The homeobox protein VentX reverts immune suppression in the tumor microenvironment

The homeobox protein VentX reverts immune suppression in the tumor microenvironment ARTICLE DOI: 10.1038/s41467-018-04567-0 OPEN The homeobox protein VentX reverts immune suppression in the tumor microenvironment 1 2 3 1 Yi Le , Hong Gao , Ronald Bleday & Zhenglun Zhu Immune suppression in the tumor microenvironment (TME) is a central obstacle to effective immunotherapy. Tumor-associated macrophages (TAMs) are key components of the TME. Although TAMs have been viewed as an ideal target of intervention to steer immunity in cancer treatment, the approach has been hampered by the lack of knowledge of how TAM plasticity is controlled by cell intrinsic factors. VentX is a homeobox protein implicated in proliferation and differentiation of human hematopoietic and immune cells. Using clinical samples obtained from cancer patients, we find that VentX expression is drastically reduced in TAMs. We show here that VentX promotes M1 differentiation of TAMs, and that VentX- regulated TAMs, in turn, revert immune suppression at the TME. Using a NSG mouse model of human colon cancers, we demonstrate that VentX regulates TAM function in tumor- igenesis in vivo. Our findings suggest a mechanism underlying immune suppression at TME and potential applications of VentX-regulated TAMs in cancer immunotherapy. 1 2 Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts, USA. Department of Medicine, Tufts Medical Center, Boston, Massachusetts, USA. Department of Surgery, Brigham and Women’s Hospital, Boston, Massachusetts, USA. Correspondence and requests for materials should be addressed to Z.Z. (email: zzhu@partners.org) NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 he role of immunity in oncogenesis has been long appre- function in tumorigenesis in vivo. Taken together, our studies ciated and increasingly exploited. Nevertheless, the efficacy suggested a key role of VentX in regulating TAM plasticity and Tof cancer immunotherapy remains limited, especially for immune status at TME. Targeting VentX, therefore, may open solid tumors. Immune suppression at the tumor microenviron- novel venue of cancer immunotherapy. ment (TME) has been accounted for cancer evasion of immune destruction and regarded as a potential venue of intervention . Results However currently, mechanisms that underlie immune suppres- VentX expression is decreased in TAMs. In advanced tumors, 2, 3 sion at TME remained largely elusive . Tumor-associated 7, 18, 19 TAMs display a pro-tumor M2-like phenotype . The macrophages (TAMs) are key components of TME and have plasticity of TAMs has been well appreciated, however, the been implicated in growth, invasion, and metastasis of nearly all transcriptional machinery that controls the TAM polarization tumors. Derived from circulating monocytes, TAMs display a remains largely unknown. Using TAMs isolated from discarded broad spectrum of phenotypes, ranging from the pro- specimens from colon cancer resection as well as macrophages inflammatory M1-like phenotype in early stages of some from normal mucosa 10 cm away from the tumor sites, we sought 4, 5 tumors to the M2-like phenotype in most advanced tumors . to explore the potential involvement of VentX in TAM plasticity. The M2-like TAMs show an elevated expression of interleukin Consistent with prior findings, we showed that TAMs express (IL)-10, matrix metallopeptidase (MMP), and vascular endothe- significant higher levels of cell surface markers associated with lial growth factor (VEGF), but decreased expression of pro- 20–23 M2 phenotypes, such as CD163 and CD206 (Fig. 1a, b). The inflammatory tumoricidal cytokines, cytotoxic inducible nitric elevated expression of M2 surface markers was accompanied by a oxides (iNOs) and reactive oxygen intermediates (ROIs), and are decrease of M1 surface markers, such as CD40 and7 CD80 thought to be supportive of tumor growth . Besides their func- (Supplementary Fig. 1); however, there was no significant dif- tions in promoting tumorigenesis, TAMs also help generate a ference in the expression of non-discriminating myeloid marker pro-tumor immune suppressive milieu at TME by altering the CD33 in TAMs and control macrophages (Fig. 1a, b). Similar to composition and function of tumor-infiltrating lymphocytes the cell surface markers, there is an increased expression of 3, 6 (TILs) . The plasticity of TAMs has been well recognized. It has markers associated with M2 differentiation but decreased been proposed that by converting the pro-tumor M2-like TAMs expression of M1 markers in TAMs (Supplementary Fig. 2). To into the anti-tumor M1-like phenotype, the TAMs might be determine whether VentX has a role in TAM plasticity, we 7, 8 converted into an effective modality of anti-tumor therapy . quantified VentX expression in TAMs by quantitative reverse- Nevertheless, this potential application of modulating TAM transcription PCR (qRT-PCR) and found that, in comparison function in cancer treatment has been hampered by our ignor- with its expression in control macrophages, VentX expression is ance of how TAMs plasticity is controlled by cell intrinsic decreased about 77% in TAMs (Fig. 1c). The decreased expression 5, 9 factors . of VentX in TAMs was further verified by western blot analysis, Dorsoventral axis formation represents coordinated cell pro- using VentX-specific antibodies (Fig. 1d). liferation and differentiation during early vertebrate embry- ogenesis. In trying to understand the molecule basis of the process, our recent work led to the appreciation of the Xenopus VentX drives TAM towards M1 phenotype. To determine homeobox protein Xom, a unstable protein of the ventral whether VentX is involved in TAM plasticity, we examined the BMP4 signaling pathway , as a novel lymphoid enhancing correlation of VentX expression with TAM phenotypes. Using factor/T-cell factor (LEF/TCF)-associated factor that antagonizes .we lipopolysaccharide (LPS) as an M1 phenotype inducer β-catenin of the dorsal Wnt signaling pathway during dorso- found that VentX expression is significantly elevated in TAMs ventral polarization of Xenopus embryos . To explore the after them being exposed to LPS (Fig. 2a). The elevated expres- potential clinical relevance of the findings, we performed sion of VentX is accompanied by elevated expression of M1 sequence homology search and functional analysis, which led to markers in TAMs, including the secretion of inflammatory the identification of human homeobox protein VentX as a human cytokines and cytotoxic iNOs (Fig. 2b, c). Similar to LPS, we 12, 13 homolog of Xom . We found that VentX is primarily found that TAMs can also be activated by pro-inflammatory expressed in hematopoietic cells and controls proliferation and cytokines, such as interferon-γ (IFNγ) (Supplementary Fig. 3). To differentiation of hematopoietic cells from early ontogenesis to determine whether VentX has a required role in TAM plasticity, 12, 14–16 terminal differentiation . Interestingly, comparative we examined the effects of knockdown VentX on TAM pheno- genomic studies showed that VentX is preserved in primates and types. As shown in Fig. 2d, treatment of TAM with VentX human but lost in mice since the evolutionary divergence of morpholino oligos (VentX-MO) led to around 80% reduction of rodent and primate lineages . Using an in vitro culture of human VentX expression. The decreased VentX expression, which is monocyte-derived macrophage model, we found that VentX verified by western blot analysis (Supplementary Fig. 4), is promotes and is required for M1 but not M2 activation . These accompanied by decreased secretion of inflammatory cytokines findings inspired us to explore whether VentX has a role in and cytotoxic iNOs (Fig. 2d–f). Corresponding to the changes of regulating plasticity and function of TAMs. M1 phenotype, there is also a change of the M2 phenotype. As Using clinical samples obtained from primary colorectal cancer shown in Fig. 2g, we found that the effect of LPS on the patients, our current studies showed that VentX expression is expression of CD206, a M2 marker highly expressed in TAMs drastically reduced in TAMs in comparison with its expression in (Supplementary Fig. 1a) , was abolished by VentX-MO macrophages isolated from normal mucosa of the same patients. treatment. We found that TAM VentX expression profile correlates with The correlation between VentX expression levels and TAM TAM phenotypes, and that ectopic expression of VentX con- phenotypes prompted us to explore the potential direct effect of verted the M2-like phenotype of TAMs into M1-like phenotype. VentX on TAM phenotype. Using transfection studies, we found Moreover, we found that VentX-regulated TAMs revert immune that TAMs transfected with GFP-VentX displayed a characteristic suppression at TME by inhibiting regulatory T-cell (Treg) dif- M1 morphology with elongated/fibroblast-like cell shape, whereas ferentiation and promoting CD8 TIL activation. Using a NOD there was no such changes in TAMs transfected with the control scid γ (NSG) mouse model of patient-derived xenograft (NSG- green fluorescent protein (GFP) (Fig. 3a). In comparison with the PDX) of colon cancers, we showed that VentX-modulated-TAMs control GFP-transfected TAMs, the surface expression of M1 2 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE a b Normal mucosal CD68 CD163 CD206 CD33 Normal mucosal Tumor 60 ** ** ** Tumor CD163 CD206 CD33 CD68 Fluorescence intensity 1.2 Normal Tumor** Normal 0.8 Tumor Mean 1.000 0.2260 Std. deviation 0.0 0.1146 0.6 0.4 0.2 Patient a Patient b Patient c N TT N N T WB:VentX 25 kDa WB:β-actin 37 kDa Fig. 1 VentX expression is downregulated in TAMs. a, b FACS analysis and the percentage of surface expression of CD33, CD68, CD163, and CD206 on macrophages isolated from normal mucosa and tumors. Shaded peaks indicated isotope control. Representative figures of five independent experiments were shown, n= 5. Error bars represent SD and paired Student’s t-test was performed. **p< 0.01. c Paired comparison of RT-PCR measurement of VentX mRNA expression in macrophages isolated from normal control tissues and TAMs of 15 patients. The relative VentX mRNA expression levels in normal macrophages were arbitrarily designated as 1. Paired Student’s t-test was performed, **p< 0.01 TAMs vs. normal macrophages. Data shown in the box are mean ± SD of 15 patients. d Western blot analysis of endogenous VentX protein levels in macrophages isolated from normal control tissues and TAMs of three patients marker CD40, CD80, and CD86 was significantly increased in VentX-regulated TAMs modulate TILs differentiation. TILs TAMs transfected with GFP-VentX (Fig. 3b). In addition, the have key roles in anti-tumor immunity . The regulatory CD4 secretion of pro-inflammatory cytokines tumor necrosisfactor-α +CD25+ Treg and tumoricidal CD8+ TILs are key components (TNFα), IL-1β, and IL-12 were significantly increased, whereas of TME. The ratio of CD4+CD25+ Tregs and CD8+ TILs, as secretion of the M2 cytokine, IL-10, was significantly reduced well as functional defects of CD8 TILs, have been implicated in 6, 27–30 (Fig. 3c). Using qRT-PCR analysis, we showed a characteristic pathogenesis and prognosis of solid tumors . Consistent increased expression of of M1 genes, such as IL-1β, IL-6, IL-12, with prior findings , we found that there was a significantly TNF-α, and iNOs (Fig. 3d), and decreased expression of M2 increased number of the CD4+CD25+/CD4+Foxp3+ Treg cells genes, such as CCL18, MMP9, VEGFA, and Arg1 in TAMs and a significantly decreased number of CD8+ TILs in tumor transfected with GFP-VentX (Fig. 3e). In comparison, there was tissues (Supplementary Fig. 6 and 7). As CD4+ no significant changes of non-discriminating macrophages CD25+/CD4+Foxp3+ Treg cells are derived from markers, such as the CD33 and CD68 upon the transfection of CD4+ T cells, to explore the potential involvement of VentX- GFP-VentX in TAMs (Supplementary Fig. 5a, 5b). The changes regulated TAMs in compositions of TILs, we sought to determine of cell surface markers are associated with changes of intracellular whether VentX-regulated TAMs modulate CD4+ T-cell differ- signaling molecules such as the STAT1 and STAT3, which are entiation. To attend our goal, CD4+ T cells isolated from per- associated with M1 and M2 phenotype, respectively (Supplemen- ipheral blood were co-cultured with autologus TAMs transfected tary Fig. 5c) . Taken together, our data suggested that VentX with either GFP-VentX or control GFP for 5 days. As shown in functions as a master switch of TAM plasticity and drives TAMs Fig. 4a, co-culture of CD4+ T cells with TAMs led to significant toward M1 phenotypes through alternating intracellular signaling induction of CD4+CD25+/CD4+Foxp3+ cells. In contrast, pathways involved in the process. ectopic expression of VentX in TAMs abolished the induction. To NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 3 | | | Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 CD33 Patient 6 CD68 Patient 7 Patient 8 CD163 Patient 9 CD206 Patient 10 Patient 11 Patient 12 Patient 13 Patient 14 Patient 15 Cell counts Cell counts Relative VentX mRNA level % of positive cells ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 a b c TAMs ** 1.2 1200 ** ** 0.8 0.4 0 0 0 TAMs TAMs + 0 24 h 48 h TAMs TAMs + LPS LPS Incubation time of LPS de f 1.2 50 0.8 0.6 ** ** 0.4 ** 300 10 0.2 0 0 Control VentX Control VentX Control VentX MO MO MO MO MO MO LPS stimulation LPS stimulation LPS stimulation Control LPS Control-MO + LPS VentX-MO + LPS 0.63 49.3 0 1.38 0 1.71 0.86 35.8 45.8 4.30 51.0 47.6 51.8 46.5 53.4 9.94 CD68 Fig. 2 VentX regulates TAM plasticity. a VentX expression in TAMs after LPS stimulation. Isolated TAMs were cultured in RPMI medium and stimulated with LPS. VentX mRNA levels were measured with real-time PCR at indicated times. b TNF-α and c nitrate level from TAMs after LPS treatment were determined by ELISA and Griess reagent. d Effects of VentX-MO on VentX expression. Isolated TAMs were transfected with MO oligonucleotides targeting VentX or control. Twenty-four hours post transfection, LPS was added to culture medium. The cells were collected 48 h later and VentX expressions were determined by real-time RT-PCR. e TNF level and f nitrate level from VentX-MO- or control-MO-treated TAMs after LPS stimulation. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05, **p< 0.01. g Effects of VentX-MO or control-MO on percentage of CD206+CD68+ TAMs after LPS stimulation. Representative data of at least three independent experiments were shown corroborate the functional relevance of the findings, CD4+ T cells control GFP for 5 days. CD4+CD25+Treg cells and CD8+ TILs isolated from normal tissues were co-cultured with autologus were then isolated and quantified by flow cytometry, following 32, 33 TAMs transfected with either GFP-VentX or control GFP for established protocols (Supplementary Fig. 9) . As shown in 5 days. We found that TAMs induced T-cell expression of inhi- Fig. 5a, b, co-culture of the tumors with GFP-VentX-modified bitory IL-13, but the induction function was abolished by ectopic TAMs led to a significant decrease of the CD4+ expression of VentX (Supplementary Fig. 8). To further deter- CD25+Tregs and a significant increase of the CD8+ TILs. In mine whether VentX-regulated TAMs modulate TIL function, addition to the significant increase in the number of CD8+ TILs, CD8+ TILs were co-cultured with TAMs transfected with GPF- there is a significant increase in the expression of IFNγ and VentX or control GFP. As shown in Fig. 4b, co-culture of CD8+ granzyme B in CD8+ cells in tumor tissues after being incubated TILs with VentX-modified TAMs led to significant activation of with VentX-modified TAMs (Fig. 5c). CD8+ TILs, as indicated by CD8+ activation markers, such as IFNγ and granzyme B. Taken together, our data suggested that VentX regulates TAM function and tumorigenesis in vivo. VentX-regulated TAMs modulate TIL differentiation and Converting the M2 phenotype of TAM into the M1 phenotype function. 5, 8 has been viewed as a promising venue for cancer treatment . Our findings that VentX promotes M1 polarization of TAM VentX-regulated TAMs modulate composition of TILs in prompted us to explore whether VentX-modulated TAM plasti- TME. Our findings that VentX-regulated TAMs modulate dif- city has a role in tumorigenesis in vivo. As VentX does not have a ferentiation and function of T cells prompted us to test the mouse homolog , to achieve our goal, we adapted a NSG mouse hypothesis that VentX-regulated TAMs control immune status at model, which support heightened engraftment of human hema- 34 35 TME by modulating the differentiation and function of TILs. To topoietic cells . Following established protocol , a NSG-PDX of test this hypothesis, en bloc primary tumor tissues were co- human colon cancers was generated by engrafting small pieces of cultured with autologous TAMs transfected with GPF-VentX or primary human colon cancer tissues into subcutaneous space on 4 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | Relative VentX Relative VentX CD206 mRNA level mRNA level TNFα (pg/mL) TNFα (pg/mL) Nitrite (unit) Nitrite (unit) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE ab 4.63 0.39 0.61 10.3 7.05 10.4 Green fluorescence TNFα IL-1β IL-12p70 IL-10 120 * 400 400 40 200 0 0 0 0 d e GFP GFP iNOS ARG1 GFP-VentX * GFP-VentX TNF CCL18 IL-12 MMP9 IL-6 * VEGFA IL-1b 024 68 0.0 0.3 0.6 0.9 1.2 1.5 mRNA expression level mRNA expression level Fig. 3 VentX polarizes TAMs towards M1 phenotype. a Morphology of TAMs transfected with GFP or GFP-VentX. Scale bar: 20 μm. b Surface expression of CD40, CD80, and CD86 in TAMs transfected with GFP or GFP-VentX as determined by flow cytometry analysis. Representative data of three independent experiments were shown. c ELISA measurement of secretion of pro-inflammatory cytokines in TAMs transfected with GFP and GFP-VentX. Results represent mean ± SD of four independent experiments and paired Student’s t-test was used. Expression of M1 (d) and M2 (e) macrophage markers in TAMs transfected with GFP or GFP-VentX as determined by qRT-PCR. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05 the abdominal side of the NSG mice. The growth of the tumors in Discussion the NSG mice was observed for 8 weeks (Fig. 6a). The tumors Tumor immunology is now a promising field for exploration of were then dissected out and sectioned and the growth of human pathogenesis and treatment of cancers. Macrophages have colon cancers was confirmed by hematoxylin and eosin (H&E) executor role in immunity and have been recognized as key and CK20 staining (Fig. 6b). To test the potential effects of components of tumor mass. The search for factors that can be VentX-regulated TAMs on tumor growth 1 week post implan- manipulated to steer macrophage plasticity has proved to be a tation of fragments of colon cancers, the NSG-PDX mice were fundamental challenge. Signaling pathways such as the JAK/ tail-vein injected with TAMs transfected with GFP-VentX or STAT1, JNK/STAT6, AKT1/AKT2, and PI3Kγ, transcriptional control GFP. As shown in Fig. 6c, we found that, in comparison regulators such as PPARγ and PPARδ,NFκB, C/EBP, and IRFs, with the GFP-transfected TAMs, GFP-VentX-transfected TAMs as well as microRNAs such as microRNA-155 and miR-142-3p exerted strong inhibition on tumor growth in the NSG-PDX have been shown to modulate macrophage 9, 24, 36, 37 mice. To further determine whether the inhibition of tumor- differentiation . Nevertheless, whether these factors can igenesis in the NSG-PDX mice is related to VentX-regulated be targeted to turn TAMs into a tumoricidal cells remains TAM polarity 1 week post implantation of fragments of colon unclear. As a unique human hematopoietic transcriptional factor cancers, the NSG-PDX mice were tail-vein injected with in vitro that does not have a murine homolog, VentX was found to have M1-differentiated TAMs transfected with either VentX-MO or essential role in controlling proliferation and differentiation of control-MO. Consistent with the results of over expression stu- hematopoietic cells and function as a p53-independent tumor 12, 15, 16, 38–40 dies, the M1-TAMs transfected with control-MO exerts strong suppressor . Data of our current study suggested inhibition of tumor growth, but the inhibition was abolished by VentX controls TAM plasticity, which in turn, reverts immune knocking-down VentX expression with VentX-MO (Fig. 6d). suppression at the TME through regulating TIL differentiation NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 5 | | | GFP GFP-VX GFP GFP-VX GFP GFP-VX GFP GFP-VX GFP-VentX GFP pg/mL pg/mL GFP GFP-VentX CD40 CD40 pg/mL CD80 CD80 pg/mL CD86 CD86 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 Co-culture of CD4 with/without modified TAMs No TAM GFP GFP-VentX ** 2.99 9.19 3.14 No TAM GFP GFP-VentX CD4 No TAM GFP GFP-VentX ** 3.98 8.41 4.86 CD4 No TAM GFP GFP-VentX Co-culture of CD8 with modified TAMs GFP GFP-VentX 37.2 12.1 ** ** GFP GFP-VentX 9.87 43.5 IFN-γ Granzyme CD8 Fig. 4 VentX modulates TAM induction of TILs differentiation and activation. a VentX modulates TAM induction of T-cell differentiation. CD4+ T cells were incubated with autologus TAMs transfected with GFP or GFP-VentX for 5 days. Differentiation of the CD4+ T cells were determined by CD25 and Foxp3 analysis. The results were mean ± SD of five independent experiments. b VentX modulates TAM effects on CD8 activation. CD8 TILs isolated from tumors were incubated with autologus TAMs transfected with GFP or GFP-VentX for 5 days. CD8+TIL activation was measured by FACS analysis of IFNγ and granzyme B. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05, **p< 0.01 and function, which was supported by both in vitro and in vivo of colon cancer specimens used for this study were listed in Table 1. Around 5–10 g tissues were collected from tumor mass, or normal mucosa 10 cm away from tumor studies (Figs. 4, 5 and Supplementary Fig. 11). Interestingly, mass. consistent with prior findings that macrophages may affect T-cell differentiation indirectly , we found that TAMs do not need to Preparation of intraepithelial lymphocytes. Lymphocytes were isolated following accumulate in significant numbers inside tumor tissues to exert 14, 42 previously described techniques with modification . In brief, dissected fresh its function (Supplementary Fig. 12). Our findings revealed a 2+ 2 mucosa and tumor mass were rinsed in 10-cm Petri dish with Ca -free and Mg novel mechanism underlying immune suppression at TME and -free Hank’s balanced salt solution (HBSS) (Life Technologies) containing 2% fetal bovine serum (FBS) and 2 mM Dithiothreitol (DTT) (Sigma-Aldrich). The suggested the potential application of VentX-regulated TAMs as a mucosa and tumor were then cut into around 0.1 cm pieces by a razor blade and novel modality of immunotherapy, especially for solid tumors, incubated in 5 mL HBSS containing 5 mM EDTA (Sigma-Aldrich) at 37 °C for 1 h, which are refractory to current available immunotherapy. Cur- then passed through a gray-mesh (100 μm). The flowthrough contains intrae- rently, the cause of decreased VentX expression in TAMs and the pithelial lymphocytes and epithelial cells and was analysis by a flow cytometer. potential clinical application of VentX-modulated TAMs in cancer treatment remain to be further defined. Isolation of macrophages from tumor mass and normal mucosa. Lamina pro- pria mononuclear cells (LPMCs) were isolated following established protocol, which 14, 42, 43 does not lead to activation of macrophages .Briefly, normal mucosa and tumor 2+ 2+ Methods tissues were rinsed with HBSS and then incubated in HBSS (with Ca and Mg ), Collection of colon tissue samples. A total of 42 patients with colon cancer, who containing 2% FBS, 1.5 mg/mL Collagenase D (Roche), 0.1 mg/mL Dnase I at 37 °C were scheduled for surgical resection at Brigham and Women’s Hospital, were for 1 h. Digested tissues were then passed through a gray-mesh (70 μm) filter. The consented to have a portion of tissues and blood collected for research purposes. flowthrough were collected, washed, and resuspended in a RPMI 1640 medium. The All patients signed an informed consent document that was approved by the cells were layered on Ficoll-Paque Plus media (GE Healthcare), and then centrifuged Institutional Review Board of Brigham and Women’s Hospital. The characteristics at 2000 r.p.m. for 30 min without brake. LPMCs at the interface were collected. 6 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | Foxp3 CD25 Granzyme B IFN-γ % of activated CD8+ T cells CD4+Foxp3+ (%) CD4+CD25+ (%) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE Co-culture of tumor with modified TAMs GFP GFP-VentX 12.8 46.2 14.8 28.7 28.0 13.0 49.2 7.23 CD4 GFP GFP-VentX Co-culture of tumor with modified TAMs GFP GFP-VentX 40 CD8+ CD8+ 27.8 13.5 10 GFP GFP-VentX CD8 GFP GFP-VentX 13.0 64.6 GFP GFP-VentX 15.1 74.9 IFN-γ Granzyme B CD8 Fig. 5 VentX-regulated TAMs modulate composition and function of TILs in tumor microenvironment. a, b Composition of CD4+CD25+ Tregs (a) and CD8+ TILs (b) in tumors after en bloc co-culture of tumor tissues with autologus TAMs transfected with GFP or GFP-VentX for 5 days. c CD8 TILs were isolated from tumor tissues after being incubated with autologus TAMs transfected with GFP or GFP-VentX for 5 days. CD8+TIL activation was measured by FACS analysis of IFNγ and granzyme B. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05 Normal mucosal macrophages and TAMs were further purified from LPMCs using warmed Human Monocyte Nucleofector Medium containing 2 mM glutamine and EasySep™ Human Monocyte/Macrophage Enrichment kit without CD16 depletion 10% FBS. Cells were then resuspended into complete RPMI medium and treated (StemCell Technologies), according to the manufacturer’s instructions. The isolation with appropriate cytokines to induce differentiation into macrophages. process does not lead to activation of macrophages and the purify of isolated intestinal 14, 42, 43 macrophages was routinely more than 95% . More than 98% of cells isolated by the techniques were viable by propidium iodide staining. FACS analysis. Phenotypic analysis of TAMs and other lymphocytes was per- formed using flow cytometry after immunolabeling of cells with fluorescence dye- conjugated antibodies. The following antibodies were used: Phycoerythrin (PE)- Transfection assays. Transfection of GFP-VentX and GFP into blood macro- conjugated anti-CD3 (OKT3), -CD25 (BC96), -CD14 (61D3), -CD68 (Y182A), phages or TAMs were carried out through lipofectamine 2000 (Life Technologies) -CD163 (GH161), and -CD206, and fluorescein isothiocyanate (FITC)-conjugated according to the manufacturer’s protocol. Forty-eight hours after transfection, cells anti-CD4 (RPA-T4) and -CD33 (HIM3-4), and Allophycocyanin (APC)-con- were filtered through a 70 μm filter for cell sorting. GFP-positive cells were sorted jugated anti-CD8 (OKT8) and -CD4 (OKT4) (eBioscience, Inc). Intracellular by BD FACSAria II under the Baker Bio-Protect Hood in a sterile condition. After staining of Foxp3 (236 A/E7), IFNγ (4 S.B3), and Granzyme B (GB11) was per- sorting, cells were cultured in RPMI 1640 complete medium. formed with PE-conjugated antibodies following the protocol provided by manu- facturer. Isotope control labeling was performed in parallel. Antibodies were VentX knockdown. Colon TAMs or human primary monocytes were transfected diluted as recommended by the supplier. Labeled cells were collected on FACScan flow cytometer with Cell-Quest software (BD Biosciences) and analyzed by FlowJo with Morpholino oligonucleotides (MO) (Gene Tools, LLC, Philomath, OR) using the Human Monocyte Nucleofector Kit (Lonza, Walkersville, MD) as previously software. Results are expressed as the percentage of positive cells. 14 6 described . Briefly, 5 × 10 cells were resuspended into 100 µl nucleofector solu- tion with 2.5 nmol of either VentX-MO oligonucleotides (VentX-MO: 5′-TACT- CAACCCTGACATAGAGGGTAA-3′ or a standard control-MO oligonucleotides Cytokine measurement. Levels of IL-1β, IL-10, IL-13, TNF-α, and IL-12p70 were and electroporated with the Nucleofector II Device (Lonza). Cells were then quantified using enzyme-linked immunosorbent assay kits obtained from eBios- immediately removed from the device and incubated overnight with 1 ml pre- ciences. Analyses were conducted according to the manufacturer’s instructions. NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 7 | | | Granzyme B IFN-γ Cell counts CD25 % of activated CD8+ T cells % of CD8+ TIL cells % of CD4+CD25+ cells ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 a b Mice 1 Mice 2 Mice 3 Tumor 0 246 8 Weeks cd M1-TAMs-VentX-MO TAMs-GFP M1-TAMs-Control-MO TAMs-GFP-VentX 2500 2500 ** 1500 1500 ** 1000 1000 ** 500 500 ** ** ** 0 0 02468 02468 Weeks Weeks Fig. 6 VentX-regulated TAMs modulate tumorigenesis in vivo. a A NSG-PDX model of human colon cancers. Small pieces of colon cancer tissues were transplanted into subcutaneous space on the abdominal side of NSG mice and the growth of the tumors were observed for 8 weeks. Data shown were results of seven independent replicates of the experiments. b H&E and CK20 staining of tumors resected from three NSG-PDX mice. Scale bar: 50 μm. c Growth curve of tumors in NSG-PDX mice treated with TAMs transfected with GFP-VentX or control GFP. Results shown represent mean ± SD of three independently replicated experiments, n = 3, and paired Student’s t-test was performed, **P < 0.01. d Growth curve of tumors in NSG-PDX mice treated with M1-TAMs transfected with VentX-MO or control-MO. Results shown represent mean ± SD of seven independently replicated experiments, n = 7, and paired Student’s t-test was performed, **P < 0.01. Photos of excised tumors at the end of experiments were shown. Scale bar: 5 mm Quantitative RT-PCR. Total RNA was isolated by the TRIzol reagent (Life blood CD4 cells were enriched by using Easysep human CD4-negative selection kit Technologies) and RNA amounts were measured by NanoDrop 2000 (Thermo following the manufacturer’s instructions (StemCell Technologies). GFP-VentX or 6 6 Scientific). An equal amount of RNA was used for first-strand complementary GFP-transfected TAMs (0.5 × 10 ) were incubated with 5 × 10 of CD4 in com- DNA synthesis with SuperScript III First-Strand Synthesis System (Life Technol- pleted RPMI 1640 at 37 °C, 5% CO for 5 days. Cells were stained with CD4-FITC ogies) according to the manufacturer’s protocol. To amplify VentX cDNA with and CD25-PE, or permeabilized and stained with CD4-FITC and Foxp3-PE, then conventional PCR, we used the AccuPrime Taq DNA polymerase system (Life analyzed by a flow cytometer. For CD8+ TIL activation assay, CD8+ TILs were Technologies) following the manufacturer’s instructions. Quantitative measure- enriched from intraepithelial lymphocytes by CD8+ T-cell Enrichment Kit 6 6 ment of VentX and other genes cDNA were carried out with SYBR Green on a (StemCell Technologies). Cells (5 × 10 ) were then incubated with 0.5 × 10 of LightCycler (480 Real-Time PCR System; Roche). The primers used list in Sup- GFP-VentX or GFP-transfected TAMs for 5 days, followed by staining and analysis plementary Table 1. Relative expression profiles of mRNAs were then calculated with flow cyotmetry. using the comparative Ct method (DDCT method). Co-cultures of tumors and TAMs. Tumor mass were washed with 1 × phosphate- Western blot analysis. Western blot analysis was performed as described pre- buffered saline (PBS) buffer plus antibiotics and then cut into 0.5 cm pieces. viously .Briefly, total cells were lysed in 1 × RIPA buffer (Abcam, Inc.) mixed with Around 80 mg of tissues were mix cultured with 0.5 × 10 of GFP-VentX or GFP- protease inhibitor cocktails (Cell Signaling Technology). Proteins were resolved by transfected TAMs of same patient in 2 mL of RPMI 1640 completed medium, 4–15% Tris-Glycine Gel (Bio-Rad) electrophoresis. Primary antibodies used supplemented with 2.5% antibiotic–antimycotic solution (Cellgro, Manassas, VA). included GFP (eBioscience 14–6674, 1:1000), VentX (Abcam, Inc. ab105352, The cultures were incubated at 37 °C, 5% CO for 5 days. The tissues were then 1:500), and β-actin (Cell Signaling Technology 4967, 1:2000). subjected to cell isolation and analyzed by a flow cytometry or immunohis- tochemistry studies. Arginase activity and NO assays. Arginase activity was quantified in cell lysates by measuring the production of urea using the QuantiChrom Arginase Assay Kit, NSG-PDX model of human colon cancers. Animal models of primary human following the manufacturer’s instuctions (DARG-200, BioAssays Systems). Nitrite colon cancers were developed as described previously . Briefly, 8-week-old NOD. concentrations in culture supernatants were determined using Griess reagent kit scid tm1Wjl Cg-Prkdc Il2rg /SzJ mice (commonly known as NSG mice) were purchased (Molecular Probes, Eugene, OR), as described previously . from The Jackson Laboratory and maintained under specific pathogen-free con- ditions. All animal experiments were approved by the Institutional Animal Care Treg cell induction and CD8+ TIL cell activation. Treg cell inductions were and Use Committee at Harvard Medical School. Colon tumors were cut into 44, 45 performed using previously described methods with modification . Briefly, the around 0.5 cm and surgically seeded in subcutaneous space of abdominal side of 8 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | TAMs- GFP TAMs- GFP-VentX M1-TAMs- VentX-MO M1-TAMs- Control-MO 3 3 Tumor volume (mm ) Tumor volume (mm ) CK20 H&E Tumor volume (mm ) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE 6. Schreiber, R. D., Old, L. J. & Smyth, M. J. Cancer immunoediting: integrating Table 1 Characteristics of colon cancer patients used for immunity’s roles in cancer suppression and promotion. Science 331, study 1565–1570 (2011). 7. Sica, A. et al. Macrophage polarization in tumour progression. Semin. Cancer Biol. 18, 349–355 (2008). Characteristics Number (n = 42) (%) 8. Noy, R. & Pollard, J. W. Tumor-associated macrophages: from mechanisms to Age, years therapy. Immunity 41,49–61 (2014). ≤ 40 5 11.9 9. Lawrence, T. & Natoli, G. Transcriptional regulation of macrophage 40–60 16 38.1 polarization: enabling diversity with identity. Nat. Rev. Immunol. 11, 750–761 ≥ 60 21 50.0 (2011). Sex 10. Zhu, Z. & Kirschner, M. Regulated proteolysis of Xom mediates dorsoventral Male 24 57.1 pattern formation during early Xenopus development. Dev. Cell. 3, 557–568 (2002). Female 18 42.9 11. Gao, H., Wu, B., Giese, R. & Zhu, Z. Xom interacts with and stimulates Pathology type transcriptional activity of LEF1/TCFs: implications for ventral cell fate Adenocarcinoma 42 100 determination during vertebrate embryogenesis. Cell Res. 17, 345–356 (2007). Stage 12. Gao, H. et al. VentX, a novel lymphoid-enhancing factor/T-cell factor- I8 19 associated transcription repressor, is a putative tumor suppressor. Cancer Res. II 14 33.3 70, 202–211 (2010). III 15 35.7 13. Moretti, P. A. et al. Molecular cloning of a human Vent-like homeobox gene. IV 5 11.9 Genomics 76,21–29 (2001). 14. Wu, X., Gao, H., Ke, W., Giese, R. W. & Zhu, Z. The homeobox transcription factor VentX controls human macrophage terminal differentiation and proinflammatory activation. J. Clin. Invest. 121, 2599–2613 (2011). NSG mice. TAMs were transfected with VentX-MO or control-MO, and then 15. Gao, H. et al. Suppression of homeobox transcription factor VentX promotes cultured in M1-Macrophage Generation Media (PromoCell). After 1 week of expansion of human hematopoietic stem/multipotent progenitor cells. J. Biol. xenograft, 0.8 × 10 of M1-differentiated TAMs transfected with VentX-MO or Chem. 287, 29979–29987 (2012). control-MO were injected into mice through tail vein. Tumor growth was mon- 16. Wu, X., Gao, H., Bleday, R. & Zhu, Z. Homeobox transcription factor VentX itored twice a week and measured by a caliper for 8 weeks. regulates differentiation and maturation of human dendritic cells. J. Biol. In vivo tracing of injected TAMs were carried out using carboxyfluorescein Chem. 289, 14633–14643 (2014). succinimidyl ester (CFSE)-stained TAMs, according to the manufacture’s 17. Zhong, Y. F. & Holland, P. W. The dynamics of vertebrate homeobox gene instruction (Molecular Probes). Briefly, half million of TAMs were incubated in 5 evolution: gain and loss of genes in mouse and human lineages. BMC Evol. mL of 5 μM CFSE staining solution in a 37 °C water bath for 20 min. Cells were Biol. 11, 169 (2011). then incubated with 20 ml of RPMI 1640 completed medium for 5 min to remove 18. Redente, E. F. et al. Tumor progression stage and anatomical site regulate unbounded dye. After centrifugation, cells were dissolved in PBS and then tail-vein tumor-associated macrophage and bone marrow-derived monocyte injected into NSG mice. Tumor tissues and TAMs were isolated 7 days after polarization. Am. J. Pathol. 176, 2972–2985 (2010). injection and TAMs were isolated as CFSE-positive cells. 19. Biswas, S. K., Allavena, P. & Mantovani, A. Tumor-associated macrophages: functional diversity, clinical significance, and open questions. Semin. Immunohistochemistry. Immunohistochemistry were performed following Immunopathol. 35, 585–600 (2013). established protocol . Briefly, tumor or normal mucosa were fixated in formalin 20. Quatromoni, J. G. & Eruslanov, E. Tumor-associated macrophages: function, (Fisher Scientific Company, Kalamazoo, MI). The tissues were then embedded in phenotype, and link to prognosis in human lung cancer. Am. J. Transl. Res. 4, paraffin and sectioned. CK20 (Dako, Carpinteria, CA, clone Ks20.8, 1:50) and H&E 376–389 (2012). staining were performed at research pathology cores at Dana-Farber/Harvard 21. Kurahara, H. et al. Significance of M2-polarized tumor-associated macrophage cancer center. Microscopic analysis was performed with a Nikon Eclipse Ti in pancreatic cancer. J. Surg. Res. 167, e211–e219 (2011). fluorescence microscopy. Images were captured at an original magnification of × 40 22. Shabo, I., Stal, O., Olsson, H., Dore, S. & Svanvik, J. Breast cancer expression using a color camera applying the NIS Elements imaging software (Nikon). of CD163, a macrophage scavenger receptor, is related to early distant Brightness and contrast for representative images were adjusted equally among recurrence and reduced patient survival. Int. J. Cancer 123, 780–786 (2008). groups. 23. Zhang, Y., Sime, W., Juhas, M. & Sjolander, A. Crosstalk between colon cancer cells and macrophages via inflammatory mediators and CD47 promotes Statistical analysis. Student’s t-test was used for statistical analysis. Data are tumour cell migration. Eur. J. Cancer 49, 3320–3334 (2013). presented as mean ± SD. The level of significance is indicated by the p-value. In all 24. Cai, X. et al. Re-polarization of tumor-associated macrophages to pro- figures, levels of statistical significance are indicated as *p < 0.05 and **p < 0.01. inflammatory M1 macrophages by microRNA-155. J. Mol. Cell Biol. 4, 341–343 (2012). Data availability. The authors declare that all the other data supporting the 25. Luo, Y. et al. Targeting tumor-associated macrophages as a novel strategy findings of this study are available within the article and its Supplementary against breast cancer. J. Clin. Invest. 116, 2132–2141 (2006). Information files and from the corresponding author upon reasonable request. 26. Sica, A. & Bronte, V. Altered macrophage differentiation and immune dysfunction in tumor development. J. Clin. Invest. 117,1155–1166 (2007). Received: 24 August 2017 Accepted: 9 May 2018 27. Naito, Y. et al. CD8+T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res. 58, 3491–3494 (1998). 28. Sato, E. et al. Intraepithelial CD8+tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc. Natl. Acad. Sci. USA 102, 18538–18543 (2005). 29. Galon, J. et al. Type, density, and location of immune cells within human References colorectal tumors predict clinical outcome. Science 313,1960–1964 1. Munn, D. H. & Bronte, V. Immune suppressive mechanisms in the tumor (2006). microenvironment. Curr. Opin. Immunol. 39,1–6 (2016). 30. Radoja, S. & Frey, A. B. Cancer-induced defective cytotoxic T lymphocyte 2. Gonzalez-Gugel, E., Saxena, M. & Bhardwaj, N. Modulation of innate effector function: another mechanism how antigenic tumors escape immune- immunity in the tumor microenvironment. Cancer Immunol. Immunother. 65, mediated killing. Mol. Med. 6, 465–479 (2000). 1261–1268 (2016). 31. Ries, C. H. et al. Targeting tumor-associated macrophages with anti-CSF-1R 3. Gajewski, T. F. et al. Immune resistance orchestrated by the tumor antibody reveals a strategy for cancer therapy. Cancer Cell. 25, 846–859 microenvironment. Immunol. Rev. 213, 131–145 (2006). (2014). 4. Chanmee, T., Ontong, P., Konno, K. & Itano, N. Tumor-associated 32. Shin, A. S. & Camilleri, M. Diagnostic assessment of diabetic gastroparesis. macrophages as major players in the tumor microenvironment. Cancers Diabetes 62, 2667–2673 (2013). (Basel) 6, 1670–1690 (2014). 33. Liu, X. et al. Regulatory T cells trigger effector T cell DNA damage and 5. Bronte, V. & Murray, P. J. Understanding local macrophage phenotypes in senescence caused by metabolic competition. Nat. Commun. 9, 249 (2018). disease: modulating macrophage function to treat cancer. Nat. Med. 21, 34. Pearson, T., Greiner, D. L. & Shultz, L. D. Creation of “humanized” mice to 117–119 (2015). study human immunity. Curr. Protoc. Immunol.15,15–21 (2008). NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 9 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 35. Maykel, J. et al. NOD-scidIl2rg (tm1Wjl) and NOD-Rag1 (null) Il2rg Author contributions (tm1Wjl): a model for stromal cell-tumor cell interaction for human colon Y.L., H.G., and Z.Z. designed the research, analyzed the data, prepared figures. Y.L. and cancer. Dig. Dis. Sci. 59, 1169–1179 (2014). H.G. performed the experiments. R.B. and Z.Z. provided tumor samples and clinical 36. Kaneda, M. M. et al. PI3Kgamma is a molecular switch that controls immune information. Z.Z. wrote the manuscript. suppression. Nature 539, 437–442 (2016). 37. Sonda, N. et al. miR-142-3p prevents macrophage differentiation during Additional information cancer-induced myelopoiesis. Immunity 38, 1236–1249 (2013). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 38. Wu, X. et al. VentX trans-activates p53 and p16ink4a to regulate cellular 018-04567-0. senescence. J. Biol. Chem. 286, 12693–12701 (2011). 39. Rawat, V. P. et al. The vent-like homeobox gene VENTX promotes human Competing interests: The authors declare no competing interest. myeloid differentiation and is highly expressed in acute myeloid leukemia. Proc. Natl. Acad. Sci. USA 107, 16946–16951 (2010). Reprints and permission information is available online at http://npg.nature.com/ 40. Gao, H., Wu, B., Le, Y. & Zhu, Z. Homeobox protein VentX induces p53- reprintsandpermissions/ independent apoptosis in cancer cells. Oncotarget 7, 39719–39729 (2016). 41. Schmidt, A. et al. Human macrophages induce CD4(+)Foxp3(+) regulatory Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in T cells via binding and re-release of TGF-beta. Immunol. Cell Biol. 94, published maps and institutional affiliations. 747–762 (2016). 42. Kamada, N. et al. Unique CD14 intestinal macrophages contribute to the pathogenesis of Crohn disease via IL-23/IFN-gamma axis. J. Clin. Invest. 118, 2269–2280 (2008). Open Access This article is licensed under a Creative Commons 43. Rogler, G. et al. Isolation and phenotypic characterization of colonic Attribution 4.0 International License, which permits use, sharing, macrophages. Clin. Exp. Immunol. 112, 205–215 (1998). adaptation, distribution and reproduction in any medium or format, as long as you give 44. Tiemessen, M. M. et al. CD4+CD25+Foxp3+regulatory T cells induce appropriate credit to the original author(s) and the source, provide a link to the Creative alternative activation of human monocytes/macrophages. Proc. Natl. Acad. Commons license, and indicate if changes were made. The images or other third party Sci. USA 104, 19446–19451 (2007). material in this article are included in the article’s Creative Commons license, unless 45. Guo, F., Iclozan, C., Suh, W. K., Anasetti, C. & Yu, X. Z. CD28 controls indicated otherwise in a credit line to the material. If material is not included in the differentiation of regulatory T cells from naive CD4 T cells. J. Immunol. 181, article’s Creative Commons license and your intended use is not permitted by statutory 2285–2291 (2008). regulation or exceeds the permitted use, you will need to obtain permission directly from 46. Mittal, V. K., Bhullar, J. S. & Jayant, K. Animal models of human colorectal the copyright holder. To view a copy of this license, visit http://creativecommons.org/ cancer: Current status, uses and limitations. World J. Gastroenterol. 21, licenses/by/4.0/. 11854–11861 (2015). 47. Wong, H. H. & Chu, P. Immunohistochemical features of the gastrointestinal tract tumors. J. Gastrointest. Oncol. 3, 262–284 (2012). © The Author(s) 2018 Acknowledgements We thank Marc Kirschner for discussions and critical reading of the manuscript. Z.Z. was supported by grants from the National Institutes of Health, Department of Defense, American Cancer Society, and research funds from Brigham and Women’s Hospital. 10 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

The homeobox protein VentX reverts immune suppression in the tumor microenvironment

Free
10 pages
Loading next page...
 
/lp/springer_journal/the-homeobox-protein-ventx-reverts-immune-suppression-in-the-tumor-FoY6AW059C
Publisher
Nature Publishing Group UK
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
D.O.I.
10.1038/s41467-018-04567-0
Publisher site
See Article on Publisher Site

Abstract

ARTICLE DOI: 10.1038/s41467-018-04567-0 OPEN The homeobox protein VentX reverts immune suppression in the tumor microenvironment 1 2 3 1 Yi Le , Hong Gao , Ronald Bleday & Zhenglun Zhu Immune suppression in the tumor microenvironment (TME) is a central obstacle to effective immunotherapy. Tumor-associated macrophages (TAMs) are key components of the TME. Although TAMs have been viewed as an ideal target of intervention to steer immunity in cancer treatment, the approach has been hampered by the lack of knowledge of how TAM plasticity is controlled by cell intrinsic factors. VentX is a homeobox protein implicated in proliferation and differentiation of human hematopoietic and immune cells. Using clinical samples obtained from cancer patients, we find that VentX expression is drastically reduced in TAMs. We show here that VentX promotes M1 differentiation of TAMs, and that VentX- regulated TAMs, in turn, revert immune suppression at the TME. Using a NSG mouse model of human colon cancers, we demonstrate that VentX regulates TAM function in tumor- igenesis in vivo. Our findings suggest a mechanism underlying immune suppression at TME and potential applications of VentX-regulated TAMs in cancer immunotherapy. 1 2 Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts, USA. Department of Medicine, Tufts Medical Center, Boston, Massachusetts, USA. Department of Surgery, Brigham and Women’s Hospital, Boston, Massachusetts, USA. Correspondence and requests for materials should be addressed to Z.Z. (email: zzhu@partners.org) NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 he role of immunity in oncogenesis has been long appre- function in tumorigenesis in vivo. Taken together, our studies ciated and increasingly exploited. Nevertheless, the efficacy suggested a key role of VentX in regulating TAM plasticity and Tof cancer immunotherapy remains limited, especially for immune status at TME. Targeting VentX, therefore, may open solid tumors. Immune suppression at the tumor microenviron- novel venue of cancer immunotherapy. ment (TME) has been accounted for cancer evasion of immune destruction and regarded as a potential venue of intervention . Results However currently, mechanisms that underlie immune suppres- VentX expression is decreased in TAMs. In advanced tumors, 2, 3 sion at TME remained largely elusive . Tumor-associated 7, 18, 19 TAMs display a pro-tumor M2-like phenotype . The macrophages (TAMs) are key components of TME and have plasticity of TAMs has been well appreciated, however, the been implicated in growth, invasion, and metastasis of nearly all transcriptional machinery that controls the TAM polarization tumors. Derived from circulating monocytes, TAMs display a remains largely unknown. Using TAMs isolated from discarded broad spectrum of phenotypes, ranging from the pro- specimens from colon cancer resection as well as macrophages inflammatory M1-like phenotype in early stages of some from normal mucosa 10 cm away from the tumor sites, we sought 4, 5 tumors to the M2-like phenotype in most advanced tumors . to explore the potential involvement of VentX in TAM plasticity. The M2-like TAMs show an elevated expression of interleukin Consistent with prior findings, we showed that TAMs express (IL)-10, matrix metallopeptidase (MMP), and vascular endothe- significant higher levels of cell surface markers associated with lial growth factor (VEGF), but decreased expression of pro- 20–23 M2 phenotypes, such as CD163 and CD206 (Fig. 1a, b). The inflammatory tumoricidal cytokines, cytotoxic inducible nitric elevated expression of M2 surface markers was accompanied by a oxides (iNOs) and reactive oxygen intermediates (ROIs), and are decrease of M1 surface markers, such as CD40 and7 CD80 thought to be supportive of tumor growth . Besides their func- (Supplementary Fig. 1); however, there was no significant dif- tions in promoting tumorigenesis, TAMs also help generate a ference in the expression of non-discriminating myeloid marker pro-tumor immune suppressive milieu at TME by altering the CD33 in TAMs and control macrophages (Fig. 1a, b). Similar to composition and function of tumor-infiltrating lymphocytes the cell surface markers, there is an increased expression of 3, 6 (TILs) . The plasticity of TAMs has been well recognized. It has markers associated with M2 differentiation but decreased been proposed that by converting the pro-tumor M2-like TAMs expression of M1 markers in TAMs (Supplementary Fig. 2). To into the anti-tumor M1-like phenotype, the TAMs might be determine whether VentX has a role in TAM plasticity, we 7, 8 converted into an effective modality of anti-tumor therapy . quantified VentX expression in TAMs by quantitative reverse- Nevertheless, this potential application of modulating TAM transcription PCR (qRT-PCR) and found that, in comparison function in cancer treatment has been hampered by our ignor- with its expression in control macrophages, VentX expression is ance of how TAMs plasticity is controlled by cell intrinsic decreased about 77% in TAMs (Fig. 1c). The decreased expression 5, 9 factors . of VentX in TAMs was further verified by western blot analysis, Dorsoventral axis formation represents coordinated cell pro- using VentX-specific antibodies (Fig. 1d). liferation and differentiation during early vertebrate embry- ogenesis. In trying to understand the molecule basis of the process, our recent work led to the appreciation of the Xenopus VentX drives TAM towards M1 phenotype. To determine homeobox protein Xom, a unstable protein of the ventral whether VentX is involved in TAM plasticity, we examined the BMP4 signaling pathway , as a novel lymphoid enhancing correlation of VentX expression with TAM phenotypes. Using factor/T-cell factor (LEF/TCF)-associated factor that antagonizes .we lipopolysaccharide (LPS) as an M1 phenotype inducer β-catenin of the dorsal Wnt signaling pathway during dorso- found that VentX expression is significantly elevated in TAMs ventral polarization of Xenopus embryos . To explore the after them being exposed to LPS (Fig. 2a). The elevated expres- potential clinical relevance of the findings, we performed sion of VentX is accompanied by elevated expression of M1 sequence homology search and functional analysis, which led to markers in TAMs, including the secretion of inflammatory the identification of human homeobox protein VentX as a human cytokines and cytotoxic iNOs (Fig. 2b, c). Similar to LPS, we 12, 13 homolog of Xom . We found that VentX is primarily found that TAMs can also be activated by pro-inflammatory expressed in hematopoietic cells and controls proliferation and cytokines, such as interferon-γ (IFNγ) (Supplementary Fig. 3). To differentiation of hematopoietic cells from early ontogenesis to determine whether VentX has a required role in TAM plasticity, 12, 14–16 terminal differentiation . Interestingly, comparative we examined the effects of knockdown VentX on TAM pheno- genomic studies showed that VentX is preserved in primates and types. As shown in Fig. 2d, treatment of TAM with VentX human but lost in mice since the evolutionary divergence of morpholino oligos (VentX-MO) led to around 80% reduction of rodent and primate lineages . Using an in vitro culture of human VentX expression. The decreased VentX expression, which is monocyte-derived macrophage model, we found that VentX verified by western blot analysis (Supplementary Fig. 4), is promotes and is required for M1 but not M2 activation . These accompanied by decreased secretion of inflammatory cytokines findings inspired us to explore whether VentX has a role in and cytotoxic iNOs (Fig. 2d–f). Corresponding to the changes of regulating plasticity and function of TAMs. M1 phenotype, there is also a change of the M2 phenotype. As Using clinical samples obtained from primary colorectal cancer shown in Fig. 2g, we found that the effect of LPS on the patients, our current studies showed that VentX expression is expression of CD206, a M2 marker highly expressed in TAMs drastically reduced in TAMs in comparison with its expression in (Supplementary Fig. 1a) , was abolished by VentX-MO macrophages isolated from normal mucosa of the same patients. treatment. We found that TAM VentX expression profile correlates with The correlation between VentX expression levels and TAM TAM phenotypes, and that ectopic expression of VentX con- phenotypes prompted us to explore the potential direct effect of verted the M2-like phenotype of TAMs into M1-like phenotype. VentX on TAM phenotype. Using transfection studies, we found Moreover, we found that VentX-regulated TAMs revert immune that TAMs transfected with GFP-VentX displayed a characteristic suppression at TME by inhibiting regulatory T-cell (Treg) dif- M1 morphology with elongated/fibroblast-like cell shape, whereas ferentiation and promoting CD8 TIL activation. Using a NOD there was no such changes in TAMs transfected with the control scid γ (NSG) mouse model of patient-derived xenograft (NSG- green fluorescent protein (GFP) (Fig. 3a). In comparison with the PDX) of colon cancers, we showed that VentX-modulated-TAMs control GFP-transfected TAMs, the surface expression of M1 2 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE a b Normal mucosal CD68 CD163 CD206 CD33 Normal mucosal Tumor 60 ** ** ** Tumor CD163 CD206 CD33 CD68 Fluorescence intensity 1.2 Normal Tumor** Normal 0.8 Tumor Mean 1.000 0.2260 Std. deviation 0.0 0.1146 0.6 0.4 0.2 Patient a Patient b Patient c N TT N N T WB:VentX 25 kDa WB:β-actin 37 kDa Fig. 1 VentX expression is downregulated in TAMs. a, b FACS analysis and the percentage of surface expression of CD33, CD68, CD163, and CD206 on macrophages isolated from normal mucosa and tumors. Shaded peaks indicated isotope control. Representative figures of five independent experiments were shown, n= 5. Error bars represent SD and paired Student’s t-test was performed. **p< 0.01. c Paired comparison of RT-PCR measurement of VentX mRNA expression in macrophages isolated from normal control tissues and TAMs of 15 patients. The relative VentX mRNA expression levels in normal macrophages were arbitrarily designated as 1. Paired Student’s t-test was performed, **p< 0.01 TAMs vs. normal macrophages. Data shown in the box are mean ± SD of 15 patients. d Western blot analysis of endogenous VentX protein levels in macrophages isolated from normal control tissues and TAMs of three patients marker CD40, CD80, and CD86 was significantly increased in VentX-regulated TAMs modulate TILs differentiation. TILs TAMs transfected with GFP-VentX (Fig. 3b). In addition, the have key roles in anti-tumor immunity . The regulatory CD4 secretion of pro-inflammatory cytokines tumor necrosisfactor-α +CD25+ Treg and tumoricidal CD8+ TILs are key components (TNFα), IL-1β, and IL-12 were significantly increased, whereas of TME. The ratio of CD4+CD25+ Tregs and CD8+ TILs, as secretion of the M2 cytokine, IL-10, was significantly reduced well as functional defects of CD8 TILs, have been implicated in 6, 27–30 (Fig. 3c). Using qRT-PCR analysis, we showed a characteristic pathogenesis and prognosis of solid tumors . Consistent increased expression of of M1 genes, such as IL-1β, IL-6, IL-12, with prior findings , we found that there was a significantly TNF-α, and iNOs (Fig. 3d), and decreased expression of M2 increased number of the CD4+CD25+/CD4+Foxp3+ Treg cells genes, such as CCL18, MMP9, VEGFA, and Arg1 in TAMs and a significantly decreased number of CD8+ TILs in tumor transfected with GFP-VentX (Fig. 3e). In comparison, there was tissues (Supplementary Fig. 6 and 7). As CD4+ no significant changes of non-discriminating macrophages CD25+/CD4+Foxp3+ Treg cells are derived from markers, such as the CD33 and CD68 upon the transfection of CD4+ T cells, to explore the potential involvement of VentX- GFP-VentX in TAMs (Supplementary Fig. 5a, 5b). The changes regulated TAMs in compositions of TILs, we sought to determine of cell surface markers are associated with changes of intracellular whether VentX-regulated TAMs modulate CD4+ T-cell differ- signaling molecules such as the STAT1 and STAT3, which are entiation. To attend our goal, CD4+ T cells isolated from per- associated with M1 and M2 phenotype, respectively (Supplemen- ipheral blood were co-cultured with autologus TAMs transfected tary Fig. 5c) . Taken together, our data suggested that VentX with either GFP-VentX or control GFP for 5 days. As shown in functions as a master switch of TAM plasticity and drives TAMs Fig. 4a, co-culture of CD4+ T cells with TAMs led to significant toward M1 phenotypes through alternating intracellular signaling induction of CD4+CD25+/CD4+Foxp3+ cells. In contrast, pathways involved in the process. ectopic expression of VentX in TAMs abolished the induction. To NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 3 | | | Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 CD33 Patient 6 CD68 Patient 7 Patient 8 CD163 Patient 9 CD206 Patient 10 Patient 11 Patient 12 Patient 13 Patient 14 Patient 15 Cell counts Cell counts Relative VentX mRNA level % of positive cells ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 a b c TAMs ** 1.2 1200 ** ** 0.8 0.4 0 0 0 TAMs TAMs + 0 24 h 48 h TAMs TAMs + LPS LPS Incubation time of LPS de f 1.2 50 0.8 0.6 ** ** 0.4 ** 300 10 0.2 0 0 Control VentX Control VentX Control VentX MO MO MO MO MO MO LPS stimulation LPS stimulation LPS stimulation Control LPS Control-MO + LPS VentX-MO + LPS 0.63 49.3 0 1.38 0 1.71 0.86 35.8 45.8 4.30 51.0 47.6 51.8 46.5 53.4 9.94 CD68 Fig. 2 VentX regulates TAM plasticity. a VentX expression in TAMs after LPS stimulation. Isolated TAMs were cultured in RPMI medium and stimulated with LPS. VentX mRNA levels were measured with real-time PCR at indicated times. b TNF-α and c nitrate level from TAMs after LPS treatment were determined by ELISA and Griess reagent. d Effects of VentX-MO on VentX expression. Isolated TAMs were transfected with MO oligonucleotides targeting VentX or control. Twenty-four hours post transfection, LPS was added to culture medium. The cells were collected 48 h later and VentX expressions were determined by real-time RT-PCR. e TNF level and f nitrate level from VentX-MO- or control-MO-treated TAMs after LPS stimulation. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05, **p< 0.01. g Effects of VentX-MO or control-MO on percentage of CD206+CD68+ TAMs after LPS stimulation. Representative data of at least three independent experiments were shown corroborate the functional relevance of the findings, CD4+ T cells control GFP for 5 days. CD4+CD25+Treg cells and CD8+ TILs isolated from normal tissues were co-cultured with autologus were then isolated and quantified by flow cytometry, following 32, 33 TAMs transfected with either GFP-VentX or control GFP for established protocols (Supplementary Fig. 9) . As shown in 5 days. We found that TAMs induced T-cell expression of inhi- Fig. 5a, b, co-culture of the tumors with GFP-VentX-modified bitory IL-13, but the induction function was abolished by ectopic TAMs led to a significant decrease of the CD4+ expression of VentX (Supplementary Fig. 8). To further deter- CD25+Tregs and a significant increase of the CD8+ TILs. In mine whether VentX-regulated TAMs modulate TIL function, addition to the significant increase in the number of CD8+ TILs, CD8+ TILs were co-cultured with TAMs transfected with GPF- there is a significant increase in the expression of IFNγ and VentX or control GFP. As shown in Fig. 4b, co-culture of CD8+ granzyme B in CD8+ cells in tumor tissues after being incubated TILs with VentX-modified TAMs led to significant activation of with VentX-modified TAMs (Fig. 5c). CD8+ TILs, as indicated by CD8+ activation markers, such as IFNγ and granzyme B. Taken together, our data suggested that VentX regulates TAM function and tumorigenesis in vivo. VentX-regulated TAMs modulate TIL differentiation and Converting the M2 phenotype of TAM into the M1 phenotype function. 5, 8 has been viewed as a promising venue for cancer treatment . Our findings that VentX promotes M1 polarization of TAM VentX-regulated TAMs modulate composition of TILs in prompted us to explore whether VentX-modulated TAM plasti- TME. Our findings that VentX-regulated TAMs modulate dif- city has a role in tumorigenesis in vivo. As VentX does not have a ferentiation and function of T cells prompted us to test the mouse homolog , to achieve our goal, we adapted a NSG mouse hypothesis that VentX-regulated TAMs control immune status at model, which support heightened engraftment of human hema- 34 35 TME by modulating the differentiation and function of TILs. To topoietic cells . Following established protocol , a NSG-PDX of test this hypothesis, en bloc primary tumor tissues were co- human colon cancers was generated by engrafting small pieces of cultured with autologous TAMs transfected with GPF-VentX or primary human colon cancer tissues into subcutaneous space on 4 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | Relative VentX Relative VentX CD206 mRNA level mRNA level TNFα (pg/mL) TNFα (pg/mL) Nitrite (unit) Nitrite (unit) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE ab 4.63 0.39 0.61 10.3 7.05 10.4 Green fluorescence TNFα IL-1β IL-12p70 IL-10 120 * 400 400 40 200 0 0 0 0 d e GFP GFP iNOS ARG1 GFP-VentX * GFP-VentX TNF CCL18 IL-12 MMP9 IL-6 * VEGFA IL-1b 024 68 0.0 0.3 0.6 0.9 1.2 1.5 mRNA expression level mRNA expression level Fig. 3 VentX polarizes TAMs towards M1 phenotype. a Morphology of TAMs transfected with GFP or GFP-VentX. Scale bar: 20 μm. b Surface expression of CD40, CD80, and CD86 in TAMs transfected with GFP or GFP-VentX as determined by flow cytometry analysis. Representative data of three independent experiments were shown. c ELISA measurement of secretion of pro-inflammatory cytokines in TAMs transfected with GFP and GFP-VentX. Results represent mean ± SD of four independent experiments and paired Student’s t-test was used. Expression of M1 (d) and M2 (e) macrophage markers in TAMs transfected with GFP or GFP-VentX as determined by qRT-PCR. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05 the abdominal side of the NSG mice. The growth of the tumors in Discussion the NSG mice was observed for 8 weeks (Fig. 6a). The tumors Tumor immunology is now a promising field for exploration of were then dissected out and sectioned and the growth of human pathogenesis and treatment of cancers. Macrophages have colon cancers was confirmed by hematoxylin and eosin (H&E) executor role in immunity and have been recognized as key and CK20 staining (Fig. 6b). To test the potential effects of components of tumor mass. The search for factors that can be VentX-regulated TAMs on tumor growth 1 week post implan- manipulated to steer macrophage plasticity has proved to be a tation of fragments of colon cancers, the NSG-PDX mice were fundamental challenge. Signaling pathways such as the JAK/ tail-vein injected with TAMs transfected with GFP-VentX or STAT1, JNK/STAT6, AKT1/AKT2, and PI3Kγ, transcriptional control GFP. As shown in Fig. 6c, we found that, in comparison regulators such as PPARγ and PPARδ,NFκB, C/EBP, and IRFs, with the GFP-transfected TAMs, GFP-VentX-transfected TAMs as well as microRNAs such as microRNA-155 and miR-142-3p exerted strong inhibition on tumor growth in the NSG-PDX have been shown to modulate macrophage 9, 24, 36, 37 mice. To further determine whether the inhibition of tumor- differentiation . Nevertheless, whether these factors can igenesis in the NSG-PDX mice is related to VentX-regulated be targeted to turn TAMs into a tumoricidal cells remains TAM polarity 1 week post implantation of fragments of colon unclear. As a unique human hematopoietic transcriptional factor cancers, the NSG-PDX mice were tail-vein injected with in vitro that does not have a murine homolog, VentX was found to have M1-differentiated TAMs transfected with either VentX-MO or essential role in controlling proliferation and differentiation of control-MO. Consistent with the results of over expression stu- hematopoietic cells and function as a p53-independent tumor 12, 15, 16, 38–40 dies, the M1-TAMs transfected with control-MO exerts strong suppressor . Data of our current study suggested inhibition of tumor growth, but the inhibition was abolished by VentX controls TAM plasticity, which in turn, reverts immune knocking-down VentX expression with VentX-MO (Fig. 6d). suppression at the TME through regulating TIL differentiation NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 5 | | | GFP GFP-VX GFP GFP-VX GFP GFP-VX GFP GFP-VX GFP-VentX GFP pg/mL pg/mL GFP GFP-VentX CD40 CD40 pg/mL CD80 CD80 pg/mL CD86 CD86 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 Co-culture of CD4 with/without modified TAMs No TAM GFP GFP-VentX ** 2.99 9.19 3.14 No TAM GFP GFP-VentX CD4 No TAM GFP GFP-VentX ** 3.98 8.41 4.86 CD4 No TAM GFP GFP-VentX Co-culture of CD8 with modified TAMs GFP GFP-VentX 37.2 12.1 ** ** GFP GFP-VentX 9.87 43.5 IFN-γ Granzyme CD8 Fig. 4 VentX modulates TAM induction of TILs differentiation and activation. a VentX modulates TAM induction of T-cell differentiation. CD4+ T cells were incubated with autologus TAMs transfected with GFP or GFP-VentX for 5 days. Differentiation of the CD4+ T cells were determined by CD25 and Foxp3 analysis. The results were mean ± SD of five independent experiments. b VentX modulates TAM effects on CD8 activation. CD8 TILs isolated from tumors were incubated with autologus TAMs transfected with GFP or GFP-VentX for 5 days. CD8+TIL activation was measured by FACS analysis of IFNγ and granzyme B. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05, **p< 0.01 and function, which was supported by both in vitro and in vivo of colon cancer specimens used for this study were listed in Table 1. Around 5–10 g tissues were collected from tumor mass, or normal mucosa 10 cm away from tumor studies (Figs. 4, 5 and Supplementary Fig. 11). Interestingly, mass. consistent with prior findings that macrophages may affect T-cell differentiation indirectly , we found that TAMs do not need to Preparation of intraepithelial lymphocytes. Lymphocytes were isolated following accumulate in significant numbers inside tumor tissues to exert 14, 42 previously described techniques with modification . In brief, dissected fresh its function (Supplementary Fig. 12). Our findings revealed a 2+ 2 mucosa and tumor mass were rinsed in 10-cm Petri dish with Ca -free and Mg novel mechanism underlying immune suppression at TME and -free Hank’s balanced salt solution (HBSS) (Life Technologies) containing 2% fetal bovine serum (FBS) and 2 mM Dithiothreitol (DTT) (Sigma-Aldrich). The suggested the potential application of VentX-regulated TAMs as a mucosa and tumor were then cut into around 0.1 cm pieces by a razor blade and novel modality of immunotherapy, especially for solid tumors, incubated in 5 mL HBSS containing 5 mM EDTA (Sigma-Aldrich) at 37 °C for 1 h, which are refractory to current available immunotherapy. Cur- then passed through a gray-mesh (100 μm). The flowthrough contains intrae- rently, the cause of decreased VentX expression in TAMs and the pithelial lymphocytes and epithelial cells and was analysis by a flow cytometer. potential clinical application of VentX-modulated TAMs in cancer treatment remain to be further defined. Isolation of macrophages from tumor mass and normal mucosa. Lamina pro- pria mononuclear cells (LPMCs) were isolated following established protocol, which 14, 42, 43 does not lead to activation of macrophages .Briefly, normal mucosa and tumor 2+ 2+ Methods tissues were rinsed with HBSS and then incubated in HBSS (with Ca and Mg ), Collection of colon tissue samples. A total of 42 patients with colon cancer, who containing 2% FBS, 1.5 mg/mL Collagenase D (Roche), 0.1 mg/mL Dnase I at 37 °C were scheduled for surgical resection at Brigham and Women’s Hospital, were for 1 h. Digested tissues were then passed through a gray-mesh (70 μm) filter. The consented to have a portion of tissues and blood collected for research purposes. flowthrough were collected, washed, and resuspended in a RPMI 1640 medium. The All patients signed an informed consent document that was approved by the cells were layered on Ficoll-Paque Plus media (GE Healthcare), and then centrifuged Institutional Review Board of Brigham and Women’s Hospital. The characteristics at 2000 r.p.m. for 30 min without brake. LPMCs at the interface were collected. 6 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | Foxp3 CD25 Granzyme B IFN-γ % of activated CD8+ T cells CD4+Foxp3+ (%) CD4+CD25+ (%) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE Co-culture of tumor with modified TAMs GFP GFP-VentX 12.8 46.2 14.8 28.7 28.0 13.0 49.2 7.23 CD4 GFP GFP-VentX Co-culture of tumor with modified TAMs GFP GFP-VentX 40 CD8+ CD8+ 27.8 13.5 10 GFP GFP-VentX CD8 GFP GFP-VentX 13.0 64.6 GFP GFP-VentX 15.1 74.9 IFN-γ Granzyme B CD8 Fig. 5 VentX-regulated TAMs modulate composition and function of TILs in tumor microenvironment. a, b Composition of CD4+CD25+ Tregs (a) and CD8+ TILs (b) in tumors after en bloc co-culture of tumor tissues with autologus TAMs transfected with GFP or GFP-VentX for 5 days. c CD8 TILs were isolated from tumor tissues after being incubated with autologus TAMs transfected with GFP or GFP-VentX for 5 days. CD8+TIL activation was measured by FACS analysis of IFNγ and granzyme B. Data shown are mean ± SD of three independent experiments and paired Student’s t-test was performed. *p< 0.05 Normal mucosal macrophages and TAMs were further purified from LPMCs using warmed Human Monocyte Nucleofector Medium containing 2 mM glutamine and EasySep™ Human Monocyte/Macrophage Enrichment kit without CD16 depletion 10% FBS. Cells were then resuspended into complete RPMI medium and treated (StemCell Technologies), according to the manufacturer’s instructions. The isolation with appropriate cytokines to induce differentiation into macrophages. process does not lead to activation of macrophages and the purify of isolated intestinal 14, 42, 43 macrophages was routinely more than 95% . More than 98% of cells isolated by the techniques were viable by propidium iodide staining. FACS analysis. Phenotypic analysis of TAMs and other lymphocytes was per- formed using flow cytometry after immunolabeling of cells with fluorescence dye- conjugated antibodies. The following antibodies were used: Phycoerythrin (PE)- Transfection assays. Transfection of GFP-VentX and GFP into blood macro- conjugated anti-CD3 (OKT3), -CD25 (BC96), -CD14 (61D3), -CD68 (Y182A), phages or TAMs were carried out through lipofectamine 2000 (Life Technologies) -CD163 (GH161), and -CD206, and fluorescein isothiocyanate (FITC)-conjugated according to the manufacturer’s protocol. Forty-eight hours after transfection, cells anti-CD4 (RPA-T4) and -CD33 (HIM3-4), and Allophycocyanin (APC)-con- were filtered through a 70 μm filter for cell sorting. GFP-positive cells were sorted jugated anti-CD8 (OKT8) and -CD4 (OKT4) (eBioscience, Inc). Intracellular by BD FACSAria II under the Baker Bio-Protect Hood in a sterile condition. After staining of Foxp3 (236 A/E7), IFNγ (4 S.B3), and Granzyme B (GB11) was per- sorting, cells were cultured in RPMI 1640 complete medium. formed with PE-conjugated antibodies following the protocol provided by manu- facturer. Isotope control labeling was performed in parallel. Antibodies were VentX knockdown. Colon TAMs or human primary monocytes were transfected diluted as recommended by the supplier. Labeled cells were collected on FACScan flow cytometer with Cell-Quest software (BD Biosciences) and analyzed by FlowJo with Morpholino oligonucleotides (MO) (Gene Tools, LLC, Philomath, OR) using the Human Monocyte Nucleofector Kit (Lonza, Walkersville, MD) as previously software. Results are expressed as the percentage of positive cells. 14 6 described . Briefly, 5 × 10 cells were resuspended into 100 µl nucleofector solu- tion with 2.5 nmol of either VentX-MO oligonucleotides (VentX-MO: 5′-TACT- CAACCCTGACATAGAGGGTAA-3′ or a standard control-MO oligonucleotides Cytokine measurement. Levels of IL-1β, IL-10, IL-13, TNF-α, and IL-12p70 were and electroporated with the Nucleofector II Device (Lonza). Cells were then quantified using enzyme-linked immunosorbent assay kits obtained from eBios- immediately removed from the device and incubated overnight with 1 ml pre- ciences. Analyses were conducted according to the manufacturer’s instructions. NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 7 | | | Granzyme B IFN-γ Cell counts CD25 % of activated CD8+ T cells % of CD8+ TIL cells % of CD4+CD25+ cells ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 a b Mice 1 Mice 2 Mice 3 Tumor 0 246 8 Weeks cd M1-TAMs-VentX-MO TAMs-GFP M1-TAMs-Control-MO TAMs-GFP-VentX 2500 2500 ** 1500 1500 ** 1000 1000 ** 500 500 ** ** ** 0 0 02468 02468 Weeks Weeks Fig. 6 VentX-regulated TAMs modulate tumorigenesis in vivo. a A NSG-PDX model of human colon cancers. Small pieces of colon cancer tissues were transplanted into subcutaneous space on the abdominal side of NSG mice and the growth of the tumors were observed for 8 weeks. Data shown were results of seven independent replicates of the experiments. b H&E and CK20 staining of tumors resected from three NSG-PDX mice. Scale bar: 50 μm. c Growth curve of tumors in NSG-PDX mice treated with TAMs transfected with GFP-VentX or control GFP. Results shown represent mean ± SD of three independently replicated experiments, n = 3, and paired Student’s t-test was performed, **P < 0.01. d Growth curve of tumors in NSG-PDX mice treated with M1-TAMs transfected with VentX-MO or control-MO. Results shown represent mean ± SD of seven independently replicated experiments, n = 7, and paired Student’s t-test was performed, **P < 0.01. Photos of excised tumors at the end of experiments were shown. Scale bar: 5 mm Quantitative RT-PCR. Total RNA was isolated by the TRIzol reagent (Life blood CD4 cells were enriched by using Easysep human CD4-negative selection kit Technologies) and RNA amounts were measured by NanoDrop 2000 (Thermo following the manufacturer’s instructions (StemCell Technologies). GFP-VentX or 6 6 Scientific). An equal amount of RNA was used for first-strand complementary GFP-transfected TAMs (0.5 × 10 ) were incubated with 5 × 10 of CD4 in com- DNA synthesis with SuperScript III First-Strand Synthesis System (Life Technol- pleted RPMI 1640 at 37 °C, 5% CO for 5 days. Cells were stained with CD4-FITC ogies) according to the manufacturer’s protocol. To amplify VentX cDNA with and CD25-PE, or permeabilized and stained with CD4-FITC and Foxp3-PE, then conventional PCR, we used the AccuPrime Taq DNA polymerase system (Life analyzed by a flow cytometer. For CD8+ TIL activation assay, CD8+ TILs were Technologies) following the manufacturer’s instructions. Quantitative measure- enriched from intraepithelial lymphocytes by CD8+ T-cell Enrichment Kit 6 6 ment of VentX and other genes cDNA were carried out with SYBR Green on a (StemCell Technologies). Cells (5 × 10 ) were then incubated with 0.5 × 10 of LightCycler (480 Real-Time PCR System; Roche). The primers used list in Sup- GFP-VentX or GFP-transfected TAMs for 5 days, followed by staining and analysis plementary Table 1. Relative expression profiles of mRNAs were then calculated with flow cyotmetry. using the comparative Ct method (DDCT method). Co-cultures of tumors and TAMs. Tumor mass were washed with 1 × phosphate- Western blot analysis. Western blot analysis was performed as described pre- buffered saline (PBS) buffer plus antibiotics and then cut into 0.5 cm pieces. viously .Briefly, total cells were lysed in 1 × RIPA buffer (Abcam, Inc.) mixed with Around 80 mg of tissues were mix cultured with 0.5 × 10 of GFP-VentX or GFP- protease inhibitor cocktails (Cell Signaling Technology). Proteins were resolved by transfected TAMs of same patient in 2 mL of RPMI 1640 completed medium, 4–15% Tris-Glycine Gel (Bio-Rad) electrophoresis. Primary antibodies used supplemented with 2.5% antibiotic–antimycotic solution (Cellgro, Manassas, VA). included GFP (eBioscience 14–6674, 1:1000), VentX (Abcam, Inc. ab105352, The cultures were incubated at 37 °C, 5% CO for 5 days. The tissues were then 1:500), and β-actin (Cell Signaling Technology 4967, 1:2000). subjected to cell isolation and analyzed by a flow cytometry or immunohis- tochemistry studies. Arginase activity and NO assays. Arginase activity was quantified in cell lysates by measuring the production of urea using the QuantiChrom Arginase Assay Kit, NSG-PDX model of human colon cancers. Animal models of primary human following the manufacturer’s instuctions (DARG-200, BioAssays Systems). Nitrite colon cancers were developed as described previously . Briefly, 8-week-old NOD. concentrations in culture supernatants were determined using Griess reagent kit scid tm1Wjl Cg-Prkdc Il2rg /SzJ mice (commonly known as NSG mice) were purchased (Molecular Probes, Eugene, OR), as described previously . from The Jackson Laboratory and maintained under specific pathogen-free con- ditions. All animal experiments were approved by the Institutional Animal Care Treg cell induction and CD8+ TIL cell activation. Treg cell inductions were and Use Committee at Harvard Medical School. Colon tumors were cut into 44, 45 performed using previously described methods with modification . Briefly, the around 0.5 cm and surgically seeded in subcutaneous space of abdominal side of 8 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | | TAMs- GFP TAMs- GFP-VentX M1-TAMs- VentX-MO M1-TAMs- Control-MO 3 3 Tumor volume (mm ) Tumor volume (mm ) CK20 H&E Tumor volume (mm ) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 ARTICLE 6. Schreiber, R. D., Old, L. J. & Smyth, M. J. Cancer immunoediting: integrating Table 1 Characteristics of colon cancer patients used for immunity’s roles in cancer suppression and promotion. Science 331, study 1565–1570 (2011). 7. Sica, A. et al. Macrophage polarization in tumour progression. Semin. Cancer Biol. 18, 349–355 (2008). Characteristics Number (n = 42) (%) 8. Noy, R. & Pollard, J. W. Tumor-associated macrophages: from mechanisms to Age, years therapy. Immunity 41,49–61 (2014). ≤ 40 5 11.9 9. Lawrence, T. & Natoli, G. Transcriptional regulation of macrophage 40–60 16 38.1 polarization: enabling diversity with identity. Nat. Rev. Immunol. 11, 750–761 ≥ 60 21 50.0 (2011). Sex 10. Zhu, Z. & Kirschner, M. Regulated proteolysis of Xom mediates dorsoventral Male 24 57.1 pattern formation during early Xenopus development. Dev. Cell. 3, 557–568 (2002). Female 18 42.9 11. Gao, H., Wu, B., Giese, R. & Zhu, Z. Xom interacts with and stimulates Pathology type transcriptional activity of LEF1/TCFs: implications for ventral cell fate Adenocarcinoma 42 100 determination during vertebrate embryogenesis. Cell Res. 17, 345–356 (2007). Stage 12. Gao, H. et al. VentX, a novel lymphoid-enhancing factor/T-cell factor- I8 19 associated transcription repressor, is a putative tumor suppressor. Cancer Res. II 14 33.3 70, 202–211 (2010). III 15 35.7 13. Moretti, P. A. et al. Molecular cloning of a human Vent-like homeobox gene. IV 5 11.9 Genomics 76,21–29 (2001). 14. Wu, X., Gao, H., Ke, W., Giese, R. W. & Zhu, Z. The homeobox transcription factor VentX controls human macrophage terminal differentiation and proinflammatory activation. J. Clin. Invest. 121, 2599–2613 (2011). NSG mice. TAMs were transfected with VentX-MO or control-MO, and then 15. Gao, H. et al. Suppression of homeobox transcription factor VentX promotes cultured in M1-Macrophage Generation Media (PromoCell). After 1 week of expansion of human hematopoietic stem/multipotent progenitor cells. J. Biol. xenograft, 0.8 × 10 of M1-differentiated TAMs transfected with VentX-MO or Chem. 287, 29979–29987 (2012). control-MO were injected into mice through tail vein. Tumor growth was mon- 16. Wu, X., Gao, H., Bleday, R. & Zhu, Z. Homeobox transcription factor VentX itored twice a week and measured by a caliper for 8 weeks. regulates differentiation and maturation of human dendritic cells. J. Biol. In vivo tracing of injected TAMs were carried out using carboxyfluorescein Chem. 289, 14633–14643 (2014). succinimidyl ester (CFSE)-stained TAMs, according to the manufacture’s 17. Zhong, Y. F. & Holland, P. W. The dynamics of vertebrate homeobox gene instruction (Molecular Probes). Briefly, half million of TAMs were incubated in 5 evolution: gain and loss of genes in mouse and human lineages. BMC Evol. mL of 5 μM CFSE staining solution in a 37 °C water bath for 20 min. Cells were Biol. 11, 169 (2011). then incubated with 20 ml of RPMI 1640 completed medium for 5 min to remove 18. Redente, E. F. et al. Tumor progression stage and anatomical site regulate unbounded dye. After centrifugation, cells were dissolved in PBS and then tail-vein tumor-associated macrophage and bone marrow-derived monocyte injected into NSG mice. Tumor tissues and TAMs were isolated 7 days after polarization. Am. J. Pathol. 176, 2972–2985 (2010). injection and TAMs were isolated as CFSE-positive cells. 19. Biswas, S. K., Allavena, P. & Mantovani, A. Tumor-associated macrophages: functional diversity, clinical significance, and open questions. Semin. Immunohistochemistry. Immunohistochemistry were performed following Immunopathol. 35, 585–600 (2013). established protocol . Briefly, tumor or normal mucosa were fixated in formalin 20. Quatromoni, J. G. & Eruslanov, E. Tumor-associated macrophages: function, (Fisher Scientific Company, Kalamazoo, MI). The tissues were then embedded in phenotype, and link to prognosis in human lung cancer. Am. J. Transl. Res. 4, paraffin and sectioned. CK20 (Dako, Carpinteria, CA, clone Ks20.8, 1:50) and H&E 376–389 (2012). staining were performed at research pathology cores at Dana-Farber/Harvard 21. Kurahara, H. et al. Significance of M2-polarized tumor-associated macrophage cancer center. Microscopic analysis was performed with a Nikon Eclipse Ti in pancreatic cancer. J. Surg. Res. 167, e211–e219 (2011). fluorescence microscopy. Images were captured at an original magnification of × 40 22. Shabo, I., Stal, O., Olsson, H., Dore, S. & Svanvik, J. Breast cancer expression using a color camera applying the NIS Elements imaging software (Nikon). of CD163, a macrophage scavenger receptor, is related to early distant Brightness and contrast for representative images were adjusted equally among recurrence and reduced patient survival. Int. J. Cancer 123, 780–786 (2008). groups. 23. Zhang, Y., Sime, W., Juhas, M. & Sjolander, A. Crosstalk between colon cancer cells and macrophages via inflammatory mediators and CD47 promotes Statistical analysis. Student’s t-test was used for statistical analysis. Data are tumour cell migration. Eur. J. Cancer 49, 3320–3334 (2013). presented as mean ± SD. The level of significance is indicated by the p-value. In all 24. Cai, X. et al. Re-polarization of tumor-associated macrophages to pro- figures, levels of statistical significance are indicated as *p < 0.05 and **p < 0.01. inflammatory M1 macrophages by microRNA-155. J. Mol. Cell Biol. 4, 341–343 (2012). Data availability. The authors declare that all the other data supporting the 25. Luo, Y. et al. Targeting tumor-associated macrophages as a novel strategy findings of this study are available within the article and its Supplementary against breast cancer. J. Clin. Invest. 116, 2132–2141 (2006). Information files and from the corresponding author upon reasonable request. 26. Sica, A. & Bronte, V. Altered macrophage differentiation and immune dysfunction in tumor development. J. Clin. Invest. 117,1155–1166 (2007). Received: 24 August 2017 Accepted: 9 May 2018 27. Naito, Y. et al. CD8+T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res. 58, 3491–3494 (1998). 28. Sato, E. et al. Intraepithelial CD8+tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc. Natl. Acad. Sci. USA 102, 18538–18543 (2005). 29. Galon, J. et al. Type, density, and location of immune cells within human References colorectal tumors predict clinical outcome. Science 313,1960–1964 1. Munn, D. H. & Bronte, V. Immune suppressive mechanisms in the tumor (2006). microenvironment. Curr. Opin. Immunol. 39,1–6 (2016). 30. Radoja, S. & Frey, A. B. Cancer-induced defective cytotoxic T lymphocyte 2. Gonzalez-Gugel, E., Saxena, M. & Bhardwaj, N. Modulation of innate effector function: another mechanism how antigenic tumors escape immune- immunity in the tumor microenvironment. Cancer Immunol. Immunother. 65, mediated killing. Mol. Med. 6, 465–479 (2000). 1261–1268 (2016). 31. Ries, C. H. et al. Targeting tumor-associated macrophages with anti-CSF-1R 3. Gajewski, T. F. et al. Immune resistance orchestrated by the tumor antibody reveals a strategy for cancer therapy. Cancer Cell. 25, 846–859 microenvironment. Immunol. Rev. 213, 131–145 (2006). (2014). 4. Chanmee, T., Ontong, P., Konno, K. & Itano, N. Tumor-associated 32. Shin, A. S. & Camilleri, M. Diagnostic assessment of diabetic gastroparesis. macrophages as major players in the tumor microenvironment. Cancers Diabetes 62, 2667–2673 (2013). (Basel) 6, 1670–1690 (2014). 33. Liu, X. et al. Regulatory T cells trigger effector T cell DNA damage and 5. Bronte, V. & Murray, P. J. Understanding local macrophage phenotypes in senescence caused by metabolic competition. Nat. Commun. 9, 249 (2018). disease: modulating macrophage function to treat cancer. Nat. Med. 21, 34. Pearson, T., Greiner, D. L. & Shultz, L. D. Creation of “humanized” mice to 117–119 (2015). study human immunity. Curr. Protoc. Immunol.15,15–21 (2008). NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications 9 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04567-0 35. Maykel, J. et al. NOD-scidIl2rg (tm1Wjl) and NOD-Rag1 (null) Il2rg Author contributions (tm1Wjl): a model for stromal cell-tumor cell interaction for human colon Y.L., H.G., and Z.Z. designed the research, analyzed the data, prepared figures. Y.L. and cancer. Dig. Dis. Sci. 59, 1169–1179 (2014). H.G. performed the experiments. R.B. and Z.Z. provided tumor samples and clinical 36. Kaneda, M. M. et al. PI3Kgamma is a molecular switch that controls immune information. Z.Z. wrote the manuscript. suppression. Nature 539, 437–442 (2016). 37. Sonda, N. et al. miR-142-3p prevents macrophage differentiation during Additional information cancer-induced myelopoiesis. Immunity 38, 1236–1249 (2013). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 38. Wu, X. et al. VentX trans-activates p53 and p16ink4a to regulate cellular 018-04567-0. senescence. J. Biol. Chem. 286, 12693–12701 (2011). 39. Rawat, V. P. et al. The vent-like homeobox gene VENTX promotes human Competing interests: The authors declare no competing interest. myeloid differentiation and is highly expressed in acute myeloid leukemia. Proc. Natl. Acad. Sci. USA 107, 16946–16951 (2010). Reprints and permission information is available online at http://npg.nature.com/ 40. Gao, H., Wu, B., Le, Y. & Zhu, Z. Homeobox protein VentX induces p53- reprintsandpermissions/ independent apoptosis in cancer cells. Oncotarget 7, 39719–39729 (2016). 41. Schmidt, A. et al. Human macrophages induce CD4(+)Foxp3(+) regulatory Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in T cells via binding and re-release of TGF-beta. Immunol. Cell Biol. 94, published maps and institutional affiliations. 747–762 (2016). 42. Kamada, N. et al. Unique CD14 intestinal macrophages contribute to the pathogenesis of Crohn disease via IL-23/IFN-gamma axis. J. Clin. Invest. 118, 2269–2280 (2008). Open Access This article is licensed under a Creative Commons 43. Rogler, G. et al. Isolation and phenotypic characterization of colonic Attribution 4.0 International License, which permits use, sharing, macrophages. Clin. Exp. Immunol. 112, 205–215 (1998). adaptation, distribution and reproduction in any medium or format, as long as you give 44. Tiemessen, M. M. et al. CD4+CD25+Foxp3+regulatory T cells induce appropriate credit to the original author(s) and the source, provide a link to the Creative alternative activation of human monocytes/macrophages. Proc. Natl. Acad. Commons license, and indicate if changes were made. The images or other third party Sci. USA 104, 19446–19451 (2007). material in this article are included in the article’s Creative Commons license, unless 45. Guo, F., Iclozan, C., Suh, W. K., Anasetti, C. & Yu, X. Z. CD28 controls indicated otherwise in a credit line to the material. If material is not included in the differentiation of regulatory T cells from naive CD4 T cells. J. Immunol. 181, article’s Creative Commons license and your intended use is not permitted by statutory 2285–2291 (2008). regulation or exceeds the permitted use, you will need to obtain permission directly from 46. Mittal, V. K., Bhullar, J. S. & Jayant, K. Animal models of human colorectal the copyright holder. To view a copy of this license, visit http://creativecommons.org/ cancer: Current status, uses and limitations. World J. Gastroenterol. 21, licenses/by/4.0/. 11854–11861 (2015). 47. Wong, H. H. & Chu, P. Immunohistochemical features of the gastrointestinal tract tumors. J. Gastrointest. Oncol. 3, 262–284 (2012). © The Author(s) 2018 Acknowledgements We thank Marc Kirschner for discussions and critical reading of the manuscript. Z.Z. was supported by grants from the National Institutes of Health, Department of Defense, American Cancer Society, and research funds from Brigham and Women’s Hospital. 10 NATURE COMMUNICATIONS (2018) 9:2175 DOI: 10.1038/s41467-018-04567-0 www.nature.com/naturecommunications | | |

Journal

Nature CommunicationsSpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off