Systemic endocrine instigation of indolent tumor growth requires osteopontin

Abstract

The effects of primary tumors on the host systemic environment and resulting contributions of the host to tumor growth are poorly understood. Here, we find that human breast carcinomas instigate the growth of otherwise-indolent tumor cells, micrometastases, and human tumor surgical specimens located at distant anatomical sites. This systemic instigation is accompanied by incorporation of bone-marrow cells (BMCs) into the stroma of the distant, once-indolent tumors. We find that BMCs of hosts bearing instigating tumors are functionally activated prior to their mobilization; hence, when coinjected with indolent cells, these activated BMCs mimic the systemic effects imparted by instigating tumors. Secretion of osteopontin by instigating tumors is necessary for BMC activation and the subsequent outgrowth of the distant otherwise-indolent tumors. These results reveal that outgrowth of indolent tumors can be governed on a systemic level by endocrine factors released by certain instigating tumors, and hold important experimental and therapeutic implications.

Figure 1. Primary Human Breast Carcinomas Instigate Growth of Distant Indolent Tumors

(A) Scheme of bilateral implantation system. Indolently growing HMLER-HR transformed cells (Responders) are implanted subcutaneously into one flank of host mice and either Matrigel control or vigorously growing tumor cell lines (Instigators) are implanted into the contralateral flank. (B) In vivo growth kinetics of responder cells when implanted contralaterally to either Matrigel (blue) or BPLER instigators (red); n = 5 per group. (C) Final mass of responders from (B) 9 weeks after implantation opposite either Matrigel or instigating BPLER tumors. Incidence of tumor formation is shown above data bars; data include injections not resulting in responding tumor growth. (D) In vivo growth kinetics of responders when injected 30 days after implantation of either Matrigel plugs (blue) or instigating BPLER tumor cells (red); n = 5 per group. (E) Average final mass of responding tumors from (D) recovered opposite either Matrigel or BPLER instigators; incidence of responder tumor formation is indicated above data bars. (F) Final mass of responding tumors recovered opposite Matrigel or indicated tumor cell lines 9 weeks after implantation. Incidence of responding tumor formation is indicated above bars; data include mice from three separate experiments. Mass of responders recovered opposite Matrigel is significantly different from those opposite 231 (p = 0.031) and BPLER (p = 0.039).

Figure 2. Bone Marrow-Derived Cells Are Incorporated into Responding Tumor Stroma

(A) Indicated tumor cells were injected into mice that had previously been engrafted with GFP⁺ BMCs. (B) Whole mount fluorescence photomicrographs to visualize GFP⁺ BMCs recruited to indicated tumors and control tissues 4 weeks after tumor cell injections; scale bar = 2 mm. Average mass of all tissues is indicated; n = 10 mice per group. (C) Average contribution of GFP⁺ cells as a percentage of total cells (flow cytometric analysis); n = 7 tumors/tissues per group. Responding tumors opposite BPLER instigators incorporated significantly more GFP⁺ BMCs than those opposite Matrigel (p = 0.039) and PC3 noninstigating tumors (p = 0.042); responders were not statistically different from their contralateral BPLER instigators; instigating BPLER tumors were not statistically different from noninstigating PC3 tumors.

Figure 3. Instigators Functionally Activate the Bone Marrow

(A) Experimental scheme for implantation of BMCs/responding tumor cell admixtures. (B) Average mass of resulting tumors 12 weeks after implantation of mixtures of responder cells with indicated BMCs. Tumor incidence is indicated above bars; data represent mean of four separate experiments. p = 0.012 comparing BM-I with responders alone; p = 0.015 comparing BM-I with BM-C; p = 0.001 comparing BM-I with BM-NI. (C) Hematoxylin and Eosin stain to visualize histopathology of resulting tumors/tissues; scale bar = 400 μm.

Figure 4. Instigating Tumors Perturb Bone Marrow Lin⁻/Sca1⁺/cKit⁺ (LSK) Cells while Sca1⁺/cKit⁻ Cells Are Incorporated into Responding Tumors

(A) Flow cytometric analysis of LSK cells in the bone marrow of various tumor-bearing mice; n = 4 per group. (B) Quantification of Sca1⁺/cKit⁺ and Sca1⁺/cKit⁻ cells as percentage of total GFP⁺ cells in responding tumors recovered opposite instigating BPLER tumors (red) or noninstigating PC3 tumors (gray); n = 7 per group. (C) Merged photomicrographs of indicated responding tumors stained for GFP (green), Sca1 (red), and cell nuclei (blue); GFP⁺/Sca1⁺ cells appear yellow (arrows); scale bar = 25 μm.

Figure 5. Tumor-Derived Osteopontin (hOPN) Is Necessary for Systemic Instigation

(A) Concentration of hOPN in plasma of mice 9 weeks after injection of indicated cells; n = 14 for responder group; n = 5 for all other groups. Plasma OPN was significantly elevated in mice bearing BPLER (p = 0.02) and 231 (p = 0.01) tumors compared with plasma from mice injected only with responder cells. (B) The left shows in vivo growth kinetics of instigating MDA-MB-231 (parental), control 231 cells expressing shRNA against Luciferase (shLucif), and 231 derivatives expressing shRNAs against hOPN; n = 9 for parental group; n = 5 for all other groups. The right shows in vivo growth kinetics of the responding cells injected opposite the indicated tumor cell lines. (C) Flow cytometric analysis of LSK cells in the bone marrow of mice with indicated bilateral tumors; n = 4 per group. (D) Admixtures of responder cells with BMCs from mice bearing 231 instigators yielded tumors that were significantly larger than those resulting from admixtures of BMCs from mice bearing shOPN 1 tumors (p = 0.030), shOPN 5 tumors (p = 0.047), or responder cells alone (p = 0.026); n = 4 per group.

Figure 6. Instigating Tumors Instigate Outgrowth of Disseminated Lung Metastases

(A) Model of metastatic systemic instigation: GFP⁻ instigating tumor cells (1^° tumor) or Matrigel control are subcutaneously (s.c.) injected into both flanks of host mice while weakly metastatic GFP⁺ tumor cells are injected intravenously (i.v.). (B) Average numbers of micrometastatic lung foci after concurrent injection of weakly metastatic 231 cells i.v. and either instigating BPLER tumors or Matrigel control plugs s.c.; n = 4 mice per group. (C) Whole-mount fluorescent photomicrographs of 231+GFP responder lung foci in mice bearing either Matrigel control plugs (a, b) or GFP⁻ instigating 1^° tumors (c, d); scale bar = 0.5 mm. (D) Immunohistochemical staining of lung sections for GFP⁺ responder cells (red); nuclei stain, blue; scale bar = 200 μm. (E) For (E) and (F), mice received s.c. injections of either Matrigel, GFP⁻ parental 231 cells or GFP⁻ shOPN 231 cells and i.v. injection of weakly metastatic 231+GFP cells. Graph depicts numbers of micrometastatic 231+GFP lung foci from each mouse after 4 weeks; lines denote average number of foci per group; average 1^° tumor burden is indicated. Whole-mount fluorescent photomicrographs depict micrometastatic foci (arrows); scale bar = 0.5 mm. (F) Numbers of macrometastatic 231+GFP foci counted by eye in the lungs of each mouse. Whole-mount images of lungs are shown with corresponding fluorescent images; scale bar = 2 mm.

Figure 7. Human Colon Tumor Surgical Specimen Responds to Systemic Instigation

(A) The left shows human colon tumor segments (“colon responder” blue line) dissected from a single patient’s surgical specimen and implanted opposite Matrigel plugs (gray line). The center shows in vivo growth kinetics of colon responder tumor segments (blue line) implanted opposite instigating BPLER breast carcinoma cells (red line). The right shows that neither the human colon tumor segments (blue line) nor the HMLER-HR breast responder cells (orange line) are able to grow when implanted opposite one another. n = 3 mice per group. (B) Hematoxylin and Eosin stain of a colon responder recovered opposite a breast instigator, scale bar = 50 μm. (C) Staining for Ki67 (brown) to reveal proliferating responding colon tumor cells implanted opposite BPLER breast instigators, scale bar = 50 μm. (D) Model of Systemic Instigation: Instigating tumors secrete osteopontin (OPN), which perturbs primitive hematopoietic cells in the host bone marrow; cells in the bone marrow are functionally activated prior to mobilization into the circulation; release of activated bone marrow-derived cells into the circulation and their subsequent incorporation into distant responding tumor stroma serve to foster outgrowth of the once-indolent cells into growing adenocarcinomas.