Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Sep 1.
Published in final edited form as: J Immunol. 2017 Jul 28;199(5):1933–1941. doi: 10.4049/jimmunol.1700529

Versican-derived matrikines regulate Batf3-dendritic cell differentiation and promote T-cell infiltration in colorectal cancer

Chelsea Hope 1,2,*, Philip B Emmerich 1,2,*, Athanasios Papadas 1,2, Adam Pagenkopf 1,2, Kristina A Matkowskyj 2,3,5, Dana R Van De Hey 1, Susan N Payne 2, Linda Clipson 4, Natalie S Callander 1,2,5, Peiman Hematti 1,2, Shigeki Miyamoto 4, Michael G Johnson 1,2, Dustin A Deming 1,2,4,5,+, Fotis Asimakopoulos 1,2,+
PMCID: PMC5568487  NIHMSID: NIHMS891605  PMID: 28754680

Abstract

Colorectal cancer (CRC) originates within immunologically complex microenvironments. To date the benefits of immunotherapy have been modest except in neoantigen-laden mismatch repair (MMR)-deficient tumors. Approaches to enhance tumor-infiltrating lymphocytes in the tumor bed may substantially augment clinical immunotherapy responses. Here we report that proteolysis of the tolerogenic matrix proteoglycan versican (VCAN) strongly correlated with CD8+ T-cell infiltration in CRC, regardless of mismatch-repair status. Tumors displaying active VCAN proteolysis and low total VCAN were associated with robust (10-fold) CD8+ T-cell infiltration. Tumor-intrinsic WNT pathway activation was associated with CD8+ T-cell exclusion and VCAN accumulation. In addition to regulating VCAN levels at the tumor site, VCAN proteolysis results in the generation of bioactive fragments with novel functions (VCAN-derived matrikines). The VCAN-derived matrikine, versikine, enhanced the generation of CD103+CD11chiMHCIIhi conventional dendritic cells (cDC) (homologous to Batf3-DC/CD8α+ DC/CD103+ DC/cDC1 subset/intratumoral DC2) from Flt3L-mobilized primary bone marrow-derived progenitors, suggesting that VCAN proteolysis may promote differentiation of tumor-seeding DC precursors towards IRF8- and BATF3-expressing cDC. Intratumoral BATF3-dependent DC are critical determinants for T-cell anti-tumor immunity, effector T cell trafficking to the tumor site and response to immunotherapies. Our findings provide a rationale for testing VCAN proteolysis as a predictive and/or prognostic immune biomarker and VCAN-derived matrikines as novel immunotherapy agents.

Keywords: VCAN, versikine, colorectal cancer, immunotherapy, tumor-infiltrating lymphocytes, tumor-associated macrophages

INTRODUCTION

CRC is the second leading cause of cancer-related mortality in the United States (1). The 5-year survival rate for patients with metastatic disease is unacceptably low (12%), generating an impetus for rapid progress to improve outcomes. Recent advances in cancer immunotherapy have only marginally impacted outcomes in CRC (2, 3). The noteworthy exception includes patients with mismatch repair-deficient (dMMR) tumors where genetic instability generates an expanded neo-antigenic repertoire (4). In dMMR cancers, treatment with the anti-PD1 antibodies pembrolizumab and nivolumab results in deep and prolonged therapeutic responses for a large proportion of patients (2, 4, 5). Unfortunately, not all patients with dMMR CRCs respond to these agents indicating that other regulatory factors play a key role in the response of CRCs to checkpoint blockade. In addition, an effective means to utilize immuno-oncology agents for mismatch repair proficient (pMMR) CRCs, which encompass greater than 95% of all metastatic CRCs, has yet to be identified.

The presence of infiltrating lymphocytes (TILs) is linked to favorable clinical outcomes and increased response rates to immune checkpoint inhibition (5, 6). Thus, TIL infiltration possesses both prognostic and predictive biomarker utility. However, at a mechanistic level, the tumor-cell autonomous and non-autonomous networks controlling immune infiltration into the tumor bed are mostly unknown. Approaches to enhance TIL entry/activation could have a major impact on immunotherapy efficacy.

We recently demonstrated that versican (VCAN), a large matrix proteoglycan with immunoregulatory activity, accumulates in the extracellular matrix of multiple myeloma tumors (7). VCAN contributes to cancerous and non-cancerous inflammation by promoting leukocyte-derived elaboration of inflammatory mediators (8–13) but also immunodeficiency through dendritic cell (DC) dysfunction (14). Interestingly, we also detected in situ VCAN proteolysis in a pattern consistent with the activities of a-disintegrin-and-metalloproteinase-with-thrombospondin-motifs (ADAMTS) proteases (15). We hypothesized that VCAN proteolysis serves to generate bioactive fragments (“matrikines”)(16–18). Indeed, we demonstrated a fragment containing VCAN’s N-terminal 441 amino acids, “versikine” (19), elicits a transcriptional program that is predicted to promote immunogenicity, and thus, antagonize the tolerogenic actions of its parent, intact VCAN (15). However, it is unclear whether VCAN-dependent immunoregulatory mechanisms are operative in non-myeloma, or indeed non-hematopoietic, settings. We chose to investigate CRC because both myeloma and CRC are driven by chronic inflammatory networks (20) and because better understanding of CRC immunosurveillance mechanisms will likely result in improved outcomes for large patient populations. Here we demonstrate that VCAN proteolysis correlates with CD8+ T-cell infiltration in CRC, regardless of mismatch-repair status. Mechanistically, we propose that the VCAN-derived matrikine, versikine, promotes T-cell infiltration through regulation of Batf3-dependent dendritic cells (DC), a rare DC subset (21, 22) that is crucial for effector T cell trafficking (23), T cell-mediated antitumor immunity (24, 25) and response to diverse immunotherapy modalities (26–28). These results provide strong rationale for investigation of VCAN processing in immunotherapy prognostication and therapy across several solid and liquid tumor types.

MATERIALS AND METHODS

Colorectal cancer (CRC) tissue microarray (TMA)

A CRC TMA was created through the University of Wisconsin Carbone Cancer Center Translational Science Biocore Biobank. This TMA contains samples from 122 subjects with colorectal cancer across all stages. For each subject, the TMA contains 2 cores from the primary tumor and 1 core of tumor-associated normal tissue. The tumors utilized in the TMA were selected for their location and stage, such that an equal distribution of right, left and rectal tumors and stage I through IV cancers were present.

Immunohistochemical (IHC) methods and antibodies

Unstained 4–5 μm-thick TMA sections were deparaffinized and rehydrated using standard methods. Antigen retrieval was carried out in EDTA buffer (CD8 detection) or citrate (all others). The slides were treated with chondroitinase ABC prior to staining with the total VCAN antibody (29). Primary antibodies included total VCAN (HPA004726, Sigma, St. Louis, MO), αDPEAAE (PA1-1748A, Thermo Fisher, Waltham, MA), CD8 (c4-0085-80, Ebioscience, San Diego, CA, USA), phosphorylated ERK 1/2 (Thr202/Tyr204, 4370, Cell Signaling Technology, Danvers, MA), phosphorylated ribosomal protein S6 (RPS6) (Ser235/236, 4858, Cell Signaling Technology), and CTNNB1 (β-catenin, 8480, Cell Signaling Technology). The αDPEAAE neoepitope antibody has been previously validated (29).

Scoring and analysis of staining patterns

Cytoplasmic and membrane staining of the epithelium and stroma was scored for each core sample by at least three observers including a pathologist (K.A.M.) blinded to clinical parameters. Stained slides were examined using an Olympus BX43 microscope with attached Olympus DP73 digital camera (Olympus Corp, Waltham, MA). Epithelium and stroma were evaluated separately for total VCAN and αDPEAAE staining. Immunostaining for VCAN, αDPEAAE, phosphorylated ERK1/2, and phosphorylated RPS6 was assessed by scoring staining intensity (0 for no staining, 1 for low/weak staining, 2 for moderate staining and 3 for strong/intense staining) and the percentage of cells staining positive (0 for no staining, 1 for >0–10%, 2 for 11–50%, 3 for 51–75% and 4 for >75% staining). For CD8+ detection, the number of tumor infiltrating lymphocytes (TILs) per high-power field (HPF) within the malignant epithelium was calculated using a single area at 400X magnification (ocular 10× with an objective of 40x). To validate that these TMA cores were representative of the tumor a subset of cancers were selected and whole tissue sections were stained for CD8 and the CD8 scoring was compared to the TMA cores (Supplemental Table S1). Nuclear localization of β-catenin was recorded as present or absent. Tissue cores that were missing, damaged, contained staining artifacts, or had uncertain histology were excluded from the analysis.

Mismatch repair (MMR) analyses

MMR status was determined by IHC for MLH1, MSH6, MSH2, and PMS2. The following prediluted primary antibodies were utilized: MLH1 ((M1) mouse monoclonal, Ventana Medical Systems, Inc, Tucson, AZ); MSH6 ((44) mouse monoclonal, Ventana Medical Systems, Inc); MSH2 ((G219-1129) mouse monoclonal, Ventana Medical Systems, Inc); and PMS2 ((EPR3947) rabbit monoclonal, Ventana Medical Systems, Inc). Staining was performed on a BenchMark ULTRA automated slide staining system and detected using the Opitview DAB IHC detection kit. Absence of staining for these proteins was scored by independent pathology review (K.A.M.). Tumor infiltrating leukocytes were utilized as an internal control.

Ki67 proliferation index

Immunofluorescence was performed by placing the TMA slides into a humidity chamber after slides were deparaffinized and rehydrated. Slides were blocked with 5% bovine serum albumin in Tris Buffered Saline (TBS) with 0.05% Tween 20 for one hour at room temperature. Slides were then washed in TBS. The KI67 primary antibody (#11882 (Alexa Fluor 488 conjugate), Cell Signaling Technology) was diluted in PBS and incubated overnight at 4°C overnight. After incubation, coverslips were washed in TBS and mounted using Prolong Gold DAPI mounting media (#P36931, Invitrogen, Carlsbad, CA) and sealed. TMA cores were classified based on the number of KI67 positive nuclei per core.

Generation and validation of recombinant versikine

Recombinant versikine was purified from mammalian cells as previously described (15, 29) with the addition of a size-exclusion chromatography step at the end of the purification process. Versikine preparations were routinely endotoxin-tested by the LAL Chromogenic Endotoxin Assay Kit (GenScript, Piscataway, NJ) and rejected if endotoxin contamination was detected at equal or >0.25 EU/mL. Fluorophore-assisted carbohydrate electrophoresis (FACE) was used to exclude hyaluronan contamination. Circular dichroism analysis was used to confirm proper folding and secondary structure of recombinant versikine.

Bone marrow harvesting, Flt3L-mobilized cultures and flow cytometry

Bone marrow (BM) cells were harvested from C57BL/6J mice under IACUC-approved protocol M005476. Tissue from Irf8-EGFP mice was provided by Dr. Scott Abrams (Roswell Park Cancer Institute, Buffalo, NY). Total BM cells were cultured for 9 days in the presence of 200ng/mL Flt3L, as previously described (30) with the addition of 1μM recombinant versikine or vehicle at the beginning of culture. Harvested cells were resuspended in FACS buffer (PBS pH7.4, 2mM EDTA, 0.5% BSA). Cell viability was established by Trypan Blue exclusion and 2×106 live cells were stained the following antibodies: anti-CD11c (N418-PE-Cy7, Tonbo); anti-CD103 (2E7-PE, Biolegend); anti-MHCII (M5/114.152-AlexaFluor 700, Biolegend), anti-SiglecH (551-PerCP-Cy5.5, Biolegend) and anti-CD11b (P84-FITC, Biolegend) for 30 minutes on 4°C. Cells were washed, analyzed on a BD LSR II instrument and viability was assessed by DAPI staining. The instrument was calibrated daily according to manufacturer’s protocol using the BD FACSDiva (v.6) Cytometer Setting & Tracking software application. Flow cytometry data was analyzed by FlowJo version 9.7.6 software (Tree Star, Ashland, OR).

Immunoblot analysis

Whole cell lysates were prepared by boiling cells in Laemmli Sample buffer (Bio-Rad) supplemented with 100mM DTT for 10 minutes at a final concentration of 107 cells/ml. 105 cells or 20μg protein was resolved by SDS-PAGE and transferred to Immobilon-P PVDF membrane (Millipore). Membranes were blocked in 5% Milk in TBS-T [25mM Tris-HCl (pH 7.4), 0.13M NaCl, 2.7mM KCl]. Primary antibodies [anti-IRF8 (Cell Signaling Technologies, D20D8), anti-Batf3 (LSBio, B12B125)] were diluted in 5% Milk-TBST and membranes were incubated overnight at 4°C. Secondary antibody-HRP-conjugates as well as anti-GAPDH-HRP conjugate (Genscript A00192) incubations were carried out for 1 hour at room temperature. Signal detection was achieved using Amersham ECL Plus chemiluminescent solution (GE Healthcare). Blots were developed on Classic Blue Autoradiography Flim BX (MidSci).

Statistical analyses

Descriptive statistics were utilized to present the data including mean ± standard deviation. Wilcoxon rank sum and chi-square analyses were utilized where noted. A p value of ≤ 0.05 is considered statistically significant.

RESULTS

VCAN accumulation and proteolysis in normal and malignant colorectal tissue

The University of Wisconsin CRC TMA consists of 122 cases with matched cores from colorectal cancer and tumor-associated normal colon tissues. We stained the TMA with antibodies raised against a neoepitope (αDPEAAE) generated through VCAN cleavage at the Glu441-Ala442 bond of the V1-VCAN isoform (19). DPEAAE constitutes the C-terminal end of the bioactive VCAN fragment, versikine. Serial tissue TMA sections were stained with an antibody recognizing the immunoglobulin-like domain at VCAN’s N-terminal end. The latter would be expected to recognize all intact VCAN isoforms (total VCAN). Although its immunogen sequences are also included within cleaved VCAN, detection of cleaved/mobilized VCAN appears less sensitive with the latter antibody. Intense total VCAN staining was observed in tumor stroma (Fig. 1A and 1B). By contrast, highest intensity staining for the αDPEAAE neoepitope (2+, 3+) was detected within normal stroma and only variably within tumor stroma (Chi-square test, p<0.001; Fig. 1A and 1C; Supplemental Fig. S1A).

Figure 1. VCAN accumulation and processing in colorectal cancer.

Figure 1

Open in a new tab

A tissue microarray containing matched cores from colorectal cancers and the tumor-associated normal colon was stained for total VCAN and αDPEAAE, a neoepitope generated from VCAN cleavage at Glu441-Ala442 (V1-enumeration) (A). VCAN staining was observed variably within the stroma of CRCs, however overall an increase in the intensity of VCAN staining was observed in the tumor tissues compared to the normal colon (Chi-square test, p<0.001, A and B). VCAN proteolysis, as determined by αDPEAAE staining, was extensive in the stroma of normal tissue and markedly reduced in numerous CRCs (Chi-square test, p<0.001, A and C). Scale bar in A = 100μm.

There was no correlation between total VCAN staining and location of primary tumor (left/right colon, rectum) (Supplemental Fig. S1B). Increased αDPEAAE staining was observed in the rectum compared to the colon (Chi-square test, p=0.009). To determine whether VCAN processing correlated with tumor location, tumors were classified according to the degree of VCAN accumulation and processing in their stroma. Tumors were classified as “VCAN proteolysis-predominant” if their staining for total VCAN staining intensity was ≤1+ and staining for VCAN proteolysis (αDPEAAE antibody) was ≥2. Conversely, tumors were classified as “proteolysis-weak” if intact VCAN staining intensity was >1+ or αDPEAAE intensity was <2+. Despite a greater staining for αDPEAAE neoepitope being identified within the rectum, there was no significant correlation between the VCAN proteolysis-predominant classification and tumor location (Chi-square test, p=0.96; Supplementary Fig. S1C).

“VCAN proteolysis-predominant” tumors show robust CD8+ T-cell infiltration

Given the immunosuppressive properties of VCAN and immunostimulatory properties of its proteolytic product, versikine (15), we hypothesized that VCAN proteolysis-predominant tumors are primed for immune infiltration. To determine whether VCAN processing correlated with CD8+ T-cell infiltration, the TMA was stained for the effector T-cell marker, CD8, and correlated with the VCAN proteolysis classification. We detected a statistically significant correlation between proteolysis-predominant status and CD8+ T-cell infiltration. CD8+ scores in “proteolysis-predominant” tumors were on average 10-fold higher than “proteolysis-weak” tumors (mean of 22 CD8+ T-cells per HPF versus 2, respectively; Wilcoxon rank sum test, p<0.001; Fig. 2A–B, Supplemental Fig. S2). Across tumors with robust CD8+ T-cell infiltration into the malignant epithelium (defined by ≥15 CD8+ TILs per HPF), the ratio of stromal to epithelial CD8+ T-cells was assessed demonstrating that 77% of the CD8+ T-cells were in the epithelial compartment of these tumors.

Figure 2. Robust CD8+ T-cell infiltration in “VCAN proteolysis-predominant” tumors.

Figure 2

Open in a new tab

Colorectal cancers were classified as “VCAN proteolysis-predominant” if their staining for total VCAN was weak (≤1+) and staining for VCAN proteolysis was strong (αDPEAAE intensity ≥2+). Tumors that did not meet those criteria were classified as “proteolysis-weak” (A). Given the immunoregulatory properties of VCAN and the immunostimulatory properties of its proteolytic product, versikine, CD8+ T-cell infiltration was assessed comparing VCAN proteolysis-predominant cancers versus proteolysis-weak cancers. Proteolysis-predominant tumors display 10-fold higher CD8 scores on average than proteolysis-weak tumors (Wilcoxon rank sum test, p<0.001; B). CD8+ T-cell infiltration is greatest in cancers with intensive VCAN proteolysis and low total VCAN (Wilcoxon rank sum test, p<0.001, C). Scale bar in A = 100μm.

CD8+ T-cell infiltration was highest in tumors that displayed intense VCAN proteolysis together with low amounts of total VCAN (Fig. 2C). This finding suggests that low VCAN accumulation may not adequately promote T-cell infiltration unless VCAN is actively processed to generate proteolytic fragments. This observation is consistent with our hypothesis that VCAN proteolysis generates bioactive fragments with novel activities. Conversely, in tumors with high total VCAN, CD8+ T-cell infiltration may be impeded through an unfavorable stoichiometry between intact VCAN and VCAN fragments. In summary, these data suggest that VCAN proteolytic fragments are not mere markers of VCAN turnover but are endowed with important novel immunomodulatory activities. We have previously elucidated the immunoregulatory role of the VCAN fragment, versikine (15). Since tumors with greater degrees of CD8+ T-cell infiltration are known to result in a better prognosis, the association between VCAN proteolysis and tumor stage was assessed. A trend toward an increased prevalence of staining for the VCAN proteolysis-predominant classification was seen in colon cancers of earlier stage, albeit not statistically significant (Chi-square test, p=0.28; Supplemental Fig. S1D).

CD8+ T-cell infiltration correlates with VCAN proteolysis regardless of MMR status

dMMR is observed in 15% of localized CRCs and 3–4% of metastatic cases (2, 4, 5). MLH1 and MSH2 are the most commonly lost MMR proteins. These proteins can be lost secondary to somatic or germline mutations or epigenetic silencing. dMMR status has been associated with an improved prognosis and increased response to immune checkpoint blockade (2, 4, 5). Since dMMR is one of the strongest predictors of CD8+ T-cell infiltration, we next examined the potential for a correlation between VCAN proteolysis and MMR status. IHC staining for the MMR proteins MLH1, MSH2, PMS2 and MSH6 was performed to determine MMR status. Consistent with prior reports, CD8+ T-cell infiltration was increased in dMMR tumors (Wilcoxon rank sum test, p<0.001; Fig. 3A). MMR status was then correlated with VCAN and αDPEAAE staining. We observed all potential staining combinations in both pMMR and dMMR cancers (Fig. 3B). A trend towards increased intensity of VCAN staining in pMMR cancers was observed. No significant differences were observed in the proportions of tumors staining for VCAN and αDPEAAE across dMMR cancers (Fig. 3B). The correlation between VCAN proteolysis and CD8+ T-cell infiltration was maintained in both pMMR and dMMR (Fig. 3C). In both pMMR and dMMR, those tumors staining for the VCAN proteolysis-predominant classification had the greatest degree of CD8+ T-cell infiltration (Wilcoxon rank sum tests: pMMR p=0.006; dMMR p=0.03). Among the VCAN proteolysis-predominant tumors there was a greater degree of CD8+ T cell infiltration in the dMMR cancers compared to pMMR cancers (35 versus 14.8 TILs per HPF, Wilcoxon rank sum test, p=0.04).

Figure 3. Impact of VCAN proteolysis on CD8+ T-cell infiltration in MMR proficient and deficient cancers.

Figure 3

Open in a new tab

Identification of cases within the TMA with MMR deficiency was performed by IHC analysis for MLH1, MSH2, PMS2 and MSH6. Loss of staining for any of these proteins confirmed MMR deficiency. Non-tumor cells were utilized as an internal control. Increased CD8+ T-cell infiltration in dMMR cancers was confirmed in the TMA CRC cores with a mean of 11.7 CD8+ T-cells per HPF in dMMR tumors compared to 3.1 per HPF in pMMR (Wilcoxon rank sum test, p<0.001; A). The intensity of staining for both VCAN and αDPEAAE varied across both dMMR and pMMR cancers with a trend toward more intense VCAN stromal staining in pMMR cancers (B). In both pMMR and dMMR cancers, the VCAN proteolysis predominant cancers had the greatest infiltration of CD8+ T-cells (Wilcoxon rank sum test, dMMR p=0.031, pMMR p=0.006; C). Comparing the VCAN proteolysis-predominant tumors, the dMMR cancers had increased CD8+ T-cell infiltration compared to the pMMR cancers (Wilcoxon rank sum test, p=0.04; C). The proportion of VCAN proteolysis predominant tumors varies depending on the MMR status with this being more common in dMMR tumors (Wilcoxon rank sum test, p=0.01; D).

The VCAN proteolysis predominant phenotype is more common in dMMR cancers

Since the VCAN proteolysis predominant phenotype predicts CD8+ T-cell infiltration in both dMMR and pMMR cancers the prevalence of this phenotype was examined. Of the dMMR tumor samples, 25% possessed the VCAN proteolysis predominant phenotype, while this was observed in only 10% of pMMR samples (Fig. 3D, Wilcoxon rank sum test, p=0.01). In addition, another 25% of dMMR cancers demonstrated 1+ or less staining for both total VCAN and αDPEAAE, while this was observed in an additional 14% of pMMR cancers.

CD8+ T-cell exclusion is associated with WNT pathway activation in tumor cells

In a recent report by the Gajewski group (28), WNT signaling activation in melanoma tumor cells correlated with CD8 T-cell exclusion. Because activation of WNT signaling is a frequent molecular event in CRC secondary to the presence of truncating mutations in APC or activating mutations in CTNNB1 (31), we investigated whether analogous mechanisms operated in CRC. Indeed, we detected a statistically significant negative correlation between nuclear CTNNB1 (β-catenin, a marker of active WNT signaling) and CD8+ T-cell infiltration in CRC (Wilcoxon rank sum test, p=0.014; Fig. 4A). In addition, VCAN accumulation correlated with the presence of nuclear β-catenin (Chi-square test, p<0.001, Fig. 4B) and was more common in the pMMR cancers compared to dMMR tumors (53 vs. 8%, respectively, Chi-square test, p<0.001, Fig. 4C).

Figure 4. CD8+ T-cell exclusion in tumors with active WNT signaling.

Figure 4

Open in a new tab

Truncating mutations in APC are commonly encountered in CRC and activation of WNT signaling has demonstrated immunoregulatory properties (28). To examine the impact of activation of WNT signaling, IHC staining for β-catenin was performed and the presence of nuclear localization of β-catenin was assessed. Those tumors with nuclear β-catenin had a significant reduction in CD8+ T-cell infiltration (Wilcoxon rank sum test, p=0.01; A). In addition, those tumors with nuclear localization of β-catenin had a higher rate of intense staining for VCAN (Chi-square test, p<0.001; B). Nuclear β-catenin was more common in the pMMR cancers compared to dMMR tumors (53 vs. 8%, Chi-square test, p<0.001, G).

VCAN accumulation and/or proteolysis is not associated with tumor-intrinsic activation of the MAPK and PI3K pathways, nor with Ki67 index in CRC

We investigated a potential correlation between MAPK pathway activation in tumor cells (detected through ERK1/2 phosphorylation), PI3K pathway activation (detected through RPS6 phosphorylation) or tumor cell proliferation (as measured through KI67 staining). The results are shown in Supplemental Fig. S3A–E. There was no correlation between activation of these key oncogenic pathways and/or Ki67 index with VCAN processing.

Versikine promotes the generation of CD103+ cDC (Batf3-DC) from Flt3L-mobilized primary bone marrow cultures

VCAN proteolysis may impact on tumor immune contexture through regulation of intact VCAN bioavailability and/or the generation of novel bioactive fragments (matrikines). We have previously shown that versikine, a matrikine generated through VCAN proteolysis at the Glu440-Ala441 bond, activates an IRF8-dependent transcriptional program in macrophage-like cells (15). IRF8 is a terminal selector for Batf3-DC [variably referred to as cDC1 subset/CD8α+ cDC/CD103+ cDC/intratumoral DC2) (32), a DC subset with crucial roles in T-cell-mediated immunosurveillance (21–28).

Flt3L-mobilized BM cultures have long provided a faithful ex vivo model of DC differentiation (30). Addition of recombinant versikine at the onset of culture (together with Flt3L) consistently and reproducibly promoted expansion of the CD103+CD11chiMHCIIhi DC at both early and late culture timepoints (Fig. 5A, 5B). These cells were SIRPαlo, CD11blo-int and SiglecHlo confirming their identity as CD103+ conventional DC (cDC). There was no difference in the prevalence of CD11cintSiglecHhi plasmacytoid dendritic cells (pDC) at Day #4, but CD11cintSiglecHhi pDC were reduced in the presence of versikine by Day #9 (Fig. 5C). Addition of the TLR2/6 ligand FSL-1 (Pam2CGDPKHPKSF) together with Flt3L, at the onset of culture, conferred a disadvantage to CD103+ DC development (Fig. 5D). Because intact VCAN is thought to act through TLR2/6 heterodimers (13), these results suggest that versikine may signal through pathways other than those triggered by intact VCAN. They also suggest that intact VCAN may exert tolerogenic actions by preventing Batf3-DC differentiation. Taken together, our data suggest that tumor-seeding, bone-marrow-derived DC precursors may preferentially develop into immunogenic Batf3-DC in tumor microenvironments undergoing active VCAN proteolysis.

Fig. 5. Versikine, a VCAN-derived matrikine, promotes CD103+CD11chiMHCIIhi DC generation from Flt3L-mobilized bone marrow progenitors.

Fig. 5

Open in a new tab

Bone marrow (BM) from C57BL/6J animals was isolated and cultured in the presence of 200 ng/mL Flt3L for 9 days, as previously described (30). At conclusion of culture, a mixture of DC precursors and mature DC is obtained in this well-characterized system. Addition of versikine (1μM) at Day #0, alongside Flt3L, resulted in reproducible expansion of CD103+CD11chiMHCIIhi DC (results shown are representative of at least 6 independent experiments). Although the total number of CD11c+ cells was similar between vehicle- and versikine-supplemented cultures, there was a consistent skewing towards CD103+ differentiation, measurable at both earlier culture timepoints (4 days, A) and later culture timepoints (9 days, B). CD103+MHCIIhi cells were SIRPαlo, CD11blo-int and SiglecHlo confirming their identity as CD103+ conventional DC (cDC). Versikine resulted in disadvantage to plasmacytoid DC (pDC) development (CD11cintSiglecHhi) at Day #9 (C). Intact VCAN acts through TLR2/6 heterodimers. Addition of the TLR2/6 ligand, FSL-1, to Flt3L- supplemented cultures results in CD103+MHCIIhi differentiation impediment, suggesting that versikine and intact VCAN may exert opposing actions on DC differentiation (D). Veh= vehicle; Vkine= versikine; flow= flow cytometry.

Versikine promotes the expansion of IRF8- and BATF3-expressing DC

Versikine-treated, Flt3L-mobilized cultures displayed increased expression of Irf8 and Batf3 at Day #9 (Fig. 6A). To more precisely map versikine’s effects on DC development, we added versikine both at culture initiation (Day #0) together with Flt3L or at Day #0 and Day #6. The latter timepoint was chosen because pre-DC in Flt3L-mobilized cultures have been shown to peak at Day #6 (33). Versikine added at Day #0 likely acts on pre-DC already present within the explanted bone marrow. Addition of versikine at both timepoints further increased Batf3-DC generation compared to single administration at Day #0. These data provide indirect support to the hypothesis that versikine acts on pre-DC to favor Batf3-DC differentiation. Because Batf3-DC are extremely sparse (21) but critical for several aspects of T-cell-mediated immunity, even modest expansions at the tumor site may have profound consequences for T-cell anti-tumor immunity (24). We wanted to know whether the increase in total Irf8 expression reflected an expansion of Irf8-expressing Batf3-DC. To this end, we used bone marrow cells derived from Irf8-EGFP reporter mice. Indeed, versikine-induced CD103+ cDC uniformly expressed the cDC1 (Batf3-DC) terminal selector, Irf8 (Fig. 6C). Moreover, these results suggest that versikine does not induce CD103 expression in an unrelated cell type but results in expansion of bona fide cDC1 subset DCs (Batf3-DC).

Figure 6. Versikine promotes the expansion of Irf8- and Batf3- expressing DC from Flt3L-mobized bone marrow progenitors.

Figure 6

Open in a new tab

Versikine-supplemented Flt3L-mobilized BM cultures demonstrate increased expression of the CD103+ DC terminal selector, Irf8, as well as transcription factor Batf3 at Day #9 (A). To mimic DC differentiation under conditions of extended versikine exposure, we added versikine to Flt3L-supplemented bone marrow cultures at Day #0 or both Day #0 and Day #6. The latter timepoint was chosen because pre-DC peak at Day #6 in Flt3L-driven DC cultures (33). Addition of versikine at both timepoints further boosted CD103hiMHCIIhi DC. The results provide indirect support to our hypothesis that versikine acts on pre-DC precursors to promote Batf3-DC development (B). * p<0.05; ****p<0.0001. Using BM cells from Irf8-EGFP mice, we demonstrate that versikine-induced, Flt3L- mobilized CD103+ cDC are bona fide expressors of the cDC1 (Batf3-DC) terminal selector, Irf8 (C). Note that Irf8-EGFP mice generated more Batf3-DC than standard C57BL/6J at equivalent Flt3L-supplemented culture timepoints (compare to Figure 5B). However, versikine consistently resulted in Batf3-DC DC expansion compared to vehicle control. Enhanced baseline DC differentiation in Irf8-EGFP-derived cultures may be related to genetic differences between standard C57BL/6J and Irf8-EGFP strains or to the fact that bones from the latter animals were transported overnight in PBS/1%FCS prior to bone marrow extraction in our laboratory on the next day, whereas C57BL/6J experiments were performed using freshly explanted tissue. Veh= vehicle; Vkine= versikine; flow= flow cytometry.

DISCUSSION

Colorectal cancer remains a challenging problem with significant impact for the general population. Recent advances in immunotherapy of solid tumors previously thought to be non-immunogenic, such as lung cancer, raised hopes that CRC patients might also benefit. However, CRC responses to novel immunotherapy modalities have been modest at best, with the exception of a small number of patients with mismatch repair-deficient CRC. Future challenges include the selection of patients most likely to respond (through the identification and validation of novel predictive biomarkers), as well as, the devising and testing of innovative combinatorial immunotherapy regimens that augment efficacy with acceptable toxicity. CD8+ T-cell infiltration has been associated with an improved prognosis and response to immune checkpoint blockade, especially in the setting of dMMR. However, the mechanisms regulating immune cell infiltration are largely yet to be determined.

We report here the strong association between VCAN proteolysis and CD8+ T-cell infiltration. At a mechanistic level, proteolysis of intact VCAN can be postulated to produce three alternative consequences, not mutually exclusive: Firstly, proteolysis may regulate the amount and bioavailability of tolerogenic intact VCAN at the tumor site and the resultant degree of DC dysfunction (14). Secondly, proteolysis may disrupt VCAN’s complex interactions with other immunoregulatory matrix components, such as hyaluronan or tenascin C (34). Thirdly, VCAN proteolysis generates fragments with novel activities. We recently showed that versikine, a bioactive fragment generated through VCAN proteolysis, elicits an IRF8-dependent type-I interferon transcriptional program as well as IL12 but not IL10 production from macrophage-like cells (15). These actions are predicted to enhance immunogenicity and tumor “sensing” by the immune system. Indeed, in a small myeloma panel, VCAN proteolysis was necessary, albeit not sufficient, for CD8+ T-cell infiltration (15). In this manuscript, we demonstrate that versikine promotes generation of CD103+CD11chiMHCIIhi conventional DC (Batf3-DC) from Flt3L-mobilized BM progenitors. The data support a model in which DC precursors seeding tumor sites undergoing active VCAN proteolysis may preferentially differentiate towards Batf3-DC implicated in T-cell mediated immunosurveillance and response to immunotherapies (22–28) (Fig. 7). VCAN-regulated T-cell influx mechanisms may operate additively with increased neo-antigen density in dMMR CRC (4).

Figure 7. VCAN-derived matrikines may promote “T-cell inflammation” in the tumor microenvironment through Batf3-DC regulation.

Figure 7

Open in a new tab

DC precursors develop in the bone marrow and travel via the bloodstream to seed peripheral tissues, including tumor sites. Although certain predilection towards a specific DC fate may be homeostatically programmed in the bone marrow (47), pre-cDC differentiation appears mostly influenced by local microenvironmental factors in the inflamed and malignant settings associated with emergency myelopoiesis (48). We propose that matrikines generated through VCAN proteolysis (such as versikine) promote pre-DC differentiation towards Batf3-dependent DC that are critical for T-cell immunity and response to immunotherapy modalities (21–28, 37, 49). Therefore, VCAN proteolysis may promote “T-cell inflammation” (50, 51) through Batf3-DC regulation. HSC, hematopoietic stem cell; CMP, common myeloid progenitor; MDP, myeloid-dendritic cell progenitor; CDP, common dendritic cell progenitor; pre-cDC, pre-conventional dendritic cell.

We observed intense VCAN proteolysis in normal colonic epithelium. The colon constitutes an immunologically active microenvironment that has evolved to cope with the continuous exposure to exogenous antigens provided by food processing as well as intestinal microbiota (35). The implications of this regulation are profound and bear significance well beyond the confines of the gastrointestinal tract. Importantly, a correlation between the composition of intestinal flora and degree of response to anti-tumor immunotherapy is established and beginning to be clinically exploited (36–38). The mechanisms accounting for the regulation and “fine-tuning” of immune responses in normal colonic epithelium are poorly understood (39). It is tempting to associate VCAN processing, and the resultant generation of bioactive immunoregulatory fragments, with homeostatic DC maturation in normal colon. Because the effects of intestinal microbiota on anti-tumor immunity are thought to be regulated at the level of DCs, we hypothesize that VCAN proteolysis may collaborate with the microbiome to influence immune priming against distally located tumors. Alternatively or additionally, VCAN proteolysis may specifically shape the immunological milieu of the normal epithelium located adjacent to the “expanding rim” of colonic cancers.

Moreover, VCAN accumulation and turnover may impact on the local immunoregulation of several types of solid tumors that arise in normally “sterile” sites. For example, in prostate tissue, immunosuppressive signaling from TGFβ increases expression of VCAN, reduces expression of VCAN-cleaving ADAMTS proteases and enhances expression of ADAMTS metalloproteinase inhibitor, TIMP-3 (40). Interestingly, prostate cancer constitutes another common type of solid tumor that has yet to benefit from the recent advances in immunotherapy (41). It is intriguing to hypothesize that the VCAN-versikine axis may regulate immune infiltration across a wide spectrum of solid tumors.

Our data confirm and extend previous findings regarding the mechanisms regulating T-cell infiltration or exclusion from the tumor site. In particular, we confirm previous observations implicating melanoma-intrinsic WNT signaling in T-cell exclusion and extend these findings to CRC (28). Mechanistic analyses in melanoma suggested that WNT signaling acts through CCL4 to regulate tumor infiltration by Batf3-lineage DC (CD103+ DC in peripheral tissues). Our data raise the testable hypothesis that WNT signaling enhances VCAN accumulation in the tumor microenvironment, potentially through the recruitment of immunosuppressive, VCAN-producing, macrophages or inhibition of local ADAMTS-1 production. VCAN promotes DC dysfunction through Toll-like receptor-2 (TLR2) signaling (14). It is tempting to speculate that tumor-intrinsic WNT signaling radically remodels the myeloid immune contexture of the tumor through inhibition of immunogenic, Batf3-DC together with recruitment of immunosuppressive, VCAN-producing, macrophages. We are currently testing these hypotheses.

The data presented in this manuscript suggest that VCAN processing may influence the balance between tolerogenic and immunogenic inflammation in common solid tumors. Further to our earlier work (15), corroborating evidence has lately come from different angles. A recent paper suggested a link between VCAN turnover and anti-viral T-cell responses in mice (42). We speculate the analogous mechanisms may operate during innate immune sensing of tumors (43). VCAN-producing, immunosuppressive macrophages were shown to expand post-therapy in myeloma and inhibit T-cell proliferation (44). Our results are broadly consistent with an earlier report that demonstrated inverse correlation between stromal VCAN abundance and T-cell infiltration in cervical carcinoma, although this report did not assess extracellular proteolysis (45). The abundance of VCAN in CRCs is likely regulated both at the transcriptional level through WNT signaling and post-translationally, through ADAMTS proteases encoded by loci that are epigenetically regulated upon CRC progression (46). The data provide a rationale for investigating VCAN proteolysis as a novel immune biomarker in solid tumor settings. Moreover, therapeutic manipulation of the VCAN-versikine axis through targeted proteolysis of VCAN or administration of the VCAN-matrikine, versikine, could be clinically tested for synergy with modern immunotherapy modalities against CRC regardless of mismatch repair status.

Supplementary Material

1

Acknowledgments

Grant Support: This work was supported by the American Cancer Society (RSG-15-01-LIB) (FA), the UWCCC Trillium Fund for Multiple Myeloma Research (FA), Funk Out Cancer (DAD), Bowlin’ for Colons (DAD), Cathy Wingert Colorectal Cancer Research Fund (DAD), V Foundation Scholars Award (DAD) and the NIH (P30CA014520 and T32HL007899). The authors also thank the Cellular and Molecular Pathology training program (PBE).

We wish to thank Dr. Suneel S. Apte, Cleveland Clinic Lerner Research Institute, Cleveland Clinic, Cleveland, OH; Dr. Paul Sondel and Dr. Alexander Rakhmilevich, both at UW-Madison, Madison, WI, for valuable discussions and continued collaboration; Dr. Scott Abrams at Roswell Park Cancer Institute, Buffalo, NY for provision of Irf8-EGFP tissue and valuable collaboration and advice; and Dr. Deane Mosher (UW- Madison) for help with biochemical characterization and validation of recombinant versikine.

Footnotes

Conflict of Interest: The authors have no relevant conflicts of interest to declare.

Author contributions: CH, PBE, AP1, AP2, SP, DV, and MGJ performed experiments. PBE, KAM, LC, DAD and FA analyzed experiments. NC, PH, and SM provided crucial expertise and/or reagents. FA wrote the first draft of the manuscript. All authors reviewed, critiqued and edited the manuscript.

References

  • 1.Marley AR, Nan H. Epidemiology of colorectal cancer. Int J Mol Epidemiol Genet. 2016;7:105–114. [PMC free article] [PubMed] [Google Scholar]
  • 2.Lynch D, Murphy A. The emerging role of immunotherapy in colorectal cancer. Ann Transl Med. 2016;4:305. doi: 10.21037/atm.2016.08.29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Topalian SL, Hodi FS, Brahmer JR, Gettinger SN, Smith DC, McDermott DF, Powderly JD, Carvajal RD, Sosman JA, Atkins MB, Leming PD, Spigel DR, Antonia SJ, Horn L, Drake CG, Pardoll DM, Chen L, Sharfman WH, Anders RA, Taube JM, McMiller TL, Xu H, Korman AJ, Jure-Kunkel M, Agrawal S, McDonald D, Kollia GD, Gupta A, Wigginton JM, Sznol M. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443–2454. doi: 10.1056/NEJMoa1200690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Westdorp H, Fennemann FL, Weren RD, Bisseling TM, Ligtenberg MJ, Figdor CG, Schreibelt G, Hoogerbrugge N, Wimmers F, de Vries IJ. Opportunities for immunotherapy in microsatellite instable colorectal cancer. Cancer Immunol Immunother. 2016;65:1249–1259. doi: 10.1007/s00262-016-1832-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Bupathi M, Wu C. Biomarkers for immune therapy in colorectal cancer: mismatch-repair deficiency and others. J Gastrointest Oncol. 2016;7:713–720. doi: 10.21037/jgo.2016.07.03. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pages C, Tosolini M, Camus M, Berger A, Wind P, Zinzindohoue F, Bruneval P, Cugnenc PH, Trajanoski Z, Fridman WH, Pages F. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960–1964. doi: 10.1126/science.1129139. [DOI] [PubMed] [Google Scholar]
  • 7.Hope C, Ollar SJ, Heninger E, Hebron E, Jensen JL, Kim J, Maroulakou I, Miyamoto S, Leith C, Yang DT, Callander N, Hematti P, Chesi M, Bergsagel PL, Asimakopoulos F. TPL2 kinase regulates the inflammatory milieu of the myeloma niche. Blood. 2014;123:3305–3315. doi: 10.1182/blood-2014-02-554071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Du WW, Yang W, Yee AJ. Roles of versican in cancer biology--tumorigenesis, progression and metastasis. Histol Histopathol. 2013;28:701–713. doi: 10.14670/HH-28.701. [DOI] [PubMed] [Google Scholar]
  • 9.Ricciardelli C, Sakko AJ, Ween MP, Russell DL, Horsfall DJ. The biological role and regulation of versican levels in cancer. Cancer Metastasis Rev. 2009;28:233–245. doi: 10.1007/s10555-009-9182-y. [DOI] [PubMed] [Google Scholar]
  • 10.Wight TN, Kang I, Merrilees MJ. Versican and the control of inflammation. Matrix Biol. 2014;35:152–161. doi: 10.1016/j.matbio.2014.01.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Zhang Z, Miao L, Wang L. Inflammation amplification by versican: the first mediator. Int J Mol Sci. 2012;13:6873–6882. doi: 10.3390/ijms13066873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Gao D, Joshi N, Choi H, Ryu S, Hahn M, Catena R, Sadik H, Argani P, Wagner P, Vahdat LT, Port JL, Stiles B, Sukumar S, Altorki NK, Rafii S, Mittal V. Myeloid progenitor cells in the premetastatic lung promote metastases by inducing mesenchymal to epithelial transition. Cancer Res. 2012;72:1384–1394. doi: 10.1158/0008-5472.CAN-11-2905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kim S, Takahashi H, Lin WW, Descargues P, Grivennikov S, Kim Y, Luo JL, Karin M. Carcinoma-produced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature. 2009;457:102–106. doi: 10.1038/nature07623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Tang M, Diao J, Gu H, Khatri I, Zhao J, Cattral MS. Toll-like Receptor 2 Activation Promotes Tumor Dendritic Cell Dysfunction by Regulating IL-6 and IL-10 Receptor Signaling. Cell Rep. 2015;13:2851–2864. doi: 10.1016/j.celrep.2015.11.053. [DOI] [PubMed] [Google Scholar]
  • 15.Hope C, Foulcer S, Jagodinsky J, Chen SX, Jensen JL, Patel S, Leith C, Maroulakou I, Callander N, Miyamoto S, Hematti P, Apte SS, Asimakopoulos F. Immunoregulatory roles of versican proteolysis in the myeloma microenvironment. Blood. 2016;128:680–685. doi: 10.1182/blood-2016-03-705780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Boyd DF, Thomas PG. Towards integrating extracellular matrix and immunological pathways. Cytokine. 2017 doi: 10.1016/j.cyto.2017.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Gaggar A, Weathington N. Bioactive extracellular matrix fragments in lung health and disease. J Clin Invest. 2016;126:3176–3184. doi: 10.1172/JCI83147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Asimakopoulos F, Hope C, Johnson MG, Pagenkopf A, Gromek K, Nagel B. Extracellular matrix and the myeloid-in-myeloma compartment: balancing tolerogenic and immunogenic inflammation in the myeloma niche. J Leukoc Biol. 2017 doi: 10.1189/jlb.3MR1116-468R. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Nandadasa S, Foulcer S, Apte SS. The multiple, complex roles of versican and its proteolytic turnover by ADAMTS proteases during embryogenesis. Matrix Biol. 2014;35:34–41. doi: 10.1016/j.matbio.2014.01.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Wang K, Karin M. Tumor-Elicited Inflammation and Colorectal Cancer. Adv Cancer Res. 2015;128:173–196. doi: 10.1016/bs.acr.2015.04.014. [DOI] [PubMed] [Google Scholar]
  • 21.Lavin Y, Kobayashi S, Leader A, Amir ED, Elefant N, Bigenwald C, Remark R, Sweeney R, Becker CD, Levine JH, Meinhof K, Chow A, Kim-Shulze S, Wolf A, Medaglia C, Li H, Rytlewski JA, Emerson RO, Solovyov A, Greenbaum BD, Sanders C, Vignali M, Beasley MB, Flores R, Gnjatic S, Pe’er D, Rahman A, Amit I, Merad M. Innate Immune Landscape in Early Lung Adenocarcinoma by Paired Single-Cell Analyses. Cell. 2017;169:750–765. e717. doi: 10.1016/j.cell.2017.04.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Laoui D, Keirsse J, Morias Y, Van Overmeire E, Geeraerts X, Elkrim Y, Kiss M, Bolli E, Lahmar Q, Sichien D, Serneels J, Scott CL, Boon L, De Baetselier P, Mazzone M, Guilliams M, Van Ginderachter JA. The tumour microenvironment harbours ontogenically distinct dendritic cell populations with opposing effects on tumour immunity. Nat Commun. 2016;7:13720. doi: 10.1038/ncomms13720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Spranger S, Dai D, Horton B, Gajewski TF. Tumor-Residing Batf3 Dendritic Cells Are Required for Effector T Cell Trafficking and Adoptive T Cell Therapy. Cancer Cell. 2017;31:711–723. e714. doi: 10.1016/j.ccell.2017.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Broz ML, Binnewies M, Boldajipour B, Nelson AE, Pollack JL, Erle DJ, Barczak A, Rosenblum MD, Daud A, Barber DL, Amigorena S, Van’t Veer LJ, Sperling AI, Wolf DM, Krummel MF. Dissecting the tumor myeloid compartment reveals rare activating antigen-presenting cells critical for T cell immunity. Cancer Cell. 2014;26:638–652. doi: 10.1016/j.ccell.2014.09.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ruffell B, Chang-Strachan D, Chan V, Rosenbusch A, Ho CM, Pryer N, Daniel D, Hwang ES, Rugo HS, Coussens LM. Macrophage IL-10 blocks CD8+ T cell-dependent responses to chemotherapy by suppressing IL-12 expression in intratumoral dendritic cells. Cancer Cell. 2014;26:623–637. doi: 10.1016/j.ccell.2014.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Salmon H, Idoyaga J, Rahman A, Leboeuf M, Remark R, Jordan S, Casanova-Acebes M, Khudoynazarova M, Agudo J, Tung N, Chakarov S, Rivera C, Hogstad B, Bosenberg M, Hashimoto D, Gnjatic S, Bhardwaj N, Palucka AK, Brown BD, Brody J, Ginhoux F, Merad M. Expansion and Activation of CD103(+) Dendritic Cell Progenitors at the Tumor Site Enhances Tumor Responses to Therapeutic PD-L1 and BRAF Inhibition. Immunity. 2016;44:924–938. doi: 10.1016/j.immuni.2016.03.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Sanchez-Paulete AR, Cueto FJ, Martinez-Lopez M, Labiano S, Morales-Kastresana A, Rodriguez-Ruiz ME, Jure-Kunkel M, Azpilikueta A, Aznar MA, Quetglas JI, Sancho D, Melero I. Cancer Immunotherapy with Immunomodulatory Anti-CD137 and Anti-PD-1 Monoclonal Antibodies Requires BATF3-Dependent Dendritic Cells. Cancer Discov. 2016;6:71–79. doi: 10.1158/2159-8290.CD-15-0510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Spranger S, Bao R, Gajewski TF. Melanoma-intrinsic beta-catenin signalling prevents anti-tumour immunity. Nature. 2015;523:231–235. doi: 10.1038/nature14404. [DOI] [PubMed] [Google Scholar]
  • 29.Foulcer SJ, Day AJ, Apte SS. Isolation and purification of versican and analysis of versican proteolysis. Methods Mol Biol. 2015;1229:587–604. doi: 10.1007/978-1-4939-1714-3_46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Brasel K, De Smedt T, Smith JL, Maliszewski CR. Generation of murine dendritic cells from flt3-ligand-supplemented bone marrow cultures. Blood. 2000;96:3029–3039. [PubMed] [Google Scholar]
  • 31.Markowitz SD, Bertagnolli MM. Molecular origins of cancer: Molecular basis of colorectal cancer. N Engl J Med. 2009;361:2449–2460. doi: 10.1056/NEJMra0804588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sichien D, Scott CL, Martens L, Vanderkerken M, Van Gassen S, Plantinga M, Joeris T, De Prijck S, Vanhoutte L, Vanheerswynghels M, Van Isterdael G, Toussaint W, Madeira FB, Vergote K, Agace WW, Clausen BE, Hammad H, Dalod M, Saeys Y, Lambrecht BN, Guilliams M. IRF8 Transcription Factor Controls Survival and Function of Terminally Differentiated Conventional and Plasmacytoid Dendritic Cells, Respectively. Immunity. 2016 doi: 10.1016/j.immuni.2016.08.013. [DOI] [PubMed] [Google Scholar]
  • 33.Mayer CT, Ghorbani P, Nandan A, Dudek M, Arnold-Schrauf C, Hesse C, Berod L, Stuve P, Puttur F, Merad M, Sparwasser T. Selective and efficient generation of functional Batf3-dependent CD103+ dendritic cells from mouse bone marrow. Blood. 2014;124:3081–3091. doi: 10.1182/blood-2013-12-545772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Wu YJ, La Pierre DP, Wu J, Yee AJ, Yang BB. The interaction of versican with its binding partners. Cell Res. 2005;15:483–494. doi: 10.1038/sj.cr.7290318. [DOI] [PubMed] [Google Scholar]
  • 35.Kamada N, Nunez G. Regulation of the immune system by the resident intestinal bacteria. Gastroenterology. 2014;146:1477–1488. doi: 10.1053/j.gastro.2014.01.060. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Pitt JM, Vetizou M, Waldschmitt N, Kroemer G, Chamaillard M, Boneca IG, Zitvogel L. Fine-Tuning Cancer Immunotherapy: Optimizing the Gut Microbiome. Cancer Res. 2016;76:4602–4607. doi: 10.1158/0008-5472.CAN-16-0448. [DOI] [PubMed] [Google Scholar]
  • 37.Spranger S, Sivan A, Corrales L, Gajewski TF. Tumor and Host Factors Controlling Antitumor Immunity and Efficacy of Cancer Immunotherapy. Adv Immunol. 2016;130:75–93. doi: 10.1016/bs.ai.2015.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Goldszmid RS, Dzutsev A, Viaud S, Zitvogel L, Restifo NP, Trinchieri G. Microbiota modulation of myeloid cells in cancer therapy. Cancer Immunol Res. 2015;3:103–109. doi: 10.1158/2326-6066.CIR-14-0225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Coombes JL, Powrie F. Dendritic cells in intestinal immune regulation. Nat Rev Immunol. 2008;8:435–446. doi: 10.1038/nri2335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Cross NA, Chandrasekharan S, Jokonya N, Fowles A, Hamdy FC, Buttle DJ, Eaton CL. The expression and regulation of ADAMTS-1, -4, -5, -9, and -15, and TIMP-3 by TGFbeta1 in prostate cells: relevance to the accumulation of versican. Prostate. 2005;63:269–275. doi: 10.1002/pros.20182. [DOI] [PubMed] [Google Scholar]
  • 41.Rekoske BT, McNeel DG. Immunotherapy for prostate cancer: False promises or true hope? Cancer. 2016;122:3598–3607. doi: 10.1002/cncr.30250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.McMahon M, Ye S, Izzard L, Dlugolenski D, Tripp RA, Bean AG, McCulloch DR, Stambas J. ADAMTS5 Is a Critical Regulator of Virus-Specific T Cell Immunity. PLoS Biol. 2016;14:e1002580. doi: 10.1371/journal.pbio.1002580. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Woo SR, Corrales L, Gajewski TF. Innate immune recognition of cancer. Annu Rev Immunol. 2015;33:445–474. doi: 10.1146/annurev-immunol-032414-112043. [DOI] [PubMed] [Google Scholar]
  • 44.Arana P, Zabaleta A, Lasa M, Maiso P, Alignani D, Jelinek T, Segura V, van den Bossche W, Teodosio C, Damasceno D, Rodriguez-Otero P, Prosper F, Mateos M-V, Lahuerta JJ, Bladé J, van Dongen JJM, San Miguel J, Orfao A, Paiva B. High-Throughput Characterization and New Insight into the Role of Tumor Associated Macrophages (TAMs) in Multiple Myeloma (MM) Blood. 2016;128:482–482. [Google Scholar]
  • 45.Gorter A, Zijlmans HJ, van Gent H, Trimbos JB, Fleuren GJ, Jordanova ES. Versican expression is associated with tumor-infiltrating CD8-positive T cells and infiltration depth in cervical cancer. Mod Pathol. 2010;23:1605–1615. doi: 10.1038/modpathol.2010.154. [DOI] [PubMed] [Google Scholar]
  • 46.Lind GE, Kleivi K, Meling GI, Teixeira MR, Thiis-Evensen E, Rognum TO, Lothe RA. ADAMTS1, CRABP1, and NR3C1 identified as epigenetically deregulated genes in colorectal tumorigenesis. Cell Oncol. 2006;28:259–272. doi: 10.1155/2006/949506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Schlitzer A, Sivakamasundari V, Chen J, Sumatoh HR, Schreuder J, Lum J, Malleret B, Zhang S, Larbi A, Zolezzi F, Renia L, Poidinger M, Naik S, Newell EW, Robson P, Ginhoux F. Identification of cDC1- and cDC2-committed DC progenitors reveals early lineage priming at the common DC progenitor stage in the bone marrow. Nat Immunol. 2015;16:718–728. doi: 10.1038/ni.3200. [DOI] [PubMed] [Google Scholar]
  • 48.Rutkowski MR, Svoronos N, Perales-Puchalt A, Conejo-Garcia JR. The Tumor Macroenvironment: Cancer-Promoting Networks Beyond Tumor Beds. Adv Cancer Res. 2015;128:235–262. doi: 10.1016/bs.acr.2015.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW, Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015;11:1018–1030. doi: 10.1016/j.celrep.2015.04.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Gajewski TF. The Next Hurdle in Cancer Immunotherapy: Overcoming the Non-T-Cell-Inflamed Tumor Microenvironment. Semin Oncol. 2015;42:663–671. doi: 10.1053/j.seminoncol.2015.05.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Zitvogel L, Galluzzi L, Kepp O, Smyth MJ, Kroemer G. Type I interferons in anticancer immunity. Nat Rev Immunol. 2015;15:405–414. doi: 10.1038/nri3845. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS891605-supplement-1.pdf (596.8KB, pdf)

RESOURCES