Cancer cell-macrophage fusion increases tumor cell heterogeneity and reshapes their microenvironment

doi:10.21203/rs.3.rs-2062628/v1

Download PDF

Article

Cancer cell-macrophage fusion increases tumor cell heterogeneity and reshapes their microenvironment

https://doi.org/10.21203/rs.3.rs-2062628/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The trademark of cancer is the ability to evolve, which lays the grounds for progressive events such as metastasis and recurrence. Although genetic mutations and epigenetic changes have been implicated as the mechanisms, they don’t explain why many cancers develop leukocytic traits. Cell fusion between cancer and somatic cells, particularly macrophages, has been suggested as an alternative pathway for cancer cells to obtain new traits via acquiring exogenous genetic material. In this study, tumor-macrophage hybrid cells were generated, and two clones, both grew slowly yet with very different tumorigenicity, were selected for further study. Despite their very different abilities to form tumors in mice, both clones showed significant abilities to influence the tumor microenvironment. RNA-seq of the hybrid cell clones revealed the differential expression profiles of the hybrid cells that contributed to the biological behaviors. This study emphasizes the role of hybrid cells as potent environmental modifiers that aid tumor survival and evolution despite their minority status among the tumor cells. This study also provides an animal experimental platform to study cancer-myeloid fusion and a potential direction for novel therapeutic interventions.

Cancer progression requires that the tumor cells obtain new abilities to evolve in order to escape from immune attacks and therapeutic interventions, an ability usually achieved by mutations. Natural mutations due to environmental stress have been considered the main driving force for tumor heterogeneity. However, this process takes time and does not fully explain the speed of tumor evolution commonly associated with recurrence and metastasis ¹. It also cannot explain the observation that many cancer cells express leukocytic traits. Unlike mutations caused by environmental stress, cancer-somatic cell fusion allows the tumor cells to acquire exogenous genetic material through a non-mutational mechanism capable of generating diversity in the cancer genome within a relatively short time.

Although the ability to fuse with cancer cells is not limited to bone marrow-derived cells ^2–4, myeloid cells, namely macrophages, are regarded as the most common fusion partner ⁵. Most research shows that cancer-somatic cell fusion provides an opportunity to acquire novel properties to facilitate tumor progression, particularly when the fusion partners are myeloid cells. ^{6,7 4,8−10}. Nonetheless, induction of anti-tumor events has been reported, mostly when cancer cells fuse with mesenchymal cells or stem cells ^{11 3}.

As many cancers exhibit leukocytic traits, it is postulated that tumor-leukocytic cell fusion may be an important mechanism for cancer plasticity and metastasis ^{12 13 14,15 1}. In support of this notion, tumor-leukocyte hybrids have been demonstrated in animal studies and confirmed in patient specimens ^16,17. Clinical studies further show that cancer-leukocyte hybrids potentiate tumor heterogeneity and impact patient prognosis ^15,18.

Cell-cell fusion is a physiological process required for embryonic development and the formation of skeletal muscle, osteoclasts, and syncytiotrophoblasts ¹⁹. In adult tissue, heterotypic cell-cell fusion is rare, though it can be detected during tissue repair and chronic inflammation ^{20 21}. Chemotherapy and radiation, two major treatment modalities for cancer therapy, have been shown to promote this otherwise uncommon phenomenon ^22,23. By studying biopsies from patients receiving allogeneic stem cell transplants that later developed solid tumor metastases, Pawelek’s group used forensic genetics to document a high percentage of the metastatic tumor cells carrying donor and patient DNA sequences, indicating that hybrids were present and likely the source of metastasis ¹⁷. This finding suggests that tumor-macrophage fusion may not be as rare as once thought in specific clinical scenarios and underscores the potential role of tumor-macrophage hybrids in recurrence and metastasis, particularly after cancer treatment.

Despite many reports of cell fusion, few studies have provided an experimental system that followed the fusion products from generation through their evolution. In this study, we examined cloned cancer-macrophage hybrids for in vitro and in vivo analysis. They were injected into the syngeneic mice to demonstrate their tumorigenicity and abilities to influence the microenvironment. Our results show that while not all these hybrid cells were robustly tumorigenic, they consistently exhibited significant abilities to shape the microenvironment. Thus, tumor-myeloid hybrid cells are potent environmental modifiers and have important biological implications for cancer survival and evolution.

Spontaneous fusion between squamous cell carcinoma and myeloid cells

SCCVII/SF, a squamous cell carcinoma cell line derived from C3H mice, was examined for suitability for tumor-myeloid fusion studies. SCCVII/SF-H2B-GFP cells, which express a nuclear GFP signal, were grown in the dorsum of the syngeneic mice, and the resulting tumor was excised and processed for single-cell suspensions. Since hybrid cells derived from cancer-myeloid cell fusion should carry traits from both parental cells, flow cytometry was used to gate the tumor cells based on their GFP expression and myeloid markers CD45 and CD11b. ~1% of the GFP + cells expressed the myeloid markers, which reflected the fusion rate. An example of a 1.77% fusion rate is shown here (Fig. 1a, left panel). Since phagocytosis of SCCVII/SF-H2B-GFP cells by myeloid cells could produce “pseudo-hybrids,” which would have weak GFP from tumor cell remnants in the phagosomes of the myeloid cells, only cells with high levels of GFP and myeloid markers were gated for detection of true hybrids. The images of the hybrid cells were obtained with an imaging flow cytometry to document nuclear GFP and membranous CD11b (Fig. 1a, right panel). Such specific spatial distribution of the markers provided convincing evidence that fusion between SCCVII/SF and myeloid cells occurred in vivo.

Bone marrow macrophages (BMM) require M-CSF to survive and RANKL to fuse and form multinucleated osteoclasts ²⁴. The properties of these cytokines were exploited to promote fusion between BMM and tumor cells in vitro. When M-CSF and RANKL were added to the co-culture of BMM and SCCVII/SF-H2B-GFP, the resulting multinucleated giant cells expressed not only macrophage marker PU.1, but also GFP from the tumor cells. Notably, while the PU.1 + nuclei may or may not express GFP, all the GFP + nuclei in the giant cells expressed PU.1 (Fig. 1b), indicating the occurrence of nuclear fusion between the two cell types. These multinucleated giant cells were able to survive for more than 1 month with no sign of death as long as the media was replenished. Multinucleated giant cells formed by mouse BMM mono-cultures in the same media conditions usually died after 10 days.

SCCVII/SF-BMM hybrid cell clones exhibit variable tumorigenicity

Most studies show that fusion with macrophages promotes tumor aggression ^{6,7 4,8−10}. However, tumor cells fused with mesenchymal cells or stem cells have been reported to induce anti-tumor events ^{11 3,25}. To study the outcome of fusion between SCCVII/SF and macrophages, we used the same culture conditions described above with some modifications to generate the hybrid cells for subsequent studies. Briefly, BMM were retrovirally transduced to express antibiotic resistance and cytoplasmic GFP before being co-cultured with SCCVII/SF in the presence of M-CSF and RANKL. After the culture formed multinucleated giant cells, the cytokines were withdrawn and antibiotic selection was started to enrich the hybrid cells, which were later cloned. The surviving cells acquired GFP expression and antibiotic resistance from the BMM but had only one nucleus and could survive indefinitely without the supporting cytokines.

Six clones of hybrid cells were injected into the posterior dorsum of mice. SCCVII/SF-GFP, which was SCCVII/SF transduced with the GFP retrovirus used for the hybrid cells, was used as the control. The hybrid cell clones showed variable growth abilities in mice. Clones D2 and D3 were selected for further characterization as they produced tumors that were significantly larger or smaller/not present, compared to those produced by the control cells (Fig. 2a). Most of the mice that received the D2 cells developed significantly larger tumors even though some were underestimated due to loss of tumor mass by ulceration and/or animal premature death (Fig. 2b). Unexpectedly, ~ 16% of the mice injected with the D2 cells failed to develop grossly visible tumors within the study period of 3 weeks. In contrast, the D3 cells either failed to form tumors or produced very small tumors even when the study period was extended. The D3 line never produced large tumors. The growth disparities of these two hybrid cell clones were also seen when grown in the immunocompromised NSG mice, where they grew faster, forcing the study period to be shortened to 16 days to avoid major adverse events (Fig. 2c & d).

SCCVII/SF-BMM hybrid cells have a low growth rate in vitro but exhibit increased ability to recruit stromal cells in vivo

The hybrid cell clones D2 and D3 were analyzed by a growth curve experiment in cell culture. Both cells exhibited slower growth and prolonged doubling times (D2 ~ 23.97 hours, D3 ~ 28.31 hours) than the control cells (~ 16.79 hours) after culturing for four days (Fig. 3a). The D3 cells exhibited the slowest growth, consistent with their diminished abilities to generate tumors in mice. Paradoxically, the D2 cells also showed a slow growth kinetic despite their ability to generate large tumors in mice. Unlike cell culture, the in vivo environment is complex, as non-tumor elements could also influence tumor growth and behavior. The compositions of the D2 and control SCCVII/SF-GFP tumors were examined by flow cytometry performed on tumor cell suspensions. D3 cells were not included as they either failed to form tumors or formed tumors too small to generate sufficient numbers of cells for flow cytometry. Based on the GFP expression of the D2 and control cells, the tumor cells (GFP+) were gated from the non-tumor stromal cells (GFP-) by flow cytometry.

The results showed that the D2 tumors exhibited a higher percentage of GFP- cells (Fig. 3b, left panel) than the control. As a result, a significantly lower ratio between the tumor and stromal cells, expressed as GFP+/GFP- was observed (mean value of control vs. D2: ~0.73 vs. ~0.26) (Fig. 3b, right panel), indicating that only a small portion of the D2 tumor was composed of the D2 hybrid cells. In other words, despite the large size, the bulk mass of the D2 tumor was formed not by the tumor cells but by the stromal cells. Most of the stromal cells were expected to be fibroblasts since they are the most abundant cell type in solid tumors ²⁶.

Flow cytometry was also used to examine the cell cycle distributions of the cells in the tumors. The results showed that the D2 tumor had a significantly higher percentage of tumor cells (GFP+) in the G2/M phase while a lower percentage in the G1/G0 phase compared to the control tumors (mean value of control vs. D2: G2/M: ~24% vs. ~51%. G1/G0: ~63% vs. ~41%) (Fig. 3c & d). The stromal cells (GFP-) from both tumors demonstrated similar cell cycle distributions, with the majority (~ 84%) of the cells were in G0/G1 while only a minority (~ 5%) transitioned through the G2/M phase (Supplementary Fig. S1).

Fusion with BMM increases SCCVII/SF’s ability to recruit immune cells

Tumor-infiltrating immune cells play essential roles in cancer development, and the most abundant immune cells in solid tumors are macrophages, which are present through all stages of tumorigenesis ²⁷. Macrophages can polarize to M1 or M2, a tumoricidal or immunosuppressive phenotype, respectively ²⁸. In established tumors, tumor-associated macrophages (TAM) often polarize to M2 to negatively modulate the tumor immune microenvironment. T regulatory cells (Treg), cytotoxic T cells (Tc), and NK cells are also important tumor-infiltrating immune cells. Similar to macrophages, Treg cells also have an immunosuppressive function.

Flow cytometry was used to gate the tumor cell suspensions with GFP and macrophage markers (CD45 + CD11b + F4/80+) to determine the ability of the tumor cells to recruit macrophages (Fig. 4a), expressed as the TAM/GFP + ratio. The results showed that the D2 tumor had a significantly higher ratio of TAM/GFP+ (mean value of control vs. D2: ~ 0.7 vs. 1.8), indicating that the D2 cells were more capable of recruiting TAM than the control cells (Fig. 4b, upper panel). There were no differences between the two tumors in the percentages of TAM in total leukocytes (CD45+) (Fig. 4b, lower panel). As the D3 tumor was too small to be analyzed by flow cytometry, F4/80 IHC was used to stain TAM in the tumor sections to provide a qualitative assessment. The result showed that the D3 tumor, similar to the D2 tumor, also harbored a substantial number of TAM (Fig. 4c).

Furthermore, TAM polarization was examined by flow cytometry with macrophage markers for M1 (CD11b + F4/80 + iNos+) and M2 (CD11b + F4/80 + Arg1+) (Fig. 4d, 4e). The results showed ~ 62% (mean value) of TAM in the D2 tumor and only ~ 26% of TAM in the control tumors polarized to M2 (Fig. 4e upper panel). In addition, M2 macrophages constituted ~ 14% of the total cell population in the D2 tumor while only 5% was observed in the control tumor (Fig. 4e lower panel). In contrast, M1 macrophages were not significantly present, and their numbers were not different between the two tumors (Fig. 4e lower panel).

The T cell compositions in the tumors were also analyzed by flow cytometry, including Treg (CD4 + CD25 + FoxP3), Tc (CD45 + CD8+), and Th (CD45 + CD4+) (Figs. 5a, 6a). Similar to the TAM results, the D2 tumor also exhibited an increased ability to recruit Treg and Tc, with higher numbers of these immune cells per total cells (Figs. 5b, 6b, top and middle panels). The D2 and the control tumor showed no differences in the percentages of Treg/total helper T cells (CD4+) (Fig. 5b, lower panel). In comparison, the D2 tumor had a slight increase in the percentage of Tc/total leukocytes (CD45+) (Fig. 6b, lower panel). As the D3 tumor was too small to be analyzed by flow cytometry, FoxP3 or CD8 IHC was used to provide a qualitative assessment of Treg or Tc in the tumor. The results showed that the D3 tumor, similar to the D2 tumor, also harbored higher numbers of Treg and Tc compared to the control (Fig. 5c & 6c).

The numbers of tumor-infiltrating Treg and Tc cells were less than 1/10 of the number of M2 macrophages in both tumors (Figs. 5B, 6B, middle panel). There was a scant presence of NK cells (not detectable), expected for SCC-VII/SF-derived tumors, which exhibit NK cell suppression due to high levels of TGF-β ^29,30.

Tumor angiogenesis

Tumor growth is highly dependent on neovascularization. In this study, the tumors derived from the hybrid cells recruited high numbers of M2 macrophages. Since M2 macrophages are important pro-angiogenic cells ³¹, we investigated whether their presence impacted the micro-vessels in the tumors. CD31 IHC staining in endothelial cells showed the micro-vessels in the tumors (Fig. 7a, top panel). The images were segmented (Fig. 7a, lower panel) before being analyzed by the Microvessel-Segmentation MATLAB plugin ³².

The results show that tumors formed by the hybrid cells, D2 and D3, exhibited higher microvessel densities (MVD) and thinner micro-vessel wall thickness (MVWT) than the control tumor (Fig. 7b). There were no significant differences in other angiogenesis parameters, such as blood vessel sizes and percentage of vessel area.

Differential RNA expression

RNA-seq was used to investigate the mechanisms responsible for the behavioral differences between the hybrid and control cells. Ingenuity Pathway Analysis (QIAGEN) was used to generate differential RNA expression profiles, and the unfiltered results are provided in supplementary Excel file “Upstream regulatory analysis of RNA-seq.xlsx”. The file contains: Sheet 1 − D2 hybrid cells vs. SCCVII/SF-GFP; sheet 2 − D3 hybrid cells vs. SCCVII/SF-GFP; sheet 3 − D2 hybrid cells vs. D3 hybrid cells.

Table 1 lists the representative upstream regulatory pathway results relevant to the current study. Compared with the control SCCVII/SF-GFP cells, the D2 cells exhibited increased activation in pathways involving five cytokines - tumor necrosis factor α (TNFα), interleukin (IL)-1α & β, interferon γ (IFNγ), and oncostatin M (OSM), along with two growth factors - epidermal growth factor (EGF) and vascular endothelial growth factor (VEGF)-α. All five cytokines are pro-inflammatory, and upregulation of their activities likely contributed to the increased numbers of inflammatory cells in the D2 tumor. EGF is a mitogen for many cell types, including fibroblasts and epithelial cells ^33,34 while TNFα also plays an important role in the maintenance and function of cancer-associated fibroblasts ^26,35. Thus, upregulation of TNFα and EGF pathways likely contributed to the significant presence of non-tumor cells that resulted in large sizes of the D2 tumor.

Table 1

Representative upstream regulators that influence the biological behaviors of the hybrid cells. Only pathways of cytokines, growth factors, and major cell cycle molecules are listed. The pathways predicted to be activated are listed on the left side of the z-score column, while those predicted to be inhibited are listed on the right side. A z-score ≥ 2 predicts an activation state, while a z-score ≤ -2 predicts an inhibitory state of the pathway.
RNA-Seq	Category	Upstream Regulator	Activation z-score	p-value of overlap
D2 vs SCC	Cytokine	TNF	3.663	1.77E-11
		IL1b	2.837	0.000178
		IFNg	2.811	0.0312
		OSM	2.211	0.0156
		IL1a	2.007	0.00542
	Growth factor	VEGFa	2.771	0.000839
	Growth factor	EGF	2.585	0.0374
	Cell Cycle	TP53	2.056	0.000185
D3 vs SCC	Cytokine	TNF	2.018	1.85E-11
	Growth factor	VEGFa	2.244	0.00153
		FGF7	-2.169	0.00779
		AREG	-3.352	9.53E-06
	Cell Cycle	CDKN1A	2.532	0.0000216
		TP53	2.928	1.34E-12
		CDKN2A	2.07	0.00735
D2 vs D3	Cytokine	IL6	2.021	0.0126
	Growth factor	AREG	2.917	8.3E-10
	Growth factor	FGF7	2.028	0.0114
	Cell Cycle	CDKN1A	-2.266	0.000000572
		TP53	-3.37	8.49E-14
		CCNK	2.366	0.00246

The D3 cells also exhibited upregulation of the TNFα pathway, which played important roles in diverse cellular events in tumor, including recruitment and modulation of TAM as well as angiogenesis ³⁵. The D3 cells had decreased signaling from fibroblast growth factor (FGF)-7 and amphiregulin (AREG), the biological implications of which are unclear.

VEGF-α is an essential factor for angiogenesis ³⁶. Upregulation in the VEGF-α pathway increases proliferation and migration of vascular endothelial cells. This process could be enhanced by inflammation, which likely contributed to the increase in tumor vasculature observed in the D2 and D3 tumors.

The RNA-seq results also showed that both D2 and D3 cells exhibited TP53 and cell cycle checkpoint activation. This suggests that during the process of nuclear merging, cells lost chromosomes and developed aneuploidy, leading to TP53 activation ^37,38. These events were likely responsible for the altered growth kinetics and cell cycle distribution. When compared with the control, the D3 cells showed induction of cyclin-dependent kinase inhibitor 1A (CDKN1A) and cyclin-dependent kinase inhibitor 2A (CDKN2A). When compared between the D2 and D3 cells, the D3 cells exhibited a higher level of TP53 activation and consequently CDKN1A induction, which was presumably associated with their significant presence in the G0/G1 phase (Fig. S1b) and minimal tumorgenicity (Fig. 2). In contrast, the D2 cells exhibited cyclin K (CCNK) induction, which presumably maintained the genome integrity and allowed for cell cycle progression ³⁹.

This study examined the biological implications of fusion between cancer and myeloid cells. The hybrid cancer-BMM cells were generated in vitro and began as multinucleated cells. At this stage, the cell markers from the parental cells were readily detectable. These hybrid cells had minimal chance of being successful in proliferation due to their heterokaryon status, which leads to multipolar division during chromosome segregation ⁴⁰. For proliferation to be successful, the heterokaryon-type hybrids need to merge the nuclei into one common nucleus and become synkaryons. Antibiotic selection was used as the mean to enrich the synkaryon-type hybrids in the culture and subsequently cloned for individual characterization in this study. Chromosome loss often occurs during the transition from heterokaryon to synkaryon, which leads to aneuploidy, activation of TP53 and alterations in cell cycle control ^41,42. Not surprisingly, both hybrid cell clones showed TP53 activation (Table 1) and had lost common myeloid markers. After cell-cell fusion, the loss of parental cell markers due to genetic instability is expected to make identifying fusion hybrids a great challenge. Given that markers of cell origins are often lost during this process, the frequency of tumor-myeloid fusion may be underestimated in patients.

Both hybrid cell clones exhibited slow growth kinetics (Fig. 2a). Unexpectedly, while the D3 cells showed minimal tumorigenicity, the D2 cells formed large tumors in mice. The drastic differences in tumorgenicity were observed in syngeneic C3H and in immunocompromised NSG mice, indicating that intrinsic mechanisms, not host immunity, were responsible for the growth differences. Compared with the D2 cells, the D3 cells had decreased activation of CCNK but increased activation of TP53, CDKN1A, and CDKN2A (Table 1). High levels of CDKN1A and CDKN2A inhibit cell proliferation, which contributed to the D3 cells’ significant presence in the G0/G1 phase (Supplementary Fig. S2) and likely the main cause of diminished tumorigenicity in mice. Because bone marrow-derived mesenchymal cells were present in the co-cultures used to generate the hybrid cells, the possibility that they fused with the SCCVII/SF-BMM hybrids and played roles in the tumorigenicity of the cloned hybrid cells, despite very low due to their scant presence, cannot be completely ruled out.

Despite that the D2 cells often produced large tumors, ~ 16% of the mice failed to develop grossly visible tumors within the study period. This suggests that some of the D2’s daughter cells had chromosomal instability that triggered catastrophic events, although the frequency was much less than that of the D3 cells. Such events are expected to result in aberrant mitosis and inhibit tumor growth.

This study showed that tumor-BMM fusion has the potential to greatly increase the composition of non-tumor cells, which potentially provided diversified pathways to gain survival advantages when combating an ever-changing environment. Hybrid cells also showed an increased expression of pro-inflammatory cytokines, enhancing the recruitment of inflammatory cells to the tumor environment where most of them were re-educated to develop modulatory functions to support tumor progression.

Our results indicate that tumor-myeloid fusion has important biological implications for cancer survival and evolution. While the hybrid cells gained functions in pro-tumoral pathways, they also inadvertently triggered cell cycle checkpoints and altered their growth kinetics. Despite this, the hybrid cells gained some advantages over the original tumor cells and were potent environmental modifiers. This study underscores the potential supportive roles that hybrid cells play in altering the tumoral microenvironment. The hybrid cells could form and die and do not have to become the dominant clone. Yet, their ability to modulate the tumor environment provides an important asset for tumors while waiting for the emergence of rare cells. Knowledge of tumor-myeloid fusion may help in the development of novel clinical interventions to control cancer progression.

In conclusion, hybrid cells are potent environmental modifiers. Their presence could provide important advantages, even a temporal one, over other tumor cells in promoting tumor survival and allowing the escape of rare cells to become the source of tumor recurrence or metastasis.

Animal use and ethical approval

The animal studies were approved (ARC-2017-071) by the UCLA Institutional Animal Care and Use Committee. All experiments were performed in accordance with relevant guidelines and regulations. The reporting in the manuscript follows the recommendations in the ARRIVE guidelines ⁴³. C3H/HeNCrl mice were obtained from Charles River (Hollister, CA, USA), and NOD-scid IL2Rgammanull (NSG) mice were obtained from Radiation Oncology, UCLA (originally from Jackson Laboratory, Bar Harbor, ME, USA). Mice of both sexes aged 2–4 months were used in the study. Mice were maintained under the care of the UCLA Division of Laboratory Animal Medicine.

Tumor production, tissue harvest and processing

For tumor formation, the mice received a subcutaneous injection of SCCVII/SF-derived cells (10⁶) suspended in PBS (50 µL) into the dorsal flank. Tumor length (L) and width (W) was assessed daily with a caliper, and tumor volume (V) was calculated based on the formula: V = (W² × L)/2 ⁴⁴. Twenty-four hours before euthanasia, the mice received an intraperitoneal injection of BrdU (100 µL 10 mg/mL) to label the replicating cells. Mice were euthanized on the indicated days by CO2 inhalation followed by cervical dislocation. The injection site tissue (with or without tumor formation) was excised, weighed, and processed for the subsequent studies.

For histological analysis, the harvested tumors were fixed in formalin (10%) and paraffin embedded to form tissue blocks. Tissue sections were cut, dewaxed and rehydrated before staining. For flow cytometry, the excised tumor was placed in serum-free RPMI-1640 (4 ml) and cut into small pieces. Final concentrations of liberase DL (0.28U/ml, Roche), liberase TL (0.28U/ml, Roche), and DNase I (80 ug/ml) were added to digest the tissue in a 37°C shaking water bath. After 45 min, the tissue suspension was pipetted multiple times through a 5 ml pipette to break up the cellular clumps. PBS (10 ml) containing FCS (2%) was added to stop the digestion, and the suspension was passed through a 100 µm cell strainer to generate a single-cell suspension. The cells were then spun down, resuspended in RPMI containing FCS (90%) and DMSO (10%), aliquoted, and stored in liquid nitrogen. Before usage, the cells were thawed and resuspended in PBS with FCS (2%).

Plasmid construction and retrovirus production

pMSCVpuro, pMSCVneo and pLEGFP-N1 plasmids were purchased from Takara Bio (Mountain View, CA, USA). pMSCVpuro-EGFP was constructed by subcloning the EGFP-N1 sequence from pEGFP-N1 plasmid (XhoI and NotI/Klenow) into pMSCVpuro (XhoI and HpaI). For pLEGFP-mH2B construction, primers 5’-ATACTCGAGATGCCTGAGCCTGCGAAG-3’ and 5’-CGCGGATCCTTCTGGTCTTTTGAATC-3’ were used to amplify the mouse histone H2B sequence. The PCR product was digested with XhoI and BamHI before insertion into the compatible cloning site in pLEGFPN1. This plasmid expresses a histone H2B-GFP protein in the nucleus. For retrovirus infection, supernatant from the retrovirally transfected 293FT cells was used to infect the cells as previously described ⁴⁵.

Cell-cell fusion and cloning of hybrid cells

The SCCVII/SF cell line was originally established by Dr. Herman D. Suit at Harvard University and has been widely used in C3H mice as an immunocompetent model for head and neck cancer ^46,47. This cell line was a gift from Dr. Mae St. John, UCLA Head and Neck Surgery. The cells were cultured in RPMI-1640 media supplemented with FBS (10%) and maintained in a 5% CO₂ incubator. pLEGFP-mH2B retrovirus was used to transduce SCCVII/SF to generate the SCCVII/SF-H2B-GFP cell line. For the fusion experiment shown in Fig. 2, SCCVII/SF-H2B-GFP was co-cultured with bone marrow macrophages (BMM) in MEM-α media supplemented with FBS (10%), M-CSF (50 ng/mL) and RANKL (50 ng/mL) to support BMM growth and promote cell-cell fusion. Bone marrow was harvested from long bones of euthanized mice ⁴⁵.

For cloning hybrid cell lines, pMSCVpuro-EGFP-transduced BMM were co-cultured with SCCVII/SF in MEM-α containing M-CSF and RANKL as described above. BMM and SCCVII/SF monocultures were used as the controls. The co-culture was withdrawn from M-CSF and RANKL after three days, and the media was switched to RPMI-1640 containing FBS (10%) and puromycin (5 µg/ml). No viable cells were seen in BMM or SCCVII/SF monocultures after one week. Only the SCCVII/SF cells that acquired the puromycin-resistant gene via fusion with the pMSCVpuro-EGFP-transduced BMM could survive in the co-culture. The fused cells also acquired the EGFP gene. Six clones were tested for tumorgenicity in mice. Cell lines used were within ten passages.

Flow cytometry

For cell cycle analysis, the cells were fixed in formaldehyde (4%) for 1 hour, permeabilized with TritonX-100 (0.5%) for 20 min, washed, and resuspended in PBS (100 µl) containing CaCl₂ (0.5 mM), MgCl₂ (2.5 mM), DNase I (10 units), and RNase A (40 µg). Cells were incubated for digestion at 37°C. After 1 hour, cells were washed, resuspended in PBS-FCS-Tween 20 solution (100 µl) containing FCS (2%) and Tween-20 (0.05%), and stained with APC-conjugated BrdU antibody overnight at 4°C. Cells were washed before being resuspended in PBS containing 7-Aminoactinomycin D (7-AAD) (1.25 µg/ml) and then stained for 20 min. The cells were washed and resuspended in PBS (0.5 ml) for flow cytometry analysis.

For immune cell analysis, the fixed cells were resuspended in the PBS-FCS-Tween-20 solution (100 µl) and incubated with an a-mouse CD16/32 antibody at room temperature for 30 min to block the Fc receptors. The cells were then stained with fluorochrome-conjugated antibodies against various immune markers overnight at 4°C. The stained cells were resuspended in PBS (0.5 ml) before being analyzed by flow cytometry.

For imaging flow cytometry, the fixed cells were stained with PE-conjugated CD11b antibody overnight at 4°C. The cells were washed and permeabilized with TritonX-100 (0.5%) before DNA staining with 4,6-diamidino-2-phenylindole (DAPI) (300 nM). After 30 min, the cells were washed again and resuspended in PBS (150 µl) before being analyzed by the ImageStream system equipped with IDEAS software (Amnis, Austin, TX, USA).

Immunofluorescence and immunohistochemistry

For immunofluorescence staining, cultured cells were fixed in paraformaldehyde (2%) and permeabilized with TritonX-100 (0.1%). Normal goat serum (2.5%) was used for blocking. Cells were stained overnight at 4°C with a primary antibody, and the signal was then detected by a corresponding Alexa Fluor-conjugated secondary antibody. DAPI was used to stain the nuclei.

For bright field staining, the hydrated sections were subjected to standard hematoxylin and eosin (H&E) or immunohistochemical (IHC) stains (ImmPRESS reagent, Vector Laboratories, Burlingame, CA, USA). Appropriate positive and negative controls were included.

In vitro cell proliferation

The cells were plated in quadruplicate onto a 96-well plate (200 cells/well) and allowed to attach overnight. Cell proliferation analysis was carried out daily for four days with a CellTiter-Glo luminescent cell viability assay kit (#G7570, Promega, Madison, WI, USA), following the manufacturer’s instructions. The luminescence values were used to plot against the time for cell growth curves. The cell doubling time was estimated by an online Cell Doubling Time Calculator (https://www.omnicalculator.com/biology/cell-doubling-time).

Image acquisition and analysis of tumor vascular network

The CD31 IHC-stained tumor sections were scanned with a Leica Aperio AT2 scanner under 40X to generate the whole slide images. Leica Aperio ImageScope software was used to visualize and capture the most vascularized areas (“hot spots”) within the tumor. Granulation tissue was not considered a “hot spot” and was excluded from the analysis. Depending on the size of the tissue section, images of 1 to 4 non-overlapping hot spots were acquired by ImageScope under the 10X zoom setting and exported to MATLAB platform for automatic morphometric analysis. The images were segmented and analyzed by the Microvessel-Segmentation MATLAB plugin ³². The microvessel number, vascular area, and vessel wall thickness were automatically quantified. The microvessel density (MDV), expressed as the number of microvessels per mm², was calculated.

Antibodies

Anti-CD8a (#14-0808-80), anti-CD49b/ITGA2 (clone DX5. #14-5971-85), anti-FOXP3 (#14-5773-82), anti-F4/80 (#14-4801-82), anti-BrdU-APC (#17-5071-42), anti-iNOS-APC (#17-5920-82), anti-Arginase 1-APC (17-3697-82), and Alexa Fluor-conjugated secondary antibodies (# A-11011) were purchased from ThermoFisher Scientific (Waltham, MA, USA). Anti-CD4-APC (#100412), anti-CD25-PerCP/Cy5.5 (#10191), anti-CD45-APC (#103112), anti-CD8-APC (#100712), anti-F4/80-PerCP/Cy5.5 (#123128), anti-CD11b-PE (#101208), anti-FOXP3-PE (#126404), and anti-CD16/32 (#101302) were purchased from BioLegend. Anti-CD31 (#ab28364) was purchased from Abcam (Boston, MA, USA). Anti-PU.1 (#2266) was purchased from Cell Signaling Technology (Danvers, MA, USA).

RNA-seq

RNA-seq was conducted by the UCLA Technology Center for Genomics and Bioinformatics. Briefly, total RNA was isolated from replicates of cells, followed by mRNA enrichment, reverse transcription to generate cDNA, end repair to generate blunt ends, A-tailing, adaptor ligation, and PCR amplification. Different adaptors were used for multiplexing samples in one lane. Libraries for RNA-Seq were prepared with KAPA Stranded RNA-Seq Kit (# KR1151, Roche, Wilmington, MA, USA). The data were sequenced on Illumina HiSeq 3000 for a single read 50 run. Data quality check was done on Illumina SAV. Demultiplexing was performed with Illumina Bcl2fastq2 v 2.17 program. The reads were mapped to the latest UCSC transcript set using Bowtie2 version 2.1.0 ⁴⁸ and the gene expression level was estimated using RSEM v1.2.15 ⁴⁹. TMM (trimmed mean of M-values) was used to normalize gene expression. Differentially expressed genes were identified using the edgeR program. Using Fisher’s exact test, genes showing altered expression with p < 0.05 and more than 1.5-fold changes were considered differentially expressed. The datasets are available in the Gene Expression Omnibus (GEO) repository https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE209989. Ingenuity Pathway Analysis (QIAGEN) was used for data analysis on the upstream regulatory pathways, and the unfiltered results are provided as a supplementary Excel file “Upstream regulatory analysis of RNA-seq.xlsx.”

Statistical analysis

Data were analyzed using an unpaired two-tailed Student’s t-test, except for RNA-seq (see above). p ≤ 0.05 is considered significant. Software R was used to generate the box-and-whisker and scatter plots displaying individual data points and the distribution through their quartiles.

ACKNOWLEDGMENTS

We thank Nathaniel Phillips at UCLA Technology Center for Genomics and Bioinformatics for assistance with RNA-seq analysis.

Author contributions

ML and YLL conceived the project. YLL supervised the project. ML and YLL performed experimental work. ML, JRB, SM and YLL contributed to the data analysis. YLL wrote the manuscript with input from other authors.

Data availability

The RNA-seq datasets generated and/or analyzed during the current study are available in the GEO repository https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE209989. The results of upstream regulatory pathway analyses are provided as a supplementary Excel file, “Upstream regulatory analysis of RNA-seq.xlsx.” All other data that support the findings of this study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Shabo, I. et al. Macrophage traits in cancer cells are induced by macrophage-cancer cell fusion and cannot be explained by cellular interaction. BMC Cancer 15, 922 (2015).
Powell, A.E. et al. Fusion between Intestinal epithelial cells and macrophages in a cancer context results in nuclear reprogramming. Cancer Res 71, 1497-505 (2011).
Wei, H.J. et al. FOXF1 mediates mesenchymal stem cell fusion-induced reprogramming of lung cancer cells. Oncotarget 5, 9514-29 (2014).
Xu, M.H. et al. EMT and acquisition of stem cell-like properties are involved in spontaneous formation of tumorigenic hybrids between lung cancer and bone marrow-derived mesenchymal stem cells. PLoS One 9, e87893 (2014).
Vignery, A. Macrophage fusion: are somatic and cancer cells possible partners? Trends Cell Biol 15, 188-93 (2005).
Rachkovsky, M. et al. Melanoma x macrophage hybrids with enhanced metastatic potential. Clin Exp Metastasis 16, 299-312 (1998).
Rachkovsky, M. & Pawelek, J. Acquired melanocyte stimulating hormone-inducible chemotaxis following macrophage fusion with Cloudman S91 melanoma cells. Cell Growth Differ 10, 517-24 (1999).
Chakraborty, A.K., de Freitas Sousa, J., Espreafico, E.M. & Pawelek, J.M. Human monocyte x mouse melanoma fusion hybrids express human gene. Gene 275, 103-6 (2001).
Laberge, G.S., Duvall, E., Haedicke, K. & Pawelek, J. Leukocyte(-)Cancer Cell Fusion-Genesis of a Deadly Journey. Cells 8(2019).
Manjunath, Y. et al. Tumor-Cell-Macrophage Fusion Cells as Liquid Biomarkers and Tumor Enhancers in Cancer. Int J Mol Sci 21(2020).
Chuprin, A. et al. Cell fusion induced by ERVWE1 or measles virus causes cellular senescence. Genes Dev 27, 2356-66 (2013).
Aichel, O. & Roux, W. Über Zellverschmelzung mit Qualitativ Abnormer Chro- mosomenverteilung als Ursache der Geschwulstbildung. About cell fusion with qualitatively abnormal. chromosome distribution as cause fortumor formation. 92–111 (Wilhelm Engelmann, Leipzig, Germany, 1911).
Pawelek, J.M. Tumour-cell fusion as a source of myeloid traits in cancer. Lancet Oncol 6, 988-93 (2005).
Friedl, P. Cell fusion: new mechanisms of plasticity in cancer? Lancet Oncol 6, 916-8 (2005).
Dietz, M.S. et al. Relevance of circulating hybrid cells as a non-invasive biomarker for myriad solid tumors. Sci Rep 11, 13630 (2021).
Chakraborty, A.K. et al. A spontaneous murine melanoma lung metastasis comprised of host x tumor hybrids. Cancer Res 60, 2512-9 (2000).
LaBerge, G. et al. A melanoma patient with macrophage-cancer cell hybrids in the primary tumor, a lymph node metastasis and a brain metastasis. Cancer Genet 256-257, 162-164 (2021).
Gast, C.E. et al. Cell fusion potentiates tumor heterogeneity and reveals circulating hybrid cells that correlate with stage and survival. Sci Adv 4, eaat7828 (2018).
Iosilevskii, Y. & Podbilewicz, B. Programmed cell fusion in development and homeostasis. Curr Top Dev Biol 144, 215-244 (2021).
Johansson, C.B. et al. Extensive fusion of haematopoietic cells with Purkinje neurons in response to chronic inflammation. Nat Cell Biol 10, 575-83 (2008).
Davies, P.S., Powell, A.E., Swain, J.R. & Wong, M.H. Inflammation and proliferation act together to mediate intestinal cell fusion. PLoS One 4, e6530 (2009).
Nygren, J.M. et al. Myeloid and lymphoid contribution to non-haematopoietic lineages through irradiation-induced heterotypic cell fusion. Nat Cell Biol 10, 584-92 (2008).
Yan, B., Wang, J. & Liu, L. Chemotherapy promotes tumour cell hybridization in vivo. Tumour Biol 37, 5025-30 (2016).
Feng, X. & Teitelbaum, S.L. Osteoclasts: New Insights. Bone Res 1, 11-26 (2013).
Duelli, D. & Lazebnik, Y. Cell fusion: a hidden enemy? Cancer Cell 3, 445-8 (2003).
Sahai, E. et al. A framework for advancing our understanding of cancer-associated fibroblasts. Nat Rev Cancer 20, 174-186 (2020).
Noy, R. & Pollard, J.W. Tumor-associated macrophages: from mechanisms to therapy. Immunity 41, 49-61 (2014).
Jayasingam, S.D. et al. Evaluating the Polarization of Tumor-Associated Macrophages Into M1 and M2 Phenotypes in Human Cancer Tissue: Technicalities and Challenges in Routine Clinical Practice. Front Oncol 9, 1512 (2019).
Dasgupta, S., Bhattacharya-Chatterjee, M., O'Malley, B.W., Jr. & Chatterjee, S.K. Inhibition of NK cell activity through TGF-beta 1 by down-regulation of NKG2D in a murine model of head and neck cancer. J Immunol 175, 5541-50 (2005).
Khurana, D. et al. characterization of a spontaneously arising murine squamous cell carcinoma (SCC VII) as a prerequisite for head and neck cancer immunotherapy. Head Neck 23, 899-906 (2001).
De Palma, M., Biziato, D. & Petrova, T.V. Microenvironmental regulation of tumour angiogenesis. Nat Rev Cancer 17, 457-474 (2017).
Reyes-Aldasoro, C.C., Williams, L.J., Akerman, S., Kanthou, C. & Tozer, G.M. An automatic algorithm for the segmentation and morphological analysis of microvessels in immunostained histological tumour sections. J Microsc 242, 262-78 (2011).
Schlessinger, J. et al. Regulation of cell proliferation by epidermal growth factor. CRC Crit Rev Biochem 14, 93-111 (1983).
Adamson, E.D. & Rees, A.R. Epidermal growth factor receptors. Mol Cell Biochem 34, 129-52 (1981).
Laha, D., Grant, R., Mishra, P. & Nilubol, N. The Role of Tumor Necrosis Factor in Manipulating the Immunological Response of Tumor Microenvironment. Front Immunol 12, 656908 (2021).
Carmeliet, P. VEGF as a key mediator of angiogenesis in cancer. Oncology 69 Suppl 3, 4-10 (2005).
Levine, A.J. p53, the cellular gatekeeper for growth and division. Cell 88, 323-31 (1997).
Thompson, S.L. & Compton, D.A. Proliferation of aneuploid human cells is limited by a p53-dependent mechanism. J Cell Biol 188, 369-81 (2010).
Yu, D.S. & Cortez, D. A role for CDK9-cyclin K in maintaining genome integrity. Cell Cycle 10, 28-32 (2011).
Vitale, I. et al. Illicit survival of cancer cells during polyploidization and depolyploidization. Cell Death Differ 18, 1403-13 (2011).
Bjerkvig, R., Tysnes, B.B., Aboody, K.S., Najbauer, J. & Terzis, A.J. Opinion: the origin of the cancer stem cell: current controversies and new insights. Nat Rev Cancer 5, 899-904 (2005).
Lu, X. & Kang, Y. Cell fusion as a hidden force in tumor progression. Cancer Res 69, 8536-9 (2009).
Percie du Sert, N. et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol 18, e3000410 (2020).
Faustino-Rocha, A. et al. Estimation of rat mammary tumor volume using caliper and ultrasonography measurements. Lab Anim (NY) 42, 217-24 (2013).
Lu, S.Y., Li, M. & Lin, Y.L. Mitf induction by RANKL is critical for osteoclastogenesis. Mol Biol Cell 21, 1763-71 (2010).
Suit, H.D., Sedlacek, R.S. & Zietman, A. Quantitative transplantation assays of spontaneous tumors of the C3H mouse as allografts in athymic NCr/Sed-nu/nu nude mice and isografts in C3Hf/Sed mice. Cancer Res 48, 4525-8 (1988).
O'Malley, B.W., Jr., Cope, K.A., Johnson, C.S. & Schwartz, M.R. A new immunocompetent murine model for oral cancer. Arch Otolaryngol Head Neck Surg 123, 20-4 (1997).
Langmead, B. & Salzberg, S.L. Fast gapped-read alignment with Bowtie 2. Nat Methods 9, 357-9 (2012).
Li, B. & Dewey, C.N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics12, 323 (2011).

No competing interests reported.

SupplementaryFiguresandLegends.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Cancer cell-macrophage fusion increases tumor cell heterogeneity and reshapes their microenvironment

Status:

Version 1

Abstract

Figures

Introduction

Results

Spontaneous fusion between squamous cell carcinoma and myeloid cells

SCCVII/SF-BMM hybrid cell clones exhibit variable tumorigenicity

Fusion with BMM increases SCCVII/SF’s ability to recruit immune cells

Tumor angiogenesis

Differential RNA expression

Discussion

Materials And Methods

Animal use and ethical approval

Tumor production, tissue harvest and processing

Plasmid construction and retrovirus production

Cell-cell fusion and cloning of hybrid cells

Flow cytometry

Immunofluorescence and immunohistochemistry

In vitro cell proliferation

Image acquisition and analysis of tumor vascular network

Antibodies

RNA-seq

Statistical analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1