Origins of tumor-associated macrophages and neutrophils

Abstract

Tumor-associated macrophages (TAMs) and tumor-associated neutrophils (TANs) can control cancer growth and exist in almost all solid neoplasms. The cells are known to descend from immature monocytic and granulocytic cells, respectively, which are produced in the bone marrow. However, the spleen is also a recently identified reservoir of monocytes, which can play a significant role in the inflammatory response that follows acute injury. Here, we evaluated the role of the splenic reservoir in a genetic mouse model of lung adenocarcinoma driven by activation of oncogenic Kras and inactivation of p53. We found that high numbers of TAM and TAN precursors physically relocated from the spleen to the tumor stroma, and that recruitment of tumor-promoting spleen-derived TAMs required signaling of the chemokine receptor CCR2. Also, removal of the spleen, either before or after tumor initiation, reduced TAM and TAN responses significantly and delayed tumor growth. The mechanism by which the spleen was able to maintain its reservoir capacity throughout tumor progression involved, in part, local accumulation in the splenic red pulp of typically rare extramedullary hematopoietic stem and progenitor cells, notably granulocyte and macrophage progenitors, which produced CD11b(+) Ly-6C(hi) monocytic and CD11b(+) Ly-6G(hi) granulocytic cells locally. Splenic granulocyte and macrophage progenitors and their descendants were likewise identified in clinical specimens. The present study sheds light on the origins of TAMs and TANs, and positions the spleen as an important extramedullary site, which can continuously supply growing tumors with these cells.

Conflict of interest statement

Conflict of interest statement: T.I.N. and V.K. are Alnylam Pharmaceuticals employees, and D.G.A. receives funding from, and is a consultant with, Alnylam Pharmaceuticals.

Figures

Fig. 1.

Spleen removal impairs TAM and TAN responses and tumor growth. (A) Experimental procedures. (B) Total CD11b+ and CD11b− macrophage and neutrophil counts in lungs of mice 11 wk after tumor initiation [+ adenovirus expressing Cre recombinase (AdCre), n = 17] and in mice from which the spleen was removed either immediately before tumor initiation [+AdCre SPx (wk 0), n = 19] or after tumors became apparent [+AdCre SPx (wk 8), n = 7]. Control mice (−AdCre, n = 13) were age-matched. Histograms show median values, and dots represent single mice. (C) In vivo fluorescent-mediated tomography/CT fusion images and quantification of Cathepsin B activity at week 11 after AdCre infection in KP mice with (+AdCre +SP) or without [+AdCre −SP (wk 0)] a spleen (n = 3). Data are presented as the mean ± SEM. (D and E) High-resolution CT analysis of tumor progression at weeks 7, 9, and 11 in mice with (+AdCre, n = 7) or without [+AdCre −SP (wk 0), n = 12] a spleen. (D) Number and volume of tumor nodules in each group. Histograms show median values, and dots represent single mice. (E) Representative 3D volume renderings of tumors for each group. Tumors are shown in red, and lungs are shown in green. *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant.

Fig. 2.

CCR2 silencing in splenic monocytes impairs TAM responses and tumor growth. (A) Experimental procedures. AdCre, adenovirus expressing Cre recombinase. (B) Number of macrophages and neutrophils in the lungs of KP mice treated with siCCR2 (n = 4) or control-silencing (siCON; n = 4) nanoparticles or in tumor-free mice (no cancer). (C) Volume of tumor nodules in the same KP mice identified by high-resolution CT. *P < 0.05; n.s., not significant.

Fig. 3.

Spleen contributes TAMs and TANs directly. (A) Experimental procedure for transplantation of a BrdU-labeled spleen into a splenectomized (unlabeled) KP mouse. Procedural details are provided in SI Methods. (B) Transplanted mice received an i.v. injection of the blood pool agent Angiosense-680 (ViSen Medical–PerkinElmer) to reveal perfusion of the donor organ (spleen outlined by yellow dashed line). Scale bar: 5 mm. (C) The number of BrdU+ cells was detected by flow cytometry from donor spleen mobilized to the lung 24 h after transplantation. The recipient mice were either tumor-bearing (KP mice 11 wk after tumor initiation) or tumor-free (control mice). (D) Dot plots show endogenous cells (gray) and spleen donor-derived cells (red) 24 h posttransplantation in a tumor-bearing mouse (from n = 2). Percentages indicate the contribution of newly arrived splenic cells to the preexisting lung myeloid cell pool. Lin, lineage.

Fig. 4.

Tumors induce splenic accumulation of HSCs and progenitor cells. (A) Percentage of monocytic [CD11b+ Lin− Ly-6G− (F4/80/CD11c/I-Ab)lo Ly-6C+], granulocytic (CD11b+ Gr-1+ Ly-6G+), Lin+ CD11b− and Lin− CD11b− populations in S/G2 phase in the spleen of control and tumor-bearing mice (n = 3) at week 11 after adenovirus expressing Cre recombinase (AdCre) infection. Lineage (Lin) refers to the following combination: B220/NK1.1/DX5/Ly-6G/CD90.2. (B) (Upper) Number of granulocyte/macrophage colonies in CFU assays (GM-CFU) from splenocytes of tumor-bearing (+AdCre) and control (−AdCre) mice. (Lower) Image of granulocyte/macrophage colonies. Scale bar: 2 mm. (C) Identification of splenic HSCs and GMPs by flow cytometry in KP mice on week 11. (D) Quantification of splenic HSCs and GMPs in tumor-bearing (+AdCre) KP animals (n = 17) and age-matched controls (−AdCre, n = 12). Data are presented as the mean ± SEM. ***P < 0.001.

Fig. 5.

Fate mapping of adoptively transferred GMP cells. (A) (Upper) In situ confocal image of the splenic red pulp from a tumor-bearing mouse 5 d after i.v. injection of (ACTB-eYFP)7AC5Nagy/J (EYFP+) GMPs (green). EYFP+ cell clusters are highlighted in circles, and venus sinuses are shown in red. Scale bar: 1 mm. (Lower) Higher magnification images. Scale bar: 100 μm. (B) Phenotyping and quantification of the progeny of EYFP+ GMP cells in the spleen (SP) and bone marrow (BM) of tumor-bearing [+ adenovirus expressing Cre recombinase (AdCre), n = 12] and tumor-free mice (−AdCre, n = 6). i.n., intranasal. (C) Tracking of EYFP+ GMP cells injected into tumor-free mice (−AdCre, n = 6) or tumor-bearing KP mice with (+AdCre, n = 12) or without [+AdCre −SP (wk 0), n = 4] a spleen. Data are presented as the mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; n.s., not significant.

Fig. 6.

Splenic myeloid response in human patients with invasive cancer. (A) Identification of GMP-like cells (blue) ex vivo by flow cytometry from single-cell suspensions of spleen tissues obtained from a human patient with invasive cancer. (B) Gene expression signature of purified human splenic GMP-like cells and of bone marrow GMPs, as previously reported (10). (C) Total counts of GMP-like cells in the spleen of patients with (n = 7) or without (n = 7) invasive cancer. (D) Total number of monocytes and neutrophils retrieved from spleens of the same patients. (E) Fate of human Lin− cKit+ splenic progenitor cells 5 d after injection into a nonobese diabetic SCID mouse bearing a lung carcinoma xenograft. Lin, lineage. (F) Table showing spleen weight and total numbers of splenic GMPs and monocyte- and neutrophil-like cells in tumor-bearing mice and in patients with cancer. The fold difference between species is also shown. Data are presented as the mean ± SEM. *P < 0.05.