary Fig. 2). As mice with tamoxifen, the labelling index of tissue-resident mac- hi low hi expected, numbers of monocytes and monocyte-derived cells (P1, rophages (F4/80 MHCII and MHCII ), monocytes (P1), −/− P2 and P3) in fraction II were severely diminished in Ccr2 monocyte-derived macrophages (P2–P3 fractions), eosinophils hi mice (Fig. 3b). In addition, MHCII tissue-resident macrophages and neutrophils was analysed at different time points over a low were also mostly CCR2 dependent, whereas MHCII cells were period of 5 months. As anticipated, all tested myeloid popula- not, which is likely a reflection of their markedly lower CCR2 tions, with the exception of tissue-resident macrophages, were expression level (Fig. 3b). rapidly labelled with YFP and reached a plateau by 1–2 weeks hi Interestingly, we found that tumour-resident MHCII and after the final tamoxifen injection. These data suggest rapid low MHCII macrophages both expressed relatively low levels of replacement of myeloid cell populations by BM-derived cells CCR2 compared to their LP counterparts and, consistently, their (Fig. 4). −/− hi hi numbers did not change in Ccr2 mice. Conversely, fraction II Although it is widely considered that gut F4/80 MHCII cells cells (P1, P2 and P3) in polyps behaved exactly like their WT lose their prenatal origin and are replaced by CCR2 monocytes counterparts (Fig. 3a, c). in adulthood , our data clearly demonstrated that they are in fact NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 3 | | | low MHCII high MHCII P1 P2 P3 Eosinophils Neutrophils low MHCII high MHCII P1 P2 P3 Eosinophils Neutrophils F4/80 F4/80 Ly6C Ly6C % Of CD11b myeloid cells + % Of CD11b myeloid cells ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 a Fraction I 3 Weeks 4 Weeks 6 Weeks 8 Weeks 12 months 5 5 5 5 5 10 10 10 10 10 I 2 2 1.5 3 4 4 4 4 4 10 II 10 10 10 4 7 10 11 3 5 3 3 3 12 10 10 10 10 10 0.5 0.3 0 0.5 0 0.3 0 0 0.3 III 3 4 5 3 4 5 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 CD11b E19.5 2 Days 1 Week 3 Weeks 5 5 5 5 10 10 10 10 4 4 4 4 10 10 10 10 low 80 MHCII 3 3 3 3 10 10 49 50 10 10 93 6 47 51 high MHCII 2 2 2 10 10 10 0 0 3 4 5 3 4 5 3 4 5 3 4 5 40 10 10 10 01 10 10 10 0 10 100 0 10 10 10 0 4 Weeks 6 Weeks 8 Weeks 12 Months 5 5 5 10 10 10 4 4 4 10 10 10 2 96 3 18 79 3 3 3 2 10 10 13 10 10 0 0 0 0 Age 3 4 5 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 1001 10 10 10 0 10 100 MHCII 0.5 mm 1 mm 2 mm 3–4 mm 5–6 mm low MHCII 5 5 5 5 60 5 high 10 10 10 10 MHCII 4 4 4 4 4 40 10 10 10 10 73 63 62 36 21 32 45 19 70 49 3 3 3 3 10 10 3 10 10 10 20 2 2 2 2 10 10 10 10 0 0 0 0 0 3 4 5 3 4 5 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 0 10 10 10 MHCII Tumour diameter hi hi hi low hi Fig. 2 Ageing and tumour progression influence the ratio between F4/80 MHCII and F4/80 MHCII subpopulations. a F4/80 tissue-resident macrophages persist in the colon LP after birth into adulthood, as shown by the representative flow cytometry dot plots and bar chart. Error bars represent hi low hi hi hi low the s.e.m. b The frequency of colon F4/80 MHCII cells rapidly declines after birth. Representative dot plots of F4/80 MHCII and F4/80 MHCII subpopulations obtained from fetal colon (E19.5) and from colons of mice aged 2 days, 1, 3, 4, 6 and 8 weeks and 12 months. The bar chart represents the hi hi hi low age-dependent ratio between F4/80 MHCII (blue) and F4/80 MHCII (red) subpopulations in the colon LP. E19.5: n = 1 (pool of 15 embryos obtained from 3 different pregnant mice); day 2: n = 2 (each group pool of 5 mice); 1 week: n = 2 (each group pool of 5 mice); 3 weeks: n = 6 mice; 4 weeks: n = 5 mice; 6 weeks: n = 4; 8 weeks: n = 7 and 12 months: n = 4 mice. Error bars represent the s.e.m. c Representative flow cytometry analysis (left panel) and hi hi hi low mean percentage of F4/80 MHCII and F4/80 MHCII subpopulations (right panel) in tumours of different sizes (0.5–6.0 mm in diameter) obtained Min/+ from Apc mice. 0.5 mm: n = 3; 1 mm: n = 3; 2 mm: n = 4; 3–4 mm: n = 3; 5–6 mm: n = 4. Error bars represent the s.e.m. Scale bars: 2.5 mm. Gating strategy is shown in Supplementary Fig. 4 replaced by BM-derived monocytes only very slowly, if at all. Our “frozen” to the level corresponding to the time for their entrap- fate mapping data indicated that it took several months to fully ment in this niche that became independent from new inputs hi hi low replace F4/80 MHCII cells, whereas MHCII macrophages (Fig. 4 and Fig. 5a). exhibited minimal labelling (Fig. 4), suggesting that these tissue- Of note, when tamoxifen was administered during embryogen- resident macrophages retained their original fetal seed population esis (Fig. 5b, upper panel), all myeloid cell subpopulations— for a long period. including tissue-resident macrophages in the healthy LP and colon tumours—exhibited similar YFP tagging patterns (Fig. 5b, lower panel). This result was expected, as all of these cells are the progeny of classical haematopoietic stem cells, as previously Intratumoural resident macrophages expand by self-renewal. hi low described . We noted that F4/80 MHCII tumour-resident macrophages We found clear evidence, however, for increased self-renewal in particular seemed to expand in polyps and thus wanted to hi of F4/80 macrophages that could explain their capacity to determine if this effect was due to increased self-renewal or the expand in polyps. Firstly, tumour-resident macrophages were recruitment of new cells. Again, our adult fate mapping analyses MerCreMer/R26 Min/+ clearly in an active phase of the cell cycle unlike those in the LP, in Kit Apc mice showed that the recruitment of as indicated by the expression of the Ki-67 cellproliferation new cells to DSS-accelerated adenomas was not increased and hi low marker (Fig. 6a). Secondly, RNA-sequencing (RNA-seq) that the population turnover rate of F4/80 MHCII cells was hi analyses of cell-sorted, purified tumour-resident F4/80 (both comparable to that of the LP (Fig. 5a). A decrease in YFP labelling hi low hi hi MHCII and MHCII ) cells showed a clear upregulation of was observed in tumour-resident F4/80 MHCII cells when several cell cycle regulators, including cyclin-dependent kinase compared to their LP counterparts. This is probably due to the Cdk1, cyclin-A2 and -B2, E2f2, NEK family of serine/threonine fact that the labelling index of macrophages in polyps was 4 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | 3 Weeks 4 Weeks 6 Weeks 8 Weeks 12 Months E19.5 2 Days 1 Week 3 Weeks 4 Weeks 6 Weeks 8 Weeks 12 Months 0.5 mm 1 mm 2 mm 3–4 mm 5–6 mm Ly6C F4/80 Ly6C Subset (%) Subset (%) % Of CD11b myeloid cells NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE Lamina propria Tumour low MHCII high MHCII P1 P2 P3 Eosinophils 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 CCR2 b Fraction II Fraction I low high Eosinophils MHCII MHCII 5 5 5 0.6 10 10 10 ns ns 6000 *** 4 4 300 4 20 10 10 10 10,000 II 3 3 200 10 10 10 2 2 0.5 100 10 10 0 0 0 0 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 P1 P2 P3 5 5 5 10 0.2 10 *** I *** 2500 *** 3000 1.8 4 4 4 6000 10 0.8 8 2000 22 2000 II 3 3 1500 3 10 0.4 0 28 500 0 0 0 0 0 3 4 5 3 4 5 3 4 5 0 10 10 1001 10 10 10 0 10 100 CD11b MHCII low high Eosinophils MHCII MHCII 5 5 10 10 10 *** ns ns 8000 4000 4 4 4 I 17 10 10 10 9000 11 6000 3000 II 3 3 3 10 10 10 4000 2000 2 2 9 10 10 3000 2000 1000 0 0 0 0 0 3 4 5 3 4 5 3 4 5 01 10 10 10 01 10 100 0 10 100 P1 P2 P3 5 5 5 10 10 5 ** 25,000 *** *** 4 4 4 1 20,000 10 10 10 II 23 73 15,000 6000 3 3 3 10 10 10,000 48 1000 2 2 3000 10 10 10 0.5 0 76 0 0 0 0 0 2 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 10 MHCII CD11b hi + Fig. 3 Lamina propria and intratumoural colon F4/80 CD11b macrophage subsets show differential CCR2-driven monocyte dependence. a Fluorescence- activated cell sorting histograms showing CCR2 expression profiles on distinct colon LP and intratumoural myeloid cell subpopulations: fraction I consists low hi of MHCII and MHCII cells, and fraction II consists of monocytes (P1) and monocyte-derived macrophages (P2 and P3). b Myeloid cell profiling in WT −/− and Ccr2 colon LP. Representative flow cytometry analysis with F4/80 and CD11b-expressing myeloid subpopulations, obtained from the colon LP of −/− WT and Ccr2 mice. Fractions I and II were further dissected for MHCII and Ly6C expression (left panel). The absolute numbers of eosinophils, F4/ hi hi hi low 80 MHCII and F4/80 MHCII tissue-resident macrophages and P1–P3 subpopulations are shown (right panel). The bar chart represents the mean −/− −/− number of mice in each group and the error bars represent the s.e.m. (WT; n = 7 and Ccr2 ; n = 14). c Myeloid cell profiling in WT and Ccr2 colon hi hi hi low tumours as described above for LP cells. Bar charts show the mean±s.e.m. of absolute numbers of eosinophils, F4/80 MHCII and F4/80 MHCII Min/+ Min/+ −/− tissue-resident macrophages and P1–P3 subpopulations obtained from Apc (n = 8) and Apc Ccr2 (n = 9) mice. Statistical significance was determined using an unpaired Student's t-test. **P<0.001; ***P < 0.0001; ns, not significant. Gating strategy is shown in Supplementary Fig. 4 hi low kinases and various cell division cycle and mini-chromosome because F4/80 MHCII macrophages gradually disappear in maintenance proteins family members, in comparison to their the adult LP and probably do not vigorously self-renew, and hi hi LP counterparts (Fig. 6b). Interestingly, we could not find any that F4/80 MHCII cells are predominantly maintained in the hi hi evidence for active cell cycling of F4/80 MHCII and F4/ LP by recruited BM-derived cells and, therefore, do not depend hi low 80 MHCII cells in the LP (Fig. 6a, b). This effect could be on in situ self-renewal. NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 5 | | | WT –/– Ccr2 WT –/– Ccr2 WT –/– Ccr2 Min/+ Apc –/– Ccr2 Min/+ Apc Min/+ Apc –/– Ccr2 Min/+ Apc Min/+ Apc –/– Ccr2 Min/+ Apc –/– Min/+ Min/+ Ccr2 Apc Apc –/– Ccr2 WT F4/80 F4/80 Fraction II Fraction I Ly6C Ly6C Absolute no. Absolute no. Absolute no. Absolute no. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 1 Week 2 Weeks 3 Weeks Analysis of YFP cells 4 Weeks Tamoxifen MerCreMer/R26 7 Weeks Kit 20 Weeks Tracing of BM-precursors b low high Neutrophils MHCII MHCII P1 P2 P3 Eosinophils 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 4 10 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 22.9 0.3 10 0 4.5 0 0 0 0 0 0 24.9 21.9 12.6 11.7 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 4 10 10 10 10 10 10 10 3 10 3 3 3 3 10 10 10 10 10 10 14.3 0 0.8 5.5 0 0 0 0 0 13.5 12.8 14.9 3 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 4 5 –10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 0 10 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 4 10 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 3.3 2 0.4 7.6 0 0 0 0 0 0 6.5 7.1 8.4 0 6.5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 1.3 4.1 5.7 0 0 0 0 0 0 5.3 5.2 4.8 3 3 3 3 3 3 5.8 –10 –10 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 10 10 10 10 10 10 3 3 3 3 3 3 3 10 10 10 10 10 10 10 1.9 7.8 12.3 0 0 0 0 0 0 0 10.7 11.8 11.3 10.9 3 3 3 3 3 3 3 –10 –10 –10 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 5 5 5 5 5 5 5 10 10 10 10 10 10 10 4 4 4 4 4 4 10 10 10 10 10 10 3 3 3 3 3 3 10 10 10 10 10 10 2.5 2 13.8 10 0 0 12.9 0 0 0 0 0 13.6 13.7 3 3 3 3 3 13 3 14.1 –10 –10 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –1001 10 10 10 –10010 10 10 –10010 10 10 –10 00 10 10 YFP YFP 1 Week 2 Weeks 3 Weeks 4 Weeks 7 Weeks 20 Weeks low high Neutrophils MHCII MHCII P1 P2 P3 Eosinophils Fig. 4 Intestinal tissue-resident macrophages exhibit a slower, gradual replacement by bone marrow-derived cells compared to other gut myeloid cells. a MerCreMer/R26 Schematic representation of the adult fate mapping protocol using Kit mice. Mice aged 6 weeks were injected with tamoxifen five times and groups of 4–8 animals were sacrificed 1, 2, 3, 4, 7 and 20 weeks later. b Representative flow cytometry analysis indicating the labelling efficiency of distinct low hi colon myeloid cell populations as defined in Fig. 1: MHCII -expressing and MHCII -expressing tissue-resident macrophages, monocytes and monocytes- derived macrophages (P1–P3), eosinophils and neutrophils. Neutrophils acted as internal controls for labelling efficiency and the tracings are from the same mouse. Gating strategy is shown in Supplementary Fig. 4. c The bar chart represents the mean percentage of yellow florescence protein-positive (YFP ) cells after normalization to the percentage of YFP neutrophils. The error bars represent the s.e.m. 26,27 low Tumour microenvironment alters resident macrophage phe- tumour-associated macrophages (TAMs) , with MHCII notype. Besides the upregulation of genes involved in cell pro- macrophages being the major producers of ARG1 within tumours liferation, macrophages in polyps clearly adopted a distinct (Fig. 6c–e). The expression of other urea cell cycle genes, such as metabolic signature. Similar to cancer cells, which can alter their Arginase-2 (ARG2) and Arginosuccinate synthetase 1 (AAS1), metabolism in aerobic glycolysis and production of lactate (so- was increased in tumour-resident macrophages as well (Fig. 6c). called Warburg effect) macrophages in polyps increased the Moreover, several transcripts of matrix metalloproteases (MMPs), transcript levels of key glycolytic genes, such as ENO1, GAPDH, such as MMP2, MMP9 and MMP12, involved in the degradation hi low TPI1, PGAM and LDHA, with the F4/80 MHCII macro- of extracellular matrix and promotion of metastasis were upre- phages expressing the highest levels (Fig. 6c). In addition, our gulated in both subpopulations of tissue-resident macrophages RNA-seq found that both tissue-resident subsets displayed when compared to their LP counterparts (Fig. 6f), suggesting a enhanced transcripts of ARG1, typical for M2-polarized and possible contribution of these cells in tumour progression. 6 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | YFP after normalization (%) 5 Months 7 Weeks 4 Weeks 3 Weeks 2 Weeks 1 Week CD11b F4/80 NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE ab MerCreMer/R26 Kit Tamoxifen E10.5 DSS Tamoxifen DSS MerCreMer/R26 Min/+ Kit 8 Weeks old Apc 1 Week 4 Weeks 6–7 Weeks 1 week 4 Weeks Min/+ 10 Weeks Apc Analysis of Analysis of Tumour formation Tracing of BM-precursors Tracing of BM-precursors Tumour formation + YFP cells YFP cells high low Neutrophils MHCII MHCII high low Neutrophils MHCII MHCII 5 5 5 10 10 10 5 5 5 10 10 10 4 4 4 10 10 10 4 4 4 10 10 10 3 3 3 2 10 10 7 10 3 3 3 10 10 69 10 65 0 0 0 65 0 0 3 3 3 –10 –10 –10 3 3 3 4 5 3 4 5 3 4 5 –10 –10 –10 0 10 10 10 0 10 10 10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 Eosinophils –10 10 10 10 P1 P2 P3 –10 0 10 10 10 –10 0 10 10 10 0 Eosinophils P1 P2 P3 5 5 5 5 10 10 10 10 5 5 5 5 10 10 10 10 4 4 4 4 10 10 10 10 4 4 4 4 10 10 10 10 3 3 3 3 10 10 10 10 10 10 3 3 3 3 11 64 69 10 63 10 10 10 0 0 0 0 0 0 0 3 3 3 3 –10 –10 –10 –10 3 3 3 3 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 –10 –10 –10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 YFP YFP high low high low Neutrophils MHCII MHCII Neutrophils MHCII MHCII 5 5 10 10 5 5 5 10 10 10 4 4 4 10 10 10 4 4 4 10 10 10 3 3 3 3 3 10 10 4 10 10 10 61 62 0 0 0 0 0 3 3 0 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 Eosinophils P1 P2 P3 Eosinophils P1 P2 P3 5 5 5 10 10 10 5 5 5 5 10 10 10 10 4 4 4 4 4 4 4 4 10 10 10 10 10 10 10 10 3 3 3 10 3 3 3 3 10 10 10 10 10 10 10 14 15 62 15 62 62 0 0 0 0 0 0 0 3 3 3 –10 –10 –10 –10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 3 3 4 5 3 3 4 5 3 3 4 5 3 3 4 5 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 –10 0 10 10 10 YFP YFP *** 80 *** 100 LP *** Tumour 0 0 Fig. 5 Identification of self-renewing colon macrophage subsets in healthy and tumour tissue. a Dextran sodium sulphate (DSS)-induced intestinal MerCreMer/R26 MerCreMer/R26 Min/+ adenoma formation was combined with adult fate mapping using the Kit and Kit Apc mice (schematic representation shown in upper panel). Representative flow cytometry analysis illustrates the colon myeloid cell subset labelling efficiency. Gating strategy is shown in Supplementary Fig. 4. The bar chart represents the mean percentage of yellow fluorescence protein-positive (YFP ) cells after normalization to the + MerCreMer/R26 Min/+ MerCreMer/R26 percentage of YFP neutrophils±s.e.m. A total of 15 Kit Apc and 20 Kit 6–8-week-old mice were subjected to DSS treatment. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; ***P < 0.001. For reasons of clarity, the non-significant differences between groups were not indicated. b Embryonic fate mapping confirms the prenatal origins of intestinal LP and tumoural tissue-resident F4/ hi hi hi low MerCreMer/R26 MerCreMer/R26 Min/+ 80 MHCII and F4/80 MHCII macrophages. Pregnant Kit and Kit Apc mice (E10.5) were administered tamoxifen by intraperitoneal injection, as schematically illustrated in the upper panel. The offspring were then subjected to DSS treatment at 8 weeks old and sacrificed 4 weeks later for tissue and cell isolation. Representative flow cytometry analysis shows the labelling efficiency of each myeloid cell subset analyzed. Gating strategy is shown in Supplementary Fig. 4. The bar chart represents the mean percentage of YFP cells after normalization to the + MerCreMer/R26 Min/+ MerCreMer/R26 percentage of YFP neutrophils. The error bars represent the s.e.m. A total of 4 Kit Apc and 6 Kit mice were analysed. E, embryonic day; P1, monocytes; P2 and P3, monocyte-derived macrophages. Statistical significance was determined by two-way ANOVA followed by Bonferroni test. For reasons of clarity, the non-significant differences between groups were not indicated Of note, although hierarchical clustering between distinct colon undergo a robust metabolic reprogramming during tumour hi hi hi myeloid cells clearly identified F4/80 MHCII and F4/80 MH- progression. low CII as demarcated populations, “high/low” pairs always formed more closely related couples in tumours or in the LP than “low/ low” or “high/high” comparisons between samples. Although CSF1 maintains and expands pro-tumoural resident macro- hi hi these data suggested a close relationship between F4/80 MHCII phages. CSF1 is a crucial growth factor for macrophage pro- hi low and F4/80 MHCII cells (Fig. 6g), ingenuity pathway analysis liferation and survival, and supports the self-maintenance of 26–29 (IPA ) revealed that multiple pathways involved in glycolysis, tissue-resident macrophages . Here, we compared the levels gluconeogenesis, urea cycle, hypoxia-inducible factor-1 and of CSF1 in the colon LP and colon adenomas and detected a colorectal cancer metastasis signalling were significantly upregu- progressive increase in the amounts of CSF1 as a function of hi low lated in F4/80 MHCII cells (P < 0.05 by right-tailed Fisher’s polyp size (Fig. 7a). In addition, when the effects of CSF1 were exact test, Supplementary Fig. 3, red arrows). Clearly, tumour- neutralized upon exposure to an anti-CSF1 receptor neutralizing hi low Min/+ resident F4/80 MHCII cells not only accumulate, but also antibody in tumour-bearing Apc mice, we observed a strong NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 7 | | | Neutrophils high MHCII low MHCII P1 P2 P3 Eosinophils Neutrophils high MHCII low MHCII P1 P2 P3 Eosinophils Tumour Lamina propria CD45 CD45 YFP after normalization (%) YFP after normalization (%) CD45 CD45 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 hi low reduction in the number of both F4/80 MHCII and F4/ Student's t-test) and the tumours formed were of smaller size hi hi 80 MHCII tumour-resident macrophages, whereas the num- (P < 0.001 by unpaired Student's t-test) suggesting their pro- bers of monocytes and monocyte-derived macrophages were tumoural role in established tumours (Fig. 7c, lower panel). unperturbed (Fig. 7b). All together, these data strongly suggest hi that F4/80 -resident macrophages exposed to high levels of CSF1 Discussion (and likely other tumour niche factors) in the tumour micro- The macrophages found resident in the majority of tissues are environment gain the ability to self-renew within the tumour. established during fetal development and are able to locally self- To evaluate the pro-tumoural potential of different types of maintain in adulthood without any further input from BM hae- tumour-resident macrophages, polyp numbers and their sizes matopoietic progenitors . In tissues exposed to a low-grade −/− Min/+ were assessed in Ccr2 Apc and anti-CSF1R injected inflammation, however, such as the intestine and skin, or to Min/+ Apc mice. Clearly, CCR2 deficiency, although strongly mechanical stress, such as the heart, the maintenance of these reduced monocytes and monocyte-derived macrophage P2 and cells relies not on self-renewal, but on a continuous replenish- hi P3 populations, had no effect on colonic polyp numbers in Ccr2 ment by infiltrating BM-derived CCR2-dependent Ly6C circu- −/− Min/+ Min/+ 12,14,17 Apc mice compared to their Apc littermates in the lating monocytes . DSS-accelerated tumour model (Fig. 7c, upper panel). In contrast, Here we confirmed that in adult mice the vast majority of hi depletion of tumour-resident F4/80 macrophages lowered colon LP macrophages are derived from CCR2-dependent BM hi hi significantly the polyp counts in the colon (P < 0.05 by unpaired progenies. These macrophages included F4/80 MHCII LP- a b –2.0 0.0 2.0 Tumour LP LP Tumour ** low high low high *** MHCII MHCII hi lo hi lo MHCII MHCII MHCII MHCII MHCII MHCII ns 5 5 5 5 10 10 10 10 Nek3 Nek2 4 4 4 4 10 20 10 10 10 Cdc25c Ccna2 3 3 3 3 10 10 1 10 27 10 6 10 ns 23 Cdc20 0 0 0 0 Cks1b 3 3 3 3 –10 –10 –10 –10 Ccnb2 4 5 4 5 4 5 4 5 Cdca3 0 10 10 010 10 0 10 10 0 10 10 Low High Low High Cdca71 Ki67 Prc1 LP Tumour Nek4 Cdk1 E2f2 Rrm1 Mcm5 Sample Cdca8 Arg1 *** Mcm2 *** *** Mcm3 Mcm7 15 *** Mki67 10 Mcm4 LP Tumour lo hi hi lo MHCII MHCII MHCII MHCII 0 MMP2 Low High Low High Pdk3 4 *** *** ns LP Tumour Ganc Pdha1 LP Tumour 2 Dlat * Pdk2 Arg2 0 Aldoc MMP9 *** Ass1 2 3 4 5 2 3 4 5 ns 01 10 0 10 10 01 10 0 10 10 *** Ldhb H6pd Arg1 Pdk1 2 3 4 5 2 3 4 5 ns Dcxr 0 10 10 10 10 01 10 0 10 10 Arg1 Pgam1 0 Pgm2 ns MMP12 Eno1b 5000 *** ** 8 *** * Eno1 Tpi1 Ldha 4 ns Pkm ns ns Mdh1 Pgk1 0 Low High Low High Low High Low High Gapdh LP Tumour LP Tumour LP neutrophils 1 LP neutrophils 2 LP neutrophils 3 Tumour neutrophils 1 Tumour neutrophils Tumour neutrophils 2 Tumour neutrophils 3 LP neutrophils LP monocytes 1 LP monocytes 2 Tumour monocytes LP monocytes 3 Tumour monocytes 1 LP monocytes Tumour monocytes 2 hi lo Tumour F4/80 MHCII Tumour monocytes 3 hi hi Tumour F4/80 MHCII 1 hi hi 200 hi lo Tumour F4/80 MHCII 0 Tumour F4/80 MHCII 1 hi lo hi hi Tumour F4/80 MHCII 2 LP F4/80 MHCII hi lo –50 Tumour F4/80 MHCII 3 hi lo hi hi Tumour F4/80 MHCII 2 LP F4/80 MHCII –100 –50 hi hi –100 Tumour F4/80 MHCII 3 hi hi –150 –150 LP F4/80 MHCII 1 hi hi –200 –100 0 100 200 300 LP F4/80 MHCII 2 hi hi LP F4/80 MHCII 3 PC1 hi lo LP F4/80 MHCII 1 hi lo LP F4/80 MHCII 2 hi lo LP F4/80 MHCII 3 8 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | 0.05 0.10 0.15 0.20 0.25 0.30 0.35 F4/80 PC3 high low MHCII MHCII Mean fluorescence intensity (MFI) of Arg1 Relative expression (fold change) % Of Ki67 PC2 Relative expression (fold-change) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE resident macrophages (fraction I) in addition to a classical tumourigenesis , and focused on both subsets of tissue-resident int + hi monocyte-derived series of F4/80 CD11b cells (fraction II). We F4/80 macrophages in the colon LP. found one exception to this rule, whereby LP-resident F4/ The lymphoid and myeloid composition of colon adenomas of hi low Min/+ 80 MHCII macrophages that could be detected in high Apc mice differed from that of the colon LP. In the LP, we numbers only in young mice were also clearly found in adult Ccr2 found that lymphoid cells (mainly B cells) were the predominant −/− mice. Interestingly, our fate mapping experiments showed cell type, whereas in adenomas the majority consisted of myeloid hi that LP-resident F4/80 macrophages exhibited a very slow cells and were almost devoid of lymphocytes and, notably, eosi- turnover rate compared to other myeloid populations in the LP. nophils that were the most common LP myeloid cells. Neutrophils hi low In particular, F4/80 MHCII macrophages showed minimal and monocytes were strongly increased in adenomas, but interest- hi low low labelling after adult mouse tamoxifen administration, which ingly also F4/80 MHCII macrophages. These MHCII cells are suggests that they are a fetal-derived population that gradually usually a minor population in LP, but clearly “flourished” in the disappear with age. This phenomenon is perhaps expected due to tumour environment in our model to such an extent that in large hi increased competition to occupy the niche by monocyte-derived polyps they became the dominant population among F4/80 cells. hi hi −/− cells, which can then efficiently differentiate into F4/80 MHCII Our fate mapping studies and analyses of Ccr2 mice showed that 17 low macrophages . MHCII macrophages expanded in situ in adenomas by self- Similarly to the gut, an analogous tissue-resident macrophage renewal, without any recruitment of new cells. This finding was also subpopulation that expresses low levels of MHCII was recently supported by our RNA-seq and Ki-67 analyses showing that low characterized in the skin dermis—another tissue exposed to MHCII macrophages in polyps were actively cycling, unlike commensal microbes and under tonic, low-level inflammation . macrophages found in the colon LP. Similarly, the tumour micro- Consistent with the colon tissue-resident macrophages, the der- environment was also conducive for the self-renewal of F4/ hi hi hi hi mal F4/80 MHCII fraction is clearly BM derived, whereas the 80 MHCII macrophages. This subpopulation of macrophages hi low recruitment of the F4/80 MHCII subset in the dermis occurs was also actively cycling and had become independent from CCR2- mainly during embryogenesis and rapidly declines during dependent inputs from the BM. The tumour milieu, however, did 5,12 adulthood . not favour monocyte differentiation to macrophages. In adenomas, To date, the developmental origins of intestinal TAMs and monocytes (P1) got “stuck” at the beginning of the “waterfall” their maintenance in colon adenomas have not been extensively maturation pathway to macrophages (P3), which resulted in the F4/ hi studied. As macrophages have become potential targets for cancer 80 cells becoming the dominant macrophage population. This 31 13 therapy , we therefore decided to study the dynamics of myeloid effect has also been reported in the context of mammary cancers . 13 32 populations in colon adenomas. In breast cancer , glioma and Our data clearly suggest that unique characteristics of the tumour 33 hi hi lung metastases TAMs can arise from Ly6C circulating niche can trigger self-renewal of F4/80 macrophages. CSF1 is monocytes. Their influx and positioning in the tumour is medi- certainly a key component of the tumour niche to manipulate ated by chemokines (e.g., CCL2), growth factors (e.g., CSF1) and macrophage self-renewal, but other cues can also contribute to this 31,34 complement components . In addition, a recent study process and, therefore, reconstruction of the tumour niche would be demonstrated that local proliferation of TAMs contributes to very illuminating. their accumulation in spontaneous mammary tumours . Many According to the “M1/M2 paradigm”, mainly based on in vitro have speculated that in cancers exposed to microbes, such as polarization experiments, pro-inflammatory M1macrophages colon cancer, the recruitment of macrophages and their main- display anti-tumoural characteristics, whereas alternatively acti- tenance might differ from cancers located in “sterile” sites, such as vated M2-macrophages are generally considered pro-tumoural. breast cancers . Both populations can co-exist in the tumour, although M2-like low Here, we have analysed in detail the origins and maintenance MHCII TAMs are claimed to enrich in hypoxic and M1-like hi 22,36 of macrophage subpopulations in colon tumours obtained from MHCII TAMs in normoxic areas . Due to the complex Min/+ Apc mice, which is a widely used animal model for intestinal environment of the gut, intestinal tissue-resident macrophages do Fig. 6 Tissue-resident macrophages in polyps expand by self-renewal and undergo metabolic reprogramming. a Proliferative activity of the LP and tumoural hi hi hi low tissue-resident macrophage subpopulations. Intracellular expression levels of Ki-67 in adult F4/80 MHCII and F4/80 MHCII cells, obtained from the Min/+ colon LP of wild-type mice and colon tumours from dextran sodium sulphate-treated Apc mice. Representative dot plot and bar chart representing data obtain for colon LP (n = 5 mice) and tumours (n = 10 mice). Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; **P < 0.01; ***P < 0.001; ns, not significant. b, c Myeloid subpopulations were sorted from pooled colon LP and hi − hi − + tumours of 13-15 mice in three independent experiments in an RNA-seq study: Ly6C MHCII monocytes, Ly6G F4/80 CD11b neutrophils, F4/ hi hi hi low 80 MHCII and F4/80 MHCII tissue-resident macrophage subpopulations. Gating strategy is shown in Supplementary Fig. 4. b Heat map and hi hi hi low clustering of cell cycle-associated transcripts enriched in intratumoural F4/80 MHCII and F4/80 MHCII subsets. The heat map was generated with log2 transformed RPKM values and with the row/gene median subtracted. c Heat map and clustering of glycolysis- and urea cycle-associated transcripts hi hi hi low enriched in intratumoural F4/80 MHCII and F4/80 MHCII subsets. The heat map was generated with log2 transformed FPKM values and with the hi hi hi low row/gene median subtracted. d Quantitative PCR of Arg-1 expression levels in the LP and intratumoural F4/80 MHCII and F4/80 MHCII subpopulations. The qPCR results shown are representative of sorted macrophage subpopulations obtained from pooled LP and tumours of 13-15 mice. Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; ***P<0.001 e Representative flow cytometry histograms (upper panels) and bar charts (lower panel) showing the intracellular mean of fluorescence intensity (MFI) of ARG1 expression in hi hi hi low colon LP and intratumoural F4/80 MHCII and F4/80 MHCII cells. Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; **P<0.01; ***P < 0.001; ns, not significant. f Differential expression of distinct metalloproteinases (Mmp2, Mmp9 and hi hi hi low Mmp12) in LP and intratumoural F4/80 MHCII and F4/80 MHCII subsets measured by qPCR. Error bars represent the s.e.m. Statistical significance was determined by two-way ANOVA followed by Bonferroni test; *P<0.05; ***P < 0.001; ns, not significant. g Transcriptome analysis of distinct colon LP and tumoural myeloid cell populations. Hierarchical clustering of the LP and intratumoural monocytes (P1), neutrophils and tissue-resident macrophage hi hi low (F4/80 MHCII and MHCII ) subsets NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 9 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 CSF1 d–4 d–1 Anti-CSF1R b Ab or Min/+ DSS isotype Ab Apc 1 Week 4 Weeks Tumour formation Analysis of colon tumours Fraction I Fraction II low high Eosinophils MHCII MHCII 5 5 5 10 10 *** *** 12,000 *** 4 4 4 9000 10 10 10 3000 26 72 II 3 3 6000 3 2000 18 10 10 2 2 5 1000 50 10 10 0 0 0 0 3 4 5 3 4 5 3 4 5 0 10 10 10 01 10 10 10 0 10 100 P1 P2 P3 ns 5 0.5 5 5 ns 9000 ns 3000 10 10 I 20,000 4 4 4 6000 2000 II 10 10 10 15,000 14 84 10,000 3 3 3 10 10 10 3000 1000 2 2 10 10 8 0 0 0 0 0 28 Min/+ Min/+ Apc + isotype Ab Apc + anti-CSF1R Ab 3 4 5 3 4 5 3 4 5 0 10 10 10 0 10 10 10 0 10 10 10 CD11b MHCII c Total polyps Polyps < 2 mm Polyps > 2 mm Min/+ Apc –/– Min/+ DSS 20 10 Ccr2 Apc 1 Week 4 Weeks Tumour formation Polyps count 10 5 0 0 Min/+ –/– Min/+ –/– Min/+ –/– Apc Ccr2 Apc Ccr2 Apc Ccr2 Min/+ Min/+ Min/+ Apc Apc Apc Total polyps Polyps < 2 mm Polyps > 2 mm 2 weeks 40 40 Anti-CSF1R Ab Min/+ Apc DSS or isotype Ab 30 30 1 Week 4 Weeks ** Tumour formation Polyps count 6 10 10 0 0 0 Min/+ Min/+ Min/+ Min/+ Min/+ Min/+ Apc + Apc + Apc + Apc + Apc + Apc + isotype Ab anti-CSF1R Ab isotype Ab anti-CSF1R Ab isotype Ab anti-CSF1R Ab hi Fig. 7 Pro-tumoural F4/80 resident macrophages depend on CSF1. a Intra-tumoural CSF1 levels gradually increase with tumour progression. Colon LP and polyps of different sizes (ranging from 0.5 mm to 6 mm in diameter) were processed for quantitative PCR analysis of Csf1 expression. Data were normalized to β-actin and are expressed as fold change in Csf1 expression from 0.5 mm tumours. b Anti-CSF1 receptor (CSF1R) antibody depletes intratumoural F4/ hi low hi hi Min/+ 80 MHCII and F4/80 MHCII macrophages. The 12-week-old dextran sodium sulphate (DSS)-treated Apc mice were injected intraperitoneally with rat IGg2a or anti-CSF1R blocking antibody (400 μg/mouse) at days −4 and −1 before collection of the colons, as shown in the schematic representation (upper panel). Polyps were processed and cells were analysed by flow cytometry for the presence of different myeloid cell populations hi hi defined in Fig. 1. Gating strategy is shown in Supplementary Fig. 4. The bar charts show the means±s.e.m. of absolute numbers of F4/80 MHCII and F4/ hi low 80 MHCII tissue-resident macrophage subpopulations, monocytes (P1), monocyte-derived macrophages (P2–P3) and eosinophils (control mice: n = 6 and anti-CSF1R Ab injected mice: n = 5). Statistical significance was determined using an unpaired Student’s t-test. ***P < 0.0001; ns, not significant. c Min/+ Min/+ −/− Min/+ CSF1R blockade but not CCR2 deficiency reduces tumour burden in Apc mice. Polyp counts in the colon of Apc (n = 9) and Ccr2 Apc mice Min/+ Min/+ (n = 9) (upper panel) and anti-rat IgG2a injected Apc (n = 11) and anti-CSF1R Ab injected Apc mice (n = 9) (lower panel). Statistical significance was determined using an unpaired Student’s t-test. *P < 0.05 and **P < 0.001 not follow the rigid M1–M2 classification and “classic” M1 and resident macrophage subset shows always the highest transcript 20,38 22,36,39 hi M2 features are often shared by the same cells . Similar mixed levels. As observed in other tumour models , F4/80 MH- low M1/M2 phenotype is observed in macrophages of colon tumours, CII cells, with their augmented metabolic switch together with hi hi low since both F4/80 MHCII and MHCII populations express their progressive predominance during tumour development, the classical M2 marker ARG1, but also many glycolytic genes represent a pro-tumoural macrophage subpopulation also in hi low typical for M1 macrophages, although the F4/80 MHCII - intestinal adenomas. 10 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | LP 0.5 mm 1–1.5 mm 2 mm 3–4 mm 5–6 mm Min/+ Min/+ Apc Apc + anti-CSF1R Ab + isotype Ab Relative expression F4/80 (fold change) Ly6C Polyps count/colon Polyps count/colon Polyps count/colon Polyps count/colon Absolute no. Absolute no. Polyps count/colon Polyps count/colon NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE Tumours were manually cut, minced and digested in 2% IMDM containing 1 Macrophages are not only crucial mediators of many pathol- mg/ml Collagenase D and 20 U/ml DNase I (Life Technologies, Carlsbad, CA, ogies (ranging from infectious to metabolic diseases and cancers) USA) under shaking conditions at 37 °C for 1 h. Digested tissue was subsequently but also influence the outcome of anticancer therapeutics, such as mashed through a cell strainer to obtain a cell suspension. The leukocyte 40–42 chemotherapy, radiotherapy and even immunotherapy .As population was enriched by 35% Percoll and further processed for flow cytometry and, if required, cell sorting using a FACSAria cell sorter (BD Biosciences, San Jose, such, these cells are becoming attractive candidates for ther- 43 CA, USA). Gating strategy used for sorting is shown in Supplementary Fig. 4. apeutic intervention. Manipulation of macrophage numbers and/or reprogramming their phenotype are techniques cur- rently used to target macrophages in a therapeutic context. A Transcriptomics analysis by RNA-sequencing. Total RNA was extracted using TM TM the Arcturus PicoPure RNA Isolation Kit (Cat. No. KIT0214, Thermo Fisher better understanding of their ontogeny in different tumours, such Scientific, Waltham, MA, USA) according to the manufacturer’s protocol. All as the data presented here for colon cancer, will facilitate the mouse RNAs were analysed using an Agilent Bioanalyser (Agilent, Santa Clara, CA, development of successful innovative macrophage-based ther- USA) for quality assessment; the RNA Integrity Number range was 6.5–10, with a apeutic approaches. median of 9.2. Complementary DNA (cDNA) libraries were prepared from 2 ng total RNA starting material and 1 μl of a 1:50,000 dilution of ERCC RNA Spike in Controls (Cat. No. 4456740, Ambion® Thermo Fisher Scientific) using the Methods SMARTSeq v2 protocol with the following modifications: (1) addition of 20 µM MerCreMer/R26 Mice. Kit mice were generated in our laboratory as previously TSO; and (2) use of 250 pg cDNA with 1/5 reaction of Illumina Nextera XT kit 5 −/− tm1lfc 45 Min/+ described . Ccr2 (B6.129S4-Ccr2 /J) , C57BL/6J and Apc (C57BL/6J- (Cat. No. FC-131–1024, Illumina, San Diego, CA, USA). The length distribution of 46 Min/+ ApcMin/J) mice were obtained from The Jackson Laboratory (USA). Apc the cDNA libraries was monitored using a DNA High Sensitivity Reagent Kit on MerCreMer/R26 −/− mice were backcrossed with Kit mice or Ccr2 mice to obtain the Perkin Elmer Labchip (Cat. No. CLS760672, Perkin Elmer, Waltham, MA, Min/+ MerCreMer/R26 −/− Min/+ Apc Kit and Ccr2 Apc mice, respectively. Mice were USA). All eight samples were subjected to an indexed paired-end sequencing run of specific-pathogen-free (SPF) and maintained under a 12 h light/dark cycle with 2 × 51 cycles on an Illumina HiSeq 2500 system (Illumina) under rapid run mode standard chow diet provided ad libitum. Both male and female mice (6–8 weeks (17 samples/lane). old) were used for all experiments and were equally distributed within experi- The paired-end reads were mapped to the Mouse GRCm38/mm10 reference mental and control groups. 49 genome using the STAR alignment tool . Mapped reads were summarized to the All transgenic mice were bred and housed under SPF conditions in the Nanyang gene level using featureCounts (V1.5.0-P1) software and with GENCODE gene Technological University animal facility. This study was carried out in strict 51 annotation . Genes with an average number of reads per sample <10 in all cell accordance with the recommendations of the NACLAR (National Advisory subpopulations were filtered out from subsequent analyses. For differentially Committee for Laboratory Animal Research) guidelines under the Animal & Birds expressed gene (DEG) analysis, the limma/voom pipeline was used as (Care and Use of Animals for Scientific Purposes) Rules of Singapore. The protocol 52 recommended by the MicroArray Quality Control (MAQC) project as one of the was approved by the Institutional Animal Care and Use Committee of the Nanyang best performing RNA-seq data analysis pipelines. Comparisons between different Technological University of Singapore. cell populations were performed using limma and DEGs were selected with Benjamini–Hochberg adjusted P-values < 0.05. Hierarchical clustering and Min/+ principal component analysis were performed with Log2 transformed value of Spontaneous and chemical-induced intestinal tumourigenesis. Apc mice RPKM (Reads Per Kilobase of transcript per Million mapped reads). All analyses spontaneously develop macroscopically detectable adenomas mainly in the small were carried out in R version 3.1.2 (URL http://www.R-project.org/). intestine and only few in the colon in aged mice . Therefore, for all our experi- ments (with exception of the experiments shown in Fig. 2 and Supplementary Fig. 1) we opted for a chemically induced colitis model which selectively enhances Flow cytometry. Single-cell suspensions were then stained and subsequently the development of adenomas in the colon in a relatively short time of 5 weeks. analysed by a BD Fortessa 5 laser flow cytometer (BD Bioscience). Data were Mice were terminated when showing symptoms of anaemia in combination with analysed using a FlowJo software (TreeStar, Ashland, OR, USA). The following weight loss and/or other signs of physical discomfort. antibodies were used: APC/Cy7-labelled anti-CD11b (clone: M1/70, 1:600) and In the “spontaneous” tumour model, mice were analysed at the age of BV605-labelled anti-Ly6C (HK1.4, 1:600) were purchased from Biolegend (San 6–7 months, whereas in case of chemically induced tumourigenesis, 6–8-week-old Min/+ Diego, CA, USA). PE-labelled anti-EMR1 (also known as F4/80) (clone: BM8, female and male Apc mice (and correspondent control C57BL/6J mice) were 1:600), PE/Cy7-labelled anti-CD45.2 (clone: 104, 1:600), APC-labelled anti-CD11c treated with 1.5 or 1.25% (w/v) DSS (50,000 Da; MP Biomedicals, Santa Ana, CA, (clone: N418, 1:600) and eFluor 450-labelled anti-MHCII (clone:M5/114.15.2, USA), respectively, in the drinking water for 1 week and sacrificed 4 weeks later at 1:800) were obtained from eBioscience (San Diego, CA, USA). FITC-labelled an age of 3–4 months. polyclonal sheep anti-human/mouse ARG1 (1:20) and APC-labelled anti-CCR2 (Cat. No. FAB5538A) (1:20) were obtained from R&D System (Minneapolis, MN, Colorectal polyp counts. Polyps in each colon were macroscopically counted and USA), FITC-labelled anti-Mouse/Rat Ki-67 (clone SolA15, 1:40) from Miltenyi categorized as >2 mm (large) and <2 mm (small). Colon tumours were measured Biotech (Bergisch-Gladbach, Germany) and BUV395-labelled anti-Ly6G (clone: ex vivo with the help of a sliding caliper. 1A8, 1:600) from BD (San Diego, CA, USA) MerCreMer/R26 Min/ Tamoxifen-inducible fate mapping mouse model. Kit and Apc Min/+ + MerCreMer/R26 Anti-CSF1R antibody treatment. Twelve-week-old DSS-treated Apc mice Kit mice were used for cell fate mapping to delineate the ontogeny of were injected intraperitoneally with rat Ig2a isotype control (Biolegend) or anti- intratumoural macrophages. Then, 80 mg tamoxifen (Sigma, T5648; Sigma- CSF1R blocking antibody (Clone AFS98, BioXCell, West Lebanon, NH, USA) (400 Aldrich, St. Louis, MO, USA) was dissolved in 6 ml corn oil (C8267; Sigma) and a μg/mouse) at days −4 and −1 before collection of the colons. total of 4 mg tamoxifen per mouse was administered for five consecutive days by gavage for adult labelling or by intraperitoneal injection in pregnant mothers (embryonic day 10.5) for embryonic labelling. After tamoxifen administration, c- Quantitative PCR (qPCR) analysis. cDNA was generated using SuperScript III + Min/+ MerCreMer/R26 kit BM-derived haematopoietic stem cells (HSCs) in Apc Kit + Reverse Transcriptase (Cat. No. 18080093, Invitrogen, Carlsbad, CA, USA) mice were labelled with yellow fluorescent protein (YFP ) and all cells deriving according to the manufacturer’s instructions. Quantitative real-time PCR was then from YFP BM-HSCs would maintain YFP expression. This effect enables to performed using the FAST 2× qPCR Master mix (PrecisionFAST-SY, Primerdesign distinguish whether adult intratumoural macrophages are derived from adult BM- Ltd, Cambridge, UK). Reactions were run on a real-time qPCR system (Illumina, definitive haematopoiesis or from embryonic haematopoiesis. Adult and fetal San Diego, CA, USA). Samples were normalized to β-actin, and data represent the labelling experimental protocols are shown in Fig. 5. hi hi mean of triplicate analyses and are displayed as a fold change from F4/80 MHCII LP macrophages unless otherwise stated. The primer sequences were as follows: Myeloid cell isolation. LP myeloid cells were isolated as previously described . Arg1; Fwd: gaatctgcatgggcaacc, Rev: gaatcctggtacatctgggaac; Mmp2; Fwd: taacctg- Briefly, after CO killing of the mice, the colon was opened and rinsed with gatgccgtcgt, Rev: ttcaggtaataagcacccttgaa; Mmp9; Fwd: catccagtatctgtatggtcgtg, Rev: phosphate-buffered saline (PBS) to remove the luminal contents. To remove the gctgtggttcagttgtggtg; Mmp12; Fwd: gctgctcccatgaatgaca, Rev: aagcattgcacacggttgt; epithelium, the colon was incubated in 25 ml PBS with 1.3 mM EDTA under Csf1; Fwd: ggtggaactgccagtatagaaag, Rev: tcccatatgtctccttccataaa; β-actin Fwd: shaking conditions at 37 °C for 1 h. After incubation, the colon was washed in 2% aaggccaaccgtgaaaagat, Rev: gtggtacgaccagaggcatac. Iscove’s modified Dulbecco’s medium (IMDM) to remove the EDTA solution, and then minced. The colon pieces were digested in 2% IMDM containing 1 mg/ml Collagenase D (Roche, Switzerland) under shaking conditions at 37 °C for 1.5 h. Statistical analysis. Statistical analysis was performed using GraphPad Prism The digested tissue was then gently mashed through a 150 μm cell strainer. The 6 software (GraphPad Software, La Jolla, CA, USA). All values are expressed as the leukocyte population was enriched using a 70/40% Percoll gradient (GE Healthcare mean±s.e.m as indicated in the legend. Samples were analysed by unpaired Stu- Life Science, Chicago, IL, USA). Low-density cells at the interface were harvested dent’s t-test (two-tailed) or Bonferroni two-way analysis of variance (ANOVA). A and processed further for flow cytometry. P-value of <0.05 was considered to be statistically significant. NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 11 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 Data availability. The accession number for the RNA-seq data reported in this 27. Tagliani, E. et al. Coordinate regulation of tissue macrophage and dendritic paper is GEO: GSE90153. Original flow cytometry data are deposited in the NTU cell population dynamics by CSF-1. J. Exp. Med. 208, 1901–1916 (2011). Open Access Data Repository DR-NTU (https://researchdata.ntu.edu.sg/dataset. 28. Hume, D. A., Pavli, P., Donahue, R. E. & Fidler, I. J. The effect of human xhtml?persistentId=doi:10.21979/N9/EQXCRF). All other data are available from recombinant macrophage colony-stimulating factor (CSF-1) on the murine the authors upon reasonable request. mononuclear phagocyte system in vivo. J. Immunol. 141, 3405–3409 (1988). 29. Stanley, E. R. & Chitu, V. CSF-1 receptor signaling in myeloid cells. Cold Spring Harb. Perspect. Biol. 6, a021857 (2014). Received: 11 July 2017 Accepted: 2 January 2018 30. Belkaid, Y. & Tamoutounour, S. The influence of skin microorganisms on cutaneous immunity. Nat. Rev. Immunol. 16, 353–366 (2016). 31. Noy, R. & Pollard, J. W. Tumor-associated macrophages: from mechanisms to therapy. Immunity 41,49–61 (2014). References 32. Pyonteck, S. M. et al. CSF-1R inhibition alters macrophage polarization and 1. Varol, C., Mildner, A. & Jung, S. Macrophages: development and tissue blocks glioma progression. Nat. Med. 19, 1264–1272 (2013). specialization. Annu. Rev. Immunol. 33, 643–675 (2015). 33. Qian, B. et al. A distinct macrophage population mediates metastatic breast 2. Gross, M., Salame, T. M. & Jung, S. Guardians of the gut - murine intestinal cancer cell extravasation, establishment and growth. PLoS ONE 4, e6562 macrophages and dendritic cells. Front. Immunol. 6, 254 (2015). (2009). 3. Ginhoux, F. & Guilliams, M. Tissue-resident macrophage ontogeny and 34. Bonavita, E. et al. PTX3 is an extrinsic oncosuppressor regulating homeostasis. Immunity 44, 439–449 (2016). complement-dependent inflammation in cancer. Cell 160, 700–714 (2015). 4. Perdiguero, E. G. & Geissmann, F. The development and maintenance of 35. Moser, A. R. et al. ApcMin: a mouse model for intestinal and mammary resident macrophages. Nat. Immunol. 17,2–8 (2016). tumorigenesis. Eur. J. Cancer 31A, 1061–1064 (1995). 5. Sheng, J., Ruedl, C. & Karjalainen, K. Most tissue-resident macrophages except 36. Van Overmeire, E. et al. M-CSF and GM-CSF receptor signaling differentially microglia are derived from fetal hematopoietic stem cells. Immunity 43, regulate monocyte maturation and macrophage polarization in the tumor 382–393 (2015). microenvironment. Cancer Res. 76,35–42 (2016). 6. Ginhoux, F. et al. Fate mapping analysis reveals that adult microglia derive 37. Bain, C. C. & Mowat, A. M. Macrophages in intestinal homeostasis and from primitive macrophages. Science 330, 841–845 (2010). inflammation. Immunol. Rev. 260, 102–117 (2014). 7. Schulz, C. et al. A lineage of myeloid cells independent of Myb and 38. Weber, B., Saurer, L., Schenk, M., Dickgreber, N. & Mueller, C. CX3CR1 hematopoietic stem cells. Science 336,86–90 (2012). defines functionally distinct intestinal mononuclear phagocyte subsets which 8. Yona, S. et al. Fate mapping reveals origins and dynamics of monocytes maintain their respective functions during homeostatic and inflammatory and tissue macrophages under homeostasis. Immunity 38,79–91 conditions. Eur. J. Immunol. 41, 773–779 (2011). (2013). 39. Wang, B. et al. Transition of tumor-associated macrophages from MHC class 9. Hashimoto, D. et al. Tissue-resident macrophages self-maintain locally II(hi) to MHC class II(low) mediates tumor progression in mice. BMC throughout adult life with minimal contribution from circulating monocytes. Immunol. 12, 43 (2011). Immunity 38, 792–804 (2013). 40. Colegio, O. R. et al. Functional polarization of tumour-associated 10. Guilliams, M. et al. Alveolar macrophages develop from fetal monocytes that macrophages by tumour-derived lactic acid. Nature 513, 559–563 (2014). differentiate into long-lived cells in the first week of life via GM-CSF. J. Exp. 41. Ries, C. H. et al. Targeting tumor-associated macrophages with anti-CSF-1R Med. 210, 1977–1992 (2013). antibody reveals a strategy for cancer therapy. Cancer Cell 25, 846–859 (2014). 11. Ginhoux, F., Lim, S., Hoeffel, G., Low, D. & Huber, T. Origin and 42. Mantovani, A. & Allavena, P. The interaction of anticancer therapies with differentiation of microglia. Front Cell Neurosci. 7, 45 (2013). tumor-associated macrophages. J. Exp. Med. 212, 435–445 (2015). 12. Tamoutounour, S. et al. Origins and functional specialization of macrophages 43. Engblom, C., Pfirschke, C. & Pittet, M. J. The role of myeloid cells in cancer and of conventional and monocyte-derived dendritic cells in mouse skin. therapies. Nat. Rev. Cancer 16, 447–462 (2016). Immunity 39, 925–938 (2013). 44. Schultze, J. L. Reprogramming of macrophages–new opportunities for 13. Franklin, R. A. et al. The cellular and molecular origin of tumor-associated therapeutic targeting. Curr. Opin. Pharmacol. 26,10–15 (2016). macrophages. Science 344, 921–925 (2014). 45. Boring, L. et al. Impaired monocyte migration and reduced type 1 (Th1) 14. Lavine, K. J. et al. Distinct macrophage lineages contribute to disparate cytokine responses in C-C chemokine receptor 2 knockout mice. J. Clin. patterns of cardiac recovery and remodeling in the neonatal and adult heart. Invest. 100, 2552–2561 (1997). Proc. Natl. Acad. Sci. USA 111, 16029–16034 (2014). 46. Moser, A. R., Pitot, H. C. & Dove, W. F. A dominant mutation that 15. Calderon, B. et al. The pancreas anatomy conditions the origin and properties predisposes to multiple intestinal neoplasia in the mouse. Science 247, of resident macrophages. J. Exp. Med. 212, 1497–1512 (2015). 322–324 (1990). 16. Bain, C. C. et al. Resident and pro-inflammatory macrophages in the colon 47. Muzaki, A. R. et al. Intestinal CD103(+)CD11b(-) dendritic cells restrain represent alternative context-dependent fates of the same Ly6Chi monocyte colitis via IFN-gamma-induced anti-inflammatory response in epithelial cells. precursors. Mucosal Immunol. 6, 498–510 (2013). Mucosal Immunol. 9, 336–351 (2016). 17. Bain, C. C. et al. Constant replenishment from circulating monocytes 48. Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. maintains the macrophage pool in the intestine of adult mice. Nat. Immunol. Protoc. 9, 171–181 (2014). 15, 929–937 (2014). 49. Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 18. Smythies, L. E. et al. Mucosal IL-8 and TGF-beta recruit blood monocytes: 15–21 (2013). evidence for cross-talk between the lamina propria stroma and myeloid cells. J. 50. Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose Leukoc. Biol. 80, 492–499 (2006). program for assigning sequence reads to genomic features. Bioinformatics 30, 19. Tamoutounour, S. et al. CD64 distinguishes macrophages from dendritic 923–930 (2014). cells in the gut and reveals the Th1-inducing role of mesenteric lymph node 51. Harrow, J. et al. GENCODE: producing a reference annotation for ENCODE. macrophages during colitis. Eur. J. Immunol. 42, 3150–3166 (2012). Genome Biol. 7(Suppl. 1), S4.1–S4.9 (2006). 20. Zigmond, E. et al. Macrophage-restricted interleukin-10 receptor deficiency, 52. SEQC/MAQC-III Consortium. A comprehensive assessment of RNA-seq but not IL-10 deficiency, causes severe spontaneous colitis. Immunity 40, accuracy, reproducibility and information content by the Sequencing Quality 720–733 (2014). Control Consortium. Nat. Biotechnol. 32, 903–914 (2014). 21. Shouval, D. S. et al. Interleukin-10 receptor signaling in innate immune cells regulates mucosal immune tolerance and anti-inflammatory macrophage function. Immunity 40, 706–719 (2014). 22. Movahedi, K. et al. Different tumor microenvironments contain functionally Acknowledgements distinct subsets of macrophages derived from Ly6C(high) monocytes. Cancer The authors would like to thank Monika Tetlak for providing excellent mouse man- Res. 70, 5728–5739 (2010). agement. The authors would also like to thank Insight Editing London for proofreading 23. Franklin, R. A. & Li, M. O. Ontogeny of tumor-associated macrophages and the manuscript prior to submission. This work was supported by MOE2014-T2-1-011 its implication in cancer regulation. Trends Cancer 2,20–34 (2016). and MOE2016-T2-1-012 Ministry of Education Tier2 grants to C.R. 24. Tymoszuk, P. et al. In situ proliferation contributes to accumulation of tumor- associated macrophages in spontaneous mammary tumors. Eur. J. Immunol. 44, 2247–2262 (2014). Author contributions 25. Shi, C. & Pamer, E. G. Monocyte recruitment during infection and Conceptualization: C.R.; methodology: I.S., J.S., S.F., J.L. and F.Z.; investigation: I.S., J.S. inflammation. Nat. Rev. Immunol. 11, 762–774 (2011). and Q.C.; formal analysis: I.S.; bioinformatic analysis: K.D. and M.P.; writing (original 26. Tushinski, R. J. et al. Survival of mononuclear phagocytes depends on a draft): C.R.; writing (review and editing): C.R. and K.K.; visualization: I.S.; funding lineage-specific growth factor that the differentiated cells selectively destroy. acquisition: C.R.; supervision: J.S., K.K. and C.R. Cell 28,71–81 (1982). 12 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02834-8 ARTICLE Additional information Open Access This article is licensed under a Creative Commons Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- Attribution 4.0 International License, which permits use, sharing, 018-02834-8. adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Competing interests: The authors declare no competing financial interests. Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless Reprints and permission information is available online at http://npg.nature.com/ indicated otherwise in a credit line to the material. If material is not included in the reprintsandpermissions/ article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in the copyright holder. To view a copy of this license, visit http://creativecommons.org/ published maps and institutional affiliations. licenses/by/4.0/. © The Author(s) 2018 NATURE COMMUNICATIONS (2018) 9:582 DOI: 10.1038/s41467-018-02834-8 www.nature.com/naturecommunications 13 | | |

Journal

Nature CommunicationsSpringer Journals

Published: Feb 8, 2018

There are no references for this article.