1 3

Cancer Immunol Immunother (2015) 64:1137–1149DOI 10.1007/s00262-015-1719-z

ORIGINAL ARTICLE

Modulating the innate immune activity in murine tumor microenvironment by a combination of inducer molecules attached to microparticles

Ehud Shahar1,2 · Raphael Gorodetsky2 · Elina Aizenshtein1 · Lior Lalush1,3 · Jacob Pitcovski1,3

Received: 30 December 2014 / Accepted: 21 May 2015 / Published online: 2 June 2015 © Springer-Verlag Berlin Heidelberg 2015

C5aR ligand C5a-pep on the same MP resulted in a similar inflammation activation pattern. However, interleukin-10 levels were lower, and tumor growth was significantly delayed. Mixtures of these two ligands on separate MP did not yield the same cytokine activation pattern, demonstrat-ing the importance of the cells’ dual activation. The results suggest that combining inducers of distinct innate immune activation pathways holds promise for successful redirec-tion of TME-residing IIC toward anti-tumoral activation.

Keywords Cancer immunotherapy · Innate immunity · Tumor microenvironment · Microparticle · Immunological inducer

AbbreviationsC5a-pep C5a receptor agonist hexapeptideC5aR Complement C5a receptorCCL Chemokine (C-C motif) ligandCFU Colony-forming unitsCXCL Chemokine (C-X-C motif) ligandDC Dendritic cellsELISA Enzyme-linked immunosorbent assayFACS Fluorescence-activated cell sortingFcR Fc receptorsFCS Fetal calf serumFcγR Fcγ receptori.t. IntratumorallyIIC Innate immune cellsIL InterleukinLPS LipopolysaccharideMDSC Myeloid-derived suppressor cellsmIgG Mouse IgGMP MicroparticlesROS Reactive oxygen speciesTAA Tumor-associated antigens

Abstract Targeted cancer immunotherapy is challenging due to the cellular diversity and imposed immune toler-ance in the tumor microenvironment (TME). A promising route to overcome those drawbacks may be by activat-ing innate immune cells (IIC) in the TME, toward tumor destruction. Studies have shown the ability to “re-educate” pro-tumor-activated IIC toward antitumor responses. The current research aims to stimulate such activation using a combination of innate activators loaded onto micropar-ticles (MP). Four inducers of Toll-like receptors 4 and 7, complement C5a receptor (C5aR) and gamma Fc receptor and their combinations were loaded on MP, and their influ-ence on immune cell activation evaluated. MP stimulation of immune cell activation was tested in vitro and in vivo using a subcutaneous B16-F10 melanoma model induced in C57BL6 mice. Exposure to the TLR4 ligand lipopoly-saccharide (LPS) bound to MP-induced acute inflamma-tory cytokine and chemokine activity in vitro and in vivo, with the elevation of CD45+ leukocytes in particular GR-1+ neutrophils and F4/80 macrophages in the TME. Nevertheless, LPS alone on MP was insufficient to signifi-cantly delay tumor progression. LPS combined with the

Electronic supplementary material The online version of this article (doi:10.1007/s00262-015-1719-z) contains supplementary material, which is available to authorized users.

* Jacob Pitcovski jp@migal.org.il

1 MIGAL – Galilee Research Institute, P.O. Box 831, 11016 Kiryat Shmona, Israel

2 Lab of Biotechnology and Radiobiology, Sharett Institute of Oncology, Hadassah – Hebrew University Medical Center, Jerusalem, Israel

3 Tel Hai Academic College, Upper Galilee, Israel

1138 Cancer Immunol Immunother (2015) 64:1137–1149

1 3

TAM Tumor-associated macrophagesTAN Tumor-associated neutrophilsTGF-β Transforming growth factor-βTLR Toll-like receptorsTME Tumor microenvironmentTNF-α Tumor necrosis factor-αVEGF Vascular endothelial growth factor

Introduction

Advances in immunotherapy of cancer have given rise to promising new therapies of which only a selected few have been applied clinically. A major strategy focuses on the activation of CD8 T lymphocytes, relying on the presenta-tion of tumor-associated antigens (TAA) by cancer cells [1, 2]. However, anti-tumoral immune activation toward TAA may be evaded by down-regulation or alteration in these cancer cell antigens [3]. Another major obstacle to the antitumor immune response is the presentation by cancer cells of inhibitory regulators of immune checkpoints, e.g., programmed death ligand l, together with the induction of immune tolerance by molecules secreted into the tumor microenvironment (TME) [3–6]. The TME contains a mul-titude of innate immune cells (IIC) which, in normal tissue, play a crucial role in monitoring tissue homeostasis, pro-tecting against pathogens and eliminating damaged cells; thus, they seem qualified to eliminate cancer cells [7, 8]. However, IIC found in the TME, such as tumor-associated macrophages (TAM), tumor-associated neutrophils (TAN) and myeloid-derived suppressor cells (MDSC), have been shown to contribute to tumor survival, propagation and metastatic spread [5, 8–11]. These “alternatively” activated tumor-associated IIC have been shown to promote angio-genesis, tumor scaffolding and regeneration by secreting factors such as vascular endothelial growth factor (VEGF) and transforming growth factor-β (TGF-β) together with immunosuppressive molecules that restrict antitumor acti-vation, such as interleukin (IL)-10 [5, 7, 12–14].

In spite of the above findings, TAM may still regain cytotoxicity toward tumor cells and switch to a “classi-cally” activated phenotype following “re-education” [15]. Polymorphonuclear neutrophils have been found to con-tribute to both tumor rejection and tumor promotion [11, 16]. Pro-cancerous “polarization” of these cells may be switched to antitumor behavior by defined inducers [17], and this may alter the nature of the interaction between the tumor and the immune system in the TME, thereby ena-bling the desired immunological responses.

Various molecules have been shown to enhance anti-tumor responses by inducing immune cells via various types of receptors, including Toll-like receptors (TLR), Fc receptors (FcR) and complement anaphylatoxin receptor.

TLR, preferentially expressed on dendritic cells and mac-rophages, function as molecular sensors that detect path-ogen-associated molecular patterns, triggering secretion of chemokines and cytokines to activate innate and adap-tive immune cells [18]. The roles of diverse TLR in can-cer immunotherapy have been described in many studies (reviewed by Pradere et al. [19]). In human breast can-cer cells, for example, apoptosis was induced by double-stranded RNA in a TLR3-dependent manner [20]. In mouse models, immunological antitumor activity was induced following stimulation of TLR9 by synthetic oligode-oxynucleotides with unmethylated CpG motifs [21, 22]. Tumor regression and long-term antitumor immunity were achieved by treating mice with synthetic RNA that was an agonist of TLR7 and TLR8 [23–25]. In another animal model, TLR5 activation following flagellin administration resulted in tumor growth inhibition [26]. Those studies sug-gest that TLR-mediated antitumor activity involves several mechanisms, including enabling of the immune response of CD8(+) T cells by decreasing the number of regula-tory T cells [27] and reversing the suppressive function of various types of innate and acquired immune cells [28, 29]. Another mechanism might be based on triggering the secretion of cytokines involved in the differentiation and induction of potent adaptive immunity toward tumor anti-gens [30].

FcR binds the Fc domain of antibodies that are attached to antigens, enhancing cellular uptake of the complex. This receptor is expressed on most hematopoietic cells, e.g., monocytes, macrophages, interferon-γ-activated neu-trophils, natural killer cells, mast cells and dendritic cells. Complexes of Fc–FcR may induce an antitumor response as summarized by Cassard et al. [31]. In some cancers, following IgG recognition of cancer cells, Fcγ receptor (FcγR) on IIC triggers an antitumor response [31].

Complement anaphylatoxin receptors are expressed by several myeloid-derived IIC, including monocytes, mac-rophages, dendritic cells (DC) neutrophils, basophils, mast cells and eosinophils. They bind complement compo-nents, such as C3a and C5a derived from complement cas-cade activation, mediating chemotaxis and inflammatory responses [32]. The controversial effects of C3a and C5a, displaying both pro-tumoral and anti-tumoral responses, have been reviewed by Sayegh et al. [32].

Recent studies have shown additional effects of com-binations of ligands from several activation pathways on immune activation, such as multiple TLR ligands and com-binations with complement components, antibodies and mannose [33–35].

The basic idea behind the present study was to activate immune mechanisms similar to those directed at rejecting foreign bodies (e.g., pathogens or transplants), with the aim of “re-educating” tumor-residing IIC to attract cells from

1139Cancer Immunol Immunother (2015) 64:1137–1149

1 3

both the innate and adaptive arms of the immune system to the TME and activate them toward a tumoricidal effect. We therefore stimulated TME-residing IIC with inducers of innate cell activation loaded onto microparticles (MP). Four distinct routes of activation, utilizing ligands of the membranous TLR4 [36], TLR7 found on endocytic vesi-cles [36], complement C5a receptor (C5aR) and FcγR, and their combinations, were examined for immunological and anti-tumoral responses.

Materials and methods

Primary splenocytes and cell lines

Single-cell suspensions of splenocytes were prepared from male C57BL6 mice (8–12 weeks old). Harvested spleens were dissociated in PBS using gentleMACS™ dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany). Following erythrocyte lysis with distilled water, splenocytes were passed through a 70-µm cell strainer (BD Biosciences, Bedford, MA), then washed with PBS and transferred into culture media.

Mouse melanoma cell line B16-F10, human monocyte cell line THP-1 and hybridoma-produced mouse mono-clonal IgG type-1 antibody 44-3A6 were purchased from the American Type Culture Collection (Rockville, MD). Mouse dendritic cell line DC2.4 was kindly provided by K. Rock (Dana-Farber Cancer Institute, Boston, MA).

Splenocytes and 44-3A6 hybridomas were cultured in RPMI 1640 media supplemented with 10 % fetal calf serum (FCS: Gibco, Grand Island, NY), 1 % l-glutamate and 1 % penicillin–streptomycin solution (Biological Industries, Beit Haemek, Israel). The medium for DC2.4 cells was fur-ther supplemented with 50 μM β-mercaptoethanol. THP-1 cells were grown in RPMI 1640 media supplemented with 2 % l-glutamate, 1 % sodium pyruvate and 5 % FCS, with-out antibiotics. B16-F10 cells were grown in DMEM sup-plemented with 10 % FCS, 2 % l-glutamate, 1 % sodium pyruvate and 1 % penicillin–streptomycin solution.

Preparation of immune‑stimulating ligands and MP assembly

Mouse IgG (mIgG) for the stimulation of FcγR was purified from the supernatant of the 44-3A6 culture, using a protein G Sepharose® 4B fast flow column (Sigma-Aldrich, Reho-vot, Israel). Purified mIgG was biotinylated using EZ-Link Biotin LC sulfo-NHS (Pierce, Appleton, WI), and lipopoly-saccharide (LPS; L-2654, Sigma-Aldrich) was biotinylated using EZ-Link biotin LC hydrazide (Pierce), as previously described [34, 37]. The C5a receptor agonist hexapep-tide (N-Methyl-Phe)-Lys-Pro-D-Cha-Cha-D-Arg-CO2H, described in detail by Konteatis et al. [38], was synthesized

with the addition of biotin at the C-terminal position (Gen-Script, Piscataway, NJ) and was designated C5a-pep. CL264 was purchased and biotinylated from InvivoGen (San Diego, CA). BioMag® magnetic particles (Bangslabs, Fishers, IN) were purchased covalently coated with avidin. To produce particle complexes, biotinylated inducers LPS, mIgG, C5a-pep or CL264, individually or in combination, were mixed with avidin-coated MP for 1 h and then stored at 4 °C. Each single component was designated to bind at a ratio of ~1/3 of the particles’ biotin-binding capacity as stated by the manufacturer. Prior to use, the designated MP were taken out of storage and washed twice with PBS.

THP‑1 cell induction by C5a‑pep

THP-1 cells (105) were suspended in 1-ml medium and mixed with 100 µg C5a-pep, or PBS as a control. Escheri-chia coli bacteria (104 in 100 µl) were then added with gentle mixing at 37 °C. Samples of 100 µl were withdrawn imme-diately (0 min), and after 30 and 60 min and chilled on ice. Samples were centrifuged (300g for 5 min), and supernatant was seeded on agar plates and incubated for 24 h at 37 °C. E. coli colony-forming units (CFU) were then counted.

MP induction of cytokine expression determined by semiquantitative RT‑PCR

DC2.4 cells or splenocytes were incubated overnight in 24-well plates (4 × 106 cell/well). The relevant MP were added (10 µl from stock) to the wells and incubated for 4 h. Following incubation, medium was discarded and cells were dissolved in TRI-Reagent® (Sigma-Aldrich). B16-F10 tumors were homogenized in TRI-Reagent® (1 ml for 100 mg of tissue) using gentleMACS™ dissociator.

RNA was further extracted according to the TRI-Rea-gent® manufacturer’s protocol. Reverse transcription was performed for total mRNA using the Verso™ kit protocol (Fermentas, Vilnius, Lithuania) and oligo dT primers. PCR was performed for cytokine and chemokine gene products using DreamTaq Green PCR kit (Fermentas). The primers used are listed in Supplementary Table 1.

PCR products were separated by electrophoresis on a 2.3 % agarose gel and stained with ethidium bromide. Induc-tion levels cytokine and chemokine PCR products were ana-lyzed relative to GAPDH using ChemiDoc XRS+ System and Quantity One software (both by BioRad, Hercules, CA).

Detection of secreted cytokines and chemokines in splenocyte culture by enzyme‑linked immunosorbent assay (ELISA)

Splenocytes (5 × 106 per well) were incubated overnight in 24-well plates. The relevant MP were then added (10 µl

1140 Cancer Immunol Immunother (2015) 64:1137–1149

1 3

from stock) to the wells and incubated for an additional 16 or 48 h. The medium was collected and centrifuged (500g for 5 min). The supernatant levels of the secreted cytokines IL-6, tumor necrosis factor-α (TNF-α) and IL-1β, and chemokine (C-X-C motif) ligand (CXCL) 1 were measured using a murine cytokine-detecting ELISA kit according to the manufacturer’s protocol (PeproTech, Rocky Hill, NJ).

Fluorescence‑activated cell sorting (FACS)

Tumors were dissociated using gentleMACS™ dissociator as already described and then transferred into FACS buffer made up of PBS supplemented with 0.1 % bovine serum albumin and 0.05 % sodium azide. Cells (106) were stained with antibodies against CD45 (PE/Cy7), CD3 (APC) combined with 33D1 (PE), CD49b (Pacific blue), GR-1 (PerCP) and F4/80 (APC/Cy7) or CD19 (Pacific blue), CD4 (PerCP/Cy5) and CD8 (PE) (eBioscience, San Diego, CA). FACS procedures were carried out by FACSAria (Becton–Dickinson, San Jose, CA). The FACS results were analyzed with FCS express 4 software (De Novo Software, Los Angeles, CA).

Animal studies

All animal experiments were carried out in accordance with the guidelines of the Israeli Ethics Committee. Female C57BL6 mice, 8–10 weeks of age, were kept with free access to food and water.

B16‑F10 tumor implantation and administration of MP carrying immune‑stimulating ligands

B16-F10 mouse melanoma cancer cells were cul-tured, trypsinized and washed twice with PBS. Cells (4 × 104) were suspended in 100 μl PBS and injected subcutaneously into the lateral flank. The volumes of the developing tumors were measured with a caliper [vol-ume = (width2 * length)/2]. The relevant MP were diluted twofold in PBS and injected (100 μl) intratumorally (i.t.).

Evaluation of immune activation in the TME following single injection of MP carrying immune‑stimulating ligands

Female C57BL6 mice carrying induced B16-F10 tumors were injected with MP carrying various inducing ligands when the tumor reached a volume of 200 ± 50 mm3. The mice were sacrificed 48 h post-MP injection. The tumors were harvested and weighed, then immune cell populations were determined by flow cytometry, and cytokine expres-sion was analyzed by semiquantitative RT-PCR.

Evaluation of sequential i.t. injections of MP carrying immune‑stimulating ligands

Animals carrying induced B16-F10 tumors were treated with four i.t. injections of MP carrying various induc-ing ligands. The first injection was administered when the tumor reached a volume of 50 ± 15 mm3, followed by three consecutive injections given at 2- or 4-day intervals. Ani-mals were killed when they showed distinct signs of mor-bidity or when the tumor reached a volume of >500 mm3. Tumors were harvested and analyzed as already described.

Statistical analysis

The statistical significance of the RT-PCR, ELISA and FACS results was evaluated by Student’s t test. Correla-tion coefficients between immune cell population ratios were analyzed using Pearson’s product-moment correlation analysis. Kaplan–Meier analysis was performed on mice from first injection until they showed signs of morbidity or tumor volume reached >500 mm3, and log-rank (Mantel–Cox) test was used to determine the significance of differ-ences in survival curves.

Results

To influence the immune response in the TME, four induc-ers of different innate immune activation pathways, loaded on MP, were tested: the TLR4 ligand LPS, TLR7 ligand CL264, C5a receptor ligand C5a-pep agonist and FcγR ligand mIgG.

Determination of cytokine expression induced by loaded MP

Incubation of DC2.4 cells with CL264-loaded MP resulted in elevated IL-1β levels, exceeding twofold that of soluble CL264, with a parallel rise in IL-6 transcription (Supple-mentary Fig. 1a). Exposure to C5a-pep did not significantly induce cytokine expression following incubation with any of the human or mouse immune cell lines tested (data not shown). However, in THP-1 monocyte cultures incubated with E. coli, addition of C5a-pep enhanced bacterial phago-cytosis activity (Supplementary Fig. 1b).

Immune activation of cells by MP was further tested in a primary culture of mice splenocytes. LPS-loaded MP dramatically elevated inflammatory cytokines such as IL-1β, IL-6 and TNF-α, with a rise in the levels of other associated chemokines, such as CXCL1 (murine analog to human IL-8), chemokine (C-C motif) ligand (CCL) 3 and CXCL10 (Figs. 1, 2). In addition, LPS-loaded MP trig-gered increased levels of the anti-inflammatory cytokine

1141Cancer Immunol Immunother (2015) 64:1137–1149

1 3

IL-10 (Fig. 1c). MP loaded with mIgG as sole inducer only induced a significant rise in TNF-α secretion, as shown after 16 h of incubation, with a parallel increasing trend for IL-1β and IL-6 levels (Fig. 2).

In splenocytes exposed to MP carrying a combination of mIgG and LPS (IgG–LPS-loaded MP), a significant increase in IL-6 transcription and secretion of CXCL1 and TNF-α were recorded after 48 h, although this increase was significantly lower than that induced by MP loaded only with LPS (Figs. 1, 2). When splenocytes were incubated with MP carrying C5a-pep alone or in combination with mIgG (IgG–C5a-loaded MP), there were no significant alterations in the transcription levels of any of the tested cytokines. Nevertheless, exposure to MP loaded with C5a-pep and LPS with (IgG–LPS–C5a-loaded MP) or without (LPS–C5a-loaded MP) mIgG decreased the transcription of the tested chemokines and totally abolished IL-10 tran-scription relative to exposure to LPS- or IgG–LPS-loaded MP (Fig. 1c).

Alterations in the immune cell populations in the TME following i.t. administration of MP

To monitor immune activation by MP carrying different ligands in vivo in the tumors, a low number of B16-F10

melanoma cells (4 × 104) were injected s.c. into C57BL6 mice, allowing relatively slow tumor development and more balanced leukocyte infiltration and activation. MP were injected into tumors that had reached a volume of 200 ± 50 mm3. In untreated or nonloaded MP (nMP; MP with no inducer)-treated mice, a significant negative cor-relation was observed between tumor developmental stage (as reflected by tumor volume and weight) and total leuko-cytes (CD45+), T cells (CD3+), cytotoxic T cells (CD8+) and natural killer cells (CD49b+) residing in the tumor (Fig. 3a). A similar trend was observed in untreated or nMP-treated mice for the progression of tumor develop-ment, which was accompanied by a decrease in all immune cell lineages tested (data not shown). These results could indicate either declining or stable proportions of immune cell populations in the tumor as it progresses, resulting in fewer such cells per unit volume of the tumor. Treatment of the tumors with inducer-loaded MP completely abol-ished the correlations between leukocyte number and tumor progression (Fig. 3b). Notably, the positive correlation between the percentage of tumor macrophages and neutro-phils out of the total CD45+ cell population in untreated and nMP-treated mice switched to a strong negative cor-relation in mice with inducer-treated tumors (Fig. 3). An overall change was observed in the balance of immune

Fig. 1 Cytokine transcription levels in splenocytes promoted by inducer-carrying MP. Semi-quantitative RT-PCR of cytokine levels following the incubation of mouse splenocytes with (left to right): nonloaded MP and MP carrying mIgG, C5a-pep (C5a), mIgG + C5a, LPS, mIgG + LPS, C5a + LPS or mIgG + C5a + LPS. The normalized expression of IL-6 (a), CXCL10 (b), IL-10 (c) and CCL3 (d) is presented (Mean ± SD *P < 0.05; **P < 0.01)

0

0.01

0.02

0.03

0.04

LPS

mIgGC5a

- - - - +++ +- - + + ++- -- + - + +-- +

MP composition

IL

)

6-no

rmal

ized

)IL

-10

(nor

mal

ized

)

CXC

L10

(nor

mal

ized

)C

CL3

(nor

mal

ized

)

0

0.5

1.0

1.5

LPS

mIgGC5a

- - - - +++ +- - + + ++- -- + - + +-- +

MP composition

LPS

mIgGC5a

- - - - +++ +- - + + ++- -- + - + +-- +

*

**

0

0.2

0.4

0.6

MP composition

0

0.4

0.8

1.2

1.6

LPS

mIgGC5a

- - - - +++ +- - + + ++- -- + - + +-- +

MP composition

(a) (b)

(c) (d)

1142 Cancer Immunol Immunother (2015) 64:1137–1149

1 3

Fig. 2 Cytokine secretion by splenocytes incubated with inducer-carrying MP. ELISA measuring cytokine levels after 16- or 48-h incubation of mouse splenocytes with (left to right): PBS, nonloaded MP and MP carrying LPS, mIgG or LPS + mIgG. Levels of IL-6 (a), CXCL1 (b), TNF-α (c) and IL-1β (d) are shown (Mean ± SD *P < 0.05; **P < 0.01; ***P < 0.001)

IL-6

(pg/

ml)

CXC

L1 (p

g/m

l) TN

F-α

(pg/

ml)

IL-1

β (p

g/m

l)

16 hours induction 48 hours induction

*

0

250

500

750

1000

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

0

250

500

750

1000

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

***

**

0

500

1000

1500

2000

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

*****

0

250

500

750

1000

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

***

0

500

1000

1500

2000

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

0

20

40

60

80

0100

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

0

250

500

750

1000

PBS LPSmIgG

+ +--- ++-

Controls

MP composition

(a)

(b)

(c)

(d)

1143Cancer Immunol Immunother (2015) 64:1137–1149

1 3

Fig. 3 Immune cell populations balance shift in tumors follow-ing treatments with loaded MP. Shown are correlation curves demonstrating the relationship between tumor volume meas-urements on the day of killing and (top to bottom): tumor weight, % of CD45+ cells, % of CD3+ T cells and % of CD49b+ natural killer cells from total cell counts. (a) Untreated mice or mice treated with nonloaded MP. (b) Mice treated with inducer-loaded MP. Immune cells found in the microenviron-ment were profiled by FACS. (c) The ratio between F4/80+ macrophages and GR-1+ granu-locytes, mainly neutrophils, from the CD45+ cell population (left and right correlate with a and b columns)

%C

D49

b of

tota

ltu

mor

eve

nts

%C

D3

of to

tal

tum

or e

vent

s

Tum

or

wei

ght (

mg)

%C

D45

of t

otal

tu

mor

eve

nts

%F4

/80

of t

umor

C

D45

cel

ls

Tumor volume (mm3)

%GR-1 of tumor CD45

0 200 400 600 8000

500

1000

1500r =0.837P <0.0001

0 200 400 600 8000

0.5

1.0

1.5

2.0

2.5r =-0.449P <0.01

0 200 400 600 8000

0.2

0.4

0.6

0.8r =-0.405P <0.05

0 200 400 600 8000

0.05

0.10

0.15r =-0.435P <0.05

0 5 10 15 20 250

20

40

60 r =0.418P <0.05

0 20 40 60 800

20

40

60r =-0.674P <0.0001

0 100 200 300 400 5000

0.05

0.10

0.15n.s

0 100 200 300 400 5000

0.2

0.4

0.6n.s.

0 100 200 300 400 5000

1

2

3

4n.s.

0 100 200 300 400 5000

200

400

600

800 r =0.708P <0.0001

(a) (b)

(c)

1144 Cancer Immunol Immunother (2015) 64:1137–1149

1 3

populations in the TME as a result of treatments with inducer-loaded MP. A summary of all altered correlations found between the tested immune cell populations with the inducer-loaded MP treatment as compared to untreated and nMP-treated tumors is presented in Supplementary Fig. 2.

LPS was the most potent mediator for % CD45+ cells in treated tumors, inducing an over twofold increase in overall immune cells relative to the nMP treatment (Fig. 4). Com-bining LPS with C5a-pep, CL264 or mIgG loaded on a MP did not significantly affect the influence of LPS on CD45+ levels (Fig. 4). The major cell populations that were elevated in the tumors in response to LPS-loaded MP treatments were GR-1+ neutrophils (2.8- to 5.6-fold) and F4/80+ mac-rophages (1.6- to 2-fold) compared to the nMP treatment (Fig. 4). Treating tumors with MP loaded with C5a, IgG and C5a, or CL264 did not significantly change leukocyte pop-ulation counts as compared to the nMP controls. However, treatment with C5a-pep-loaded MP showed a consistent, albeit insignificant decrease in T cell (CD3+) counts (Fig. 4).

B (CD19+), natural killer (CD49b) and dendritic (33D1+) cell populations were detected at extremely low levels in the TME and were not significantly altered by MP treatments (data not shown).

Alterations in cytokine and chemokine secretion in the TME following i.t. MP treatment

The changes in the transcription patterns of cytokines and chemokines in the TME were determined 48 h after treat-ment. Increased transcription levels of pro-inflammatory cytokines IL-1β and IL-6 were found. They seemed to be primarily influenced by the presence of LPS on the MP (Fig. 5a, b). LPS as sole inducer on MP caused no change in IL-10 transcription levels, whereas the combination of C5a-pep and LPS significantly decreased the transcription of this cytokine. Injection of C5a-loaded MP mixed 1:1 (v:v) with LPS-loaded MP resulted in significantly increased transcrip-tion of IL-10 as compared to nMP (Fig. 5c).

MP composition

- - - - LPSC5a-pepCL264mIgG

- - + -

- +

+ - - +

+ + + -

+ - -

+ - - -

- + -

+ - + -

0

1

2

3

4

CD45

*****

*

MP composition

- - - - -

- + - - +

+ - - +

+ + + -

+ - -

+ - - -

- + -

+ - + -

GR-1

0

0.5

1.0

1.5

2.0

* ***

* *

0.5

1.0

1.5

0

2.0

MP composition

- - - - -

- + - - +

+ - - +

+ + + -

+ - -

+ - - -

- + -

+ - + -

*****

*

F4/80

MP composition

- - - - LPSC5a-pep

CL264mIgG

- - + -

- +

+ - - +

+ + + -

+ - -

+ - - -

- + -

+ - + -

*

CD3

0

0.2

0.4

0.6

0.8

1.0

MP composition

- - - - -

- + - - +

+ - - +

+ + + -

+ - -

+ - - -

- + -

+ - + -

0

0.2

0.4

0.6

0.8

1.0CD4

0

0.2

0.4

0.6

0.8

1.0CD8

MP composition

- - - - -

- + - - +

+ - - +

+ + + -

+ - -

+ - - -

- + -

+ - + -

%of

tota

ltum

orev

ents

Fig. 4 Immune cell population changes in the TME after inducer-loaded MP injections. FACS analysis for immune cell popula-tions found in B16-F10 induced tumors after i.t. treatment with (left to right): nonloaded MP or MP carrying CL264, C5a-pep, mIgG + C5a-pep, LPS, mIgG + LPS, C5a-pep + LPS or

CL264 + C5a-pep + LPS. Nonloaded MP, CL264-MP, LPS–C5a-MP and LPS–C5a–CL264-MP groups consisted of 6–9 mice, and other groups consisted of three mice (Mean ± SD *P < 0.05; **P < 0.01; ***P < 0.001)

1145Cancer Immunol Immunother (2015) 64:1137–1149

1 3

Injections of inducer-loaded MP elevated the levels of the inflammatory cytokines IL-23 and IL-7 quite signifi-cantly (Fig. 5d, e). Both IL-23 and IL-7 showed slightly higher transcription levels in response to LPS combined with C5a-pep versus LPS alone (Fig. 5d, e). Increased transcription was also evident for the chemokine CCL5, though it had relatively high basal transcription levels also in nMP-injected tumors (Fig. 5f). Transcription of CXCL10 in tumors injected with inducer-carrying MP decreased mainly in response to the LPS–C5a-pep combi-nation. Additional tested cytokines or chemokines (IL-2, IL-4, IL-12, IL-13 and IL-17) showed extremely low tran-scription levels, and others (CCL2, CCL3 and CCL4) had high basal transcription levels with no significant changes induced by any MP–inducer combination (data not shown). Altogether, induction of the pro-inflammatory cytokines and chemokines resulted mainly from the presence of LPS on the MP. However, of interest were changes, such as decreased activation of the anti-inflammatory cytokine IL-10, with the use of LPS–C5a-loaded MP.

Influence of i.t. MP administration on tumor growth

The effect of i.t. MP treatments on tumor development was primarily evaluated 48 h post-injections (Supplementary Fig. 3). Tumors injected with LPS-C5a-loaded MP were significantly inhibited and showed promising molecular

and cellular immune activation (Supplementary Fig. 3, Figs. 4, 5). To evaluate the potential effects of LPS-C5a-loaded MP on tumor development, mice were treated with four consecutive injections of MP every 2 or 4 days. Tumor volume measurements showed a delay in tumor growth in mice treated with LPS–C5a-loaded MP starting approxi-mately 4 days after the first administration (Fig. 6a). Tumor volume 11 days after the first treatment was 36 and 24 % smaller when treated every 2 or 4 days, respectively, with LPS–C5a-loaded MP, relative to tumors treated every 2 days with nMP (Fig. 6a). Distinct massive areas of dis-coloration were seen on the excised tumors treated with LPS–C5a-loaded MP under both treatment schedules, as opposed to the uniform black melanin-rich appearance of the B16-F10 tumors (Fig. 6b). This seems to evidence mas-sive infiltration of leukocytes into the tumor mass. Analysis of distinct immune cell populations showed a significantly higher presence of leukocytes (CD45+) in the TME fol-lowing treatments with LPS–C5a-loaded MP as compared to treatment with nMP. Leukocyte infiltration into the TME was predominantly composed of GR-1+ neutrophils (Fig. 6c–e). Mice treated under both treatment schedules with LPS–C5a-loaded MP also exhibited significantly pro-longed survival with the tumors, as compared to control mice treated with either PBS or nMP under the two sched-ules or mice with tumors treated with LPS-loaded MP or IgG–LPS-loaded MP (Fig. 6f).

0.0

0.2

0.4

0.6 *****

*

IL-1

0 (n

orm

aliz

ed)

nMP

LPS-MP

LPS-C5a -MP

C5a-MP + LPS-MP

0.0

0.1

0.2

0.3

0.4

0.5

* ***

IL- 7

(nor

mal

ized

)

*

IL- 6

(nor

mal

ized

)

0.0

0.1

0.2

0.3

0.4

0.0

0.2

0.4

0.6

0.8

1.0 ***

CC

L5 (n

orm

aliz

ed)

0.0

0.2

0.4

0.6

0.8 * IL

- 1β

(nor

mal

ized

)

0.00

0.02

0.04

0.06

0.08

0.10

*

IL-2

3 (n

orm

aliz

ed)

0.0

0.2

0.4

0.6

0.8

1.0 *

CXC

L10

(nor

mal

ized

)

(a) (b) (c)

(f)(e)(d) (g)

Fig. 5 Influence of i.t. MP treatments on cytokine expression. Lev-els of cytokine mRNA were determined by semiquantitative RT-PCR in tumors harvested 48 h after the injection of nonloaded MP (nMP), LPS-MP, LPS–C5a-MP or a 1:1 volumetric mix of LPS-MP

and C5a-MP. The normalized expression of IL-1β (a), IL-6 (b), IL-10 (c), IL-23 (d), IL-7 (e), CCL5 (f) and CXCL10 (g) is presented. Each group consisted of six mice (Mean ± SD *P < 0.05; **P < 0.01; ***P < 0.001)

1146 Cancer Immunol Immunother (2015) 64:1137–1149

1 3

Discussion

The use of specific TAA as targets in cancer immuno-therapy is limited due to tumoral immune escape mecha-nisms as a result of antigen loss [39, 40], down-regula-tion of MHC class I presentation [40–43], suppression of immune effector cell activation by the expression of sur-face immune checkpoint ligands [4], secretion of down-regulatory cytokines [44, 45] and recruitment of cells and molecules that support tumor growth and suppress the immune response [46]. To overcome these numerous obstacles, the immune cells in the TME should be re-directed to produce an antitumor response. Such effects could derive from TLR activation of the TME-residing IIC against the tumor [19]. An advantage of innate activ-ity is that it does not rely on the recognition of TAA to elicit the immune response, thus minimizing the possibil-ity of tumor escape [47].

The basic approach presented in this study was to induce an immunological environment at the tumor site that mimics activation toward pathogen rejection and thereby induce acute inflammatory response and destruc-tive processes in the TME. To test this concept, various combinations of four different ligands of distinct IIC acti-vation pathways were loaded on MP, which served as an optimal platform for their delivery in numerous possible combinations [48].

MP carrying multiple copies of ligands combinations were shown to promote multiple signal transduction events, mediated by specific receptor cross-linking [49]. Consid-erations in choosing MP size, shape and composition were previously discussed [34]. In this study, single ligands or their different combinations were attached to ~1.5-µm iron oxide MP. The ligand’s ability to activate an immune response in vitro after biotin and MP conjugation was retained.

0 1 2 3 40.0

0.5

1.0

1.5

P <0.0001r =0.9263

%CD45 (of Total events)

%G

R-1

(of t

otal

eve

nts)

nMP LPS-C5a-MP

Every 2 days Every 4 days

nMP every 2 daysLPS-C5a-MP every 2 daysLPS-C5a-MP every 4 days

%G

R-1

+ (of

Tot

al e

vent

s)

%C

D45

+ (of

Tot

al e

vent

s)

0 5 10 15 200

50

100

****

**

4SBP daysnMP 2 daysnMP 4 daysLPS-MP 4daysIgG-LPS-MP 4daysLPS-C5a-MP 2daysLPS-C5a-MP 4days

Days post first injectionSu

rviv

al(%

)

(a)

(d) (e) (f)

(b) (c)

Fig. 6 Effect of MP treatments on tumor growth rate. B16-F10 tumors induced s.c. were treated with four injections every 2 days of nonloaded MP (nMP) or LPS–C5a-MP, or four injections every 4 days of PBS, nMP, LPS-MP, IgG–LPS-MP or LPS–C5a-MP. (a–e) Comparison of i.t. treatments of nMP with LPS–C5a-MP. (a) Tumor volume measurement calculated over 11 days. (b) Tumors har-vested after reaching 500 mm3 (arrows demonstrate extensive white regions of the tumor mass). (c) Correlation analysis demonstrating relationships between CD45+ and GR-1+ granulocyte, mainly neu-trophil, populations in tumors injected with LPS–C5a-MP. (d and e)

Flow cytometry results showing immune population analysis of (d) %GR-1+ and (e) % of total CD45+ leukocytes population in tumors (Mean ± SD **P < 0.01). (f) Kaplan–Meier analysis of tested mice induced with B16-F10 tumors and treated i.t. with MP. % Survival is defined as percentage of mice within a group with tumor volume smaller than 500 mm3. Days were counted from first MP administra-tion (at tumor volume of 50 mm3). The nMP (2 days) and LPS–C5a-MP groups consisted of eight mice, and other groups consisted of 3–4 mice (*P < 0.05; ***P < 0.001)

1147Cancer Immunol Immunother (2015) 64:1137–1149

1 3

Two of the four ligands tested in this study, LPS and CL264, which activate TLR4 and TLR7, respectively, both recognize pathogen-associated molecular patterns [30, 50, 51]. The third ligand tested, IgG, is recognized by FcγR which may induce phagocytosis, antibody-dependent cell cytotoxicity or cytokine secretion [31]. The fourth ligand tested, a synthetic C5a-pep, activates C5a receptor, which may induce secretion of reactive oxygen species (ROS), as well as chemotaxis and proliferation of immune cells. Notably, C5aR can activate the NF-κB signal pathway in monocytes or suppress this pathway in neutrophils [52, 53].

Evidence for the reciprocal influence of these receptors activation, e.g., C5aR activation that regulates differential expression of FcγR members [54], suggests that ligand combinations may impact overall cellular activity.

In this study, immune activation by sole ligands, or their combinations, loaded on MP was tested in vitro and in vivo. In vitro induction of cytokine expression patterns was tested both in cultured cell lines and in cultured iso-lated splenocytes. LPS-loaded MP induced acute pro-inflammatory cytokines and chemokines expression (e.g., IL-1β and TNF-α) accompanied by elevated expression of the anti-inflammatory cytokine IL-10, which is considered to play a major role in cancer evasion from immune-based therapies [6]. However, MP co-loaded with C5a and LPS induced expression of inflammatory cytokines, identical to that promoted by LPS alone, but this combination blocked the expression of the anti-inflammatory IL-10.

The effect of i.t. administration of ligands bound MP on innate immune activation in vivo in the TME of melanoma was tested 48 h after treatment. Cell populations, as well as cytokine and chemokine expression patterns, were deter-mined in parallel to the short-term follow-up of the tumor growth. A prominent phenomenon observed was the change from positive correlation between macrophages and neutro-phils to a negative one in the treated tumors. The expres-sion of the pro-inflammatory cytokines and chemokines, accompanied by an increase in CD45+ immune cell popu-lations, mainly composed of macrophages and neutrophils, was mostly influenced by the presence of LPS on the MP. Other combinations of inducers loaded on MP such as C5a solely did not influence these immune parameters.

Elevated levels of IL-23 and IL-7, both inhibitors of the regulatory T cell activation [55, 56], following i.t. treatment with MP loaded with LPS and C5a, may have contributed to the decrease in IL-10 levels, though barely detectable transcription levels were assayed in vitro (data not shown). This may imply that other mechanisms are also involved in this response.

The high expression levels of the chemokine CXCL10, as reported in numerous cancer types (e.g., melanoma, ovarian carcinoma and multiple myeloma) [57], which we observed in the tumors from the negative control of

nMP-treated mice, were significantly reduced following treatment with LPS–C5a-loaded MP.

The combined activation of TLR4 and C5aR seemed to induce a large-scale inflammatory response in the TME, as seen by a sole activation of TLR4’s, while repressing paral-lel induction of the tumor-beneficial anti-inflammatory reg-ulating molecules, and possibly, tissue-repair factors (e.g., VEGF). Dual receptor induction, resulting in the repression of the anti-inflammatory cytokine IL-10, while maintaining all other inflammatory features, implies redirection toward a “classical” immune activation, capable of tumor eradica-tion. Notably, activation of these receptors in the TME, fol-lowing injection of a mixture of MP, each bound to a dif-ferent ligand, was associated with elevation of IL-10 levels, indicating the importance of co-binding of these ligands to the same MP. Additionally, short-term tumor volume meas-urements indicated inhibition of tumors treated with LPS-C5a-loaded MP, which was not seen following treatment with the mixture of MP each bound to a different single ligand. Thus, simultaneous activation of the same leuko-cyte with both ligands seems to have an advantage over the effect of mixed MP, each bound to separate inducers.

The unique pattern of cytokine expression following repeated i.t. treatments with LPS–C5a-loaded MP, in par-ticular the reduction in IL-10, was accompanied by the observation of discolored sections on the excised tumors. The loss of the tumor color, possibly as result of reduc-tion in melanin levels, probably due to massive infiltra-tion of leukocytes, may indicate inflamed or necrotic areas in the TME. These observations are supported by increased CD45+ cell populations, predominantly the GR-1+ neutrophils, which were detected in these tumors’ microenvironments.

The described changes in cellular and molecular con-tents in the TME following LPS–C5a-loaded MP treat-ments resulted in significant inhibition of tumor growth as compared to two LPS-MP treatments and controls (PBS or nMP). This delay in tumor progression following LPS–C5a-loaded MP treatment, accompanied by the induc-tion of immune activation, may be explained by the inhi-bition of the anti-inflammatory agents (e.g., IL-10) effect on activated cells, disallowing their “misdirection” toward immune tolerance. Other pro-tumoral factors, such as VEGF and TGF-β, may also be down-regulated and should be explored.

Our results show the possibility of changing tumor–immune system homeostasis in the TME and influenc-ing the composition and activation of residing immune cell populations. The “polarization” of tumor-residing IIC toward a “classical” pathogen-aggressive response may be accomplished by appropriately activating the inflamma-tory response while restraining the parallel tissue-repair activity of the activated cells. The possibility of this effect

1148 Cancer Immunol Immunother (2015) 64:1137–1149

1 3

expanding to the activation of IIC in distal metastases deserves further investigation.

Our results support accumulated evidence suggesting that treating tumors with a single IIC-activating agent, e.g., TLR ligand, followed by an inflammatory response is not sufficient for successful immunotherapy. On the other hand, activation of IIC toward a tumoricidal immune response may be achieved by combining ligands originat-ing from several innate activation routes (e.g., complement receptor together with TLR activation). Such combinations displayed enhanced efficacy or an altered activation profile as compared to single ligand activity [33–35, 58–60].

In conclusion, our study suggests that cells exist in the TME with the ability to eliminate tumors needs to be re-directed toward anti-tumoral responses. The current study shows the feasibility of attracting immune cells and induc-ing an acute inflammatory response in the TME, while suppressing the pro-tumoral activity of the IIC, by using a combination of different inducers ligands carried on the same MP.

Conflict of interest The authors declare that they have no conflict of interest in publishing this research.

References

1. Blattman JN, Greenberg PD (2004) Cancer immunotherapy: a treatment for the masses. Science 305:200–205. doi:10.1126/science.1100369

2. Page DB, Naidoo J, McArthur HL (2014) Emerging immuno-therapy strategies in breast cancer. Immunotherapy 6:195–209. doi:10.2217/imt.13.166

3. Chaudhuri D, Suriano R, Mittelman A, Tiwari RK (2009) Tar-geting the immune system in cancer. Curr Pharm Biotechnol 10:166–184

4. Pardoll DM (2012) The blockade of immune checkpoints in can-cer immunotherapy. Nat Rev Cancer 12:252–264. doi:10.1038/nrc3239

5. Grivennikov SI, Greten FR, Karin M (2010) Immunity, inflammation, and cancer. Cell 140:883–899. doi:10.1016/j.cell.2010.01.025

6. Real LM, Jimenez P, Kirkin A et al (2001) Multiple mechanisms of immune evasion can coexist in melanoma tumor cell lines derived from the same patient. Cancer Immunol Immunother 49:621–628

7. Mantovani A, Sozzani S, Locati M et al (2002) Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23:549–555

8. De Visser KE, Eichten A, Coussens LM (2006) Paradoxical roles of the immune system during cancer development. Nat Rev Can-cer 6:24–37. doi:10.1038/nrc1782

9. Mantovani A, Allavena P, Sica A, Balkwill F (2008) Cancer-related inflammation. Nature 454:436–444. doi:10.1038/nature07205

10. Mantovani A, Cassatella MA, Costantini C, Jaillon S (2011) Neutrophils in the activation and regulation of innate and adap-tive immunity. Nat Rev Immunol 11:519–531. doi:10.1038/nri3024

11. Galdiero MR, Bonavita E, Barajon I et al (2013) Tumor asso-ciated macrophages and neutrophils in cancer. Immunobiology 218:1402–1410. doi:10.1016/j.imbio.2013.06.003

12. Lewis CE, Pollard JW (2006) Distinct role of macrophages in different tumor microenvironments. Cancer Res 66:605–612. doi:10.1158/0008-5472.CAN-05-4005

13. Junttila MR, de Sauvage FJ (2013) Influence of tumour micro-environment heterogeneity on therapeutic response. Nature 501:346–354. doi:10.1038/nature12626

14. Hanahan D, Coussens LM (2012) Accessories to the crime: func-tions of cells recruited to the tumor microenvironment. Cancer Cell 21:309–322. doi:10.1016/j.ccr.2012.02.022

15. Hagemann T, Lawrence T, McNeish I et al (2008) “Re-educat-ing” tumor-associated macrophages by targeting NF-kappaB. J Exp Med 205:1261–1268. doi:10.1084/jem.20080108

16. Mittendorf EA, Alatrash G, Qiao N et al (2012) Breast cancer cell uptake of the inflammatory mediator neutrophil elastase triggers an anticancer adaptive immune response. Cancer Res 72:3153–3162. doi:10.1158/0008-5472.CAN-11-4135

17. Mantovani A, Sica A, Sozzani S et al (2004) The chemokine sys-tem in diverse forms of macrophage activation and polarization. Trends Immunol 25:677–686. doi:10.1016/j.it.2004.09.015

18. Gay NJ, Gangloff M (2007) Structure and function of Toll receptors and their ligands. Annu Rev Biochem 76:141–165. doi:10.1146/annurev.biochem.76.060305.151318

19. Pradere J-P, Dapito DH, Schwabe RF (2014) The Yin and Yang of toll-like receptors in cancer. Oncogene 33:3485–3495. doi:10.1038/onc.2013.302

20. Salaun B, Coste I, Rissoan M-C et al (2006) TLR3 can directly trigger apoptosis in human cancer cells. J Immunol 176:4894–4901

21. Krieg AM (2004) Antitumor applications of stimulating toll-like receptor 9 with CpG oligodeoxynucleotides. Curr Oncol Rep 6:88–95

22. Carpentier AF, Chen L, Maltonti F, Delattre JY (1999) Oligode-oxynucleotides containing CpG motifs can induce rejection of a neuroblastoma in mice. Cancer Res 59:5429–5432

23. Heil F, Hemmi H, Hochrein H et al (2004) Species-specific rec-ognition of single-stranded RNA via toll-like receptor 7 and 8. Science 303:1526–1529. doi:10.1126/science.1093620

24. Diebold SS, Kaisho T, Hemmi H et al (2004) Innate antivi-ral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303:1529–1531. doi:10.1126/science.1093616

25. Scheel B, Aulwurm S, Probst J et al (2006) Therapeutic anti-tumor immunity triggered by injections of immunostimulat-ing single-stranded RNA. Eur J Immunol 36:2807–2816. doi:10.1002/eji.200635910

26. Sfondrini L, Rossini A, Besusso D et al (2006) Antitu-mor activity of the TLR-5 ligand flagellin in mouse mod-els of cancer. J Immunol 176:6624–6630. doi:10.4049/jimmunol.176.11.6624

27. El Andaloussi A, Sonabend AM, Han Y, Lesniak MS (2006) Stimulation of TLR9 with CpG ODN enhances apoptosis of gli-oma and prolongs the survival of mice with experimental brain tumors. Glia 54:526–535. doi:10.1002/glia.20401

28. Shime H, Matsumoto M, Oshiumi H et al (2012) Toll-like receptor 3 signaling converts tumor-supporting myeloid cells to tumoricidal effectors. Proc Natl Acad Sci USA 109:2066–2071. doi:10.1073/pnas.1113099109

29. Peng G, Guo Z, Kiniwa Y et al (2005) Toll-like receptor 8-medi-ated reversal of CD4+ regulatory T cell function. Science 309:1380–1384. doi:10.1126/science.1113401

30. Wang R-F, Miyahara Y, Wang HY (2008) Toll-like receptors and immune regulation: implications for cancer therapy. Oncogene 27:181–189. doi:10.1038/sj.onc.1210906

1149Cancer Immunol Immunother (2015) 64:1137–1149

1 3

31. Cassard L, Cohen-Solal J, Camilleri-Broët S et al (2006) Fc gamma receptors and cancer. Springer Semin Immunopathol 28:321–328. doi:10.1007/s00281-006-0058-8

32. Sayegh ET, Bloch O, Parsa AT (2014) Complement anaphyla-toxins as immune regulators in cancer. Cancer Med 3:747–758. doi:10.1002/cam4.241

33. Rudilla F, Fayolle C, Casares N et al (2012) Combination of a TLR4 ligand and anaphylatoxin C5a for the induction of anti-gen-specific cytotoxic T cell responses. Vaccine 30:2848–2858. doi:10.1016/j.vaccine.2012.02.052

34. Shahar E, Gorodetsky R, Gaberman E et al (2010) Targeted microbeads for attraction and induction of specific innate immune response in the tumor microenvironment. Vaccine 28:7279–7287. doi:10.1016/j.vaccine.2010.08.083

35. Stier S, Maletzki C, Klier U, Linnebacher M (2013) Com-binations of TLR ligands: a promising approach in can-cer immunotherapy. Clin Dev Immunol 2013:271246. doi:10.1155/2013/271246

36. Kaisho T, Akira S (2006) Toll-like receptor function and sign-aling. J Allergy Clin Immunol 117:979–987. doi:10.1016/j.jaci.2006.02.023

37. Odeyale CO, Kang YH (1988) Biotinylation of bacterial lipopol-ysaccharide and its applications to electron microscopy. J Histo-chem Cytochem 36:1131–1137. doi:10.1177/36.9.3136207

38. Konteatis ZD, Siciliano SJ, Van Riper G et al (1994) Develop-ment of C5a receptor antagonists. Differential loss of functional responses. J Immunol 153:4200–4205

39. Jäger E, Ringhoffer M, Altmannsberger M et al (1997) Immu-noselection in vivo: independent loss of MHC class I and mel-anocyte differentiation antigen expression in metastatic mela-noma. Int J Cancer 71:142–147

40. Ikeda H, Lethé B, Lehmann F et al (1997) Characterization of an antigen that is recognized on a melanoma showing partial HLA loss by CTL expressing an NK inhibitory receptor. Immunity 6:199–208

41. Del Campo AB, Kyte JA, Carretero J et al (2014) Immune escape of cancer cells with beta2-microglobulin loss over the course of metastatic melanoma. Int J Cancer 134:102–113. doi:10.1002/ijc.28338

42. Garrido F, Ruiz-Cabello F, Cabrera T et al (1997) Implications for immunosurveillance of altered HLA class I phenotypes in human tumours. Immunol Today 18:89–95

43. Lehmann F, Marchand M, Hainaut P et al (1995) Differences in the antigens recognized by cytolytic T cells on two succes-sive metastases of a melanoma patient are consistent with immune selection. Eur J Immunol 25:340–347. doi:10.1002/eji.1830250206

44. Smith DR, Kunkel SL, Burdick MD et al (1994) Production of interleukin-10 by human bronchogenic carcinoma. Am J Pathol 145:18–25

45. Huber D, Philipp J, Fontana A (1992) Protease inhibitors inter-fere with the transforming growth factor-beta-dependent but not the transforming growth factor-beta-independent path-way of tumor cell-mediated immunosuppression. J Immunol 148:277–284

46. Shimoda M, Mellody KT, Orimo A (2010) Carcinoma-asso-ciated fibroblasts are a rate-limiting determinant for tumour

progression. Semin Cell Dev Biol 21:19–25. doi:10.1016/j.semcdb.2009.10.002

47. Khong HT, Restifo NP (2002) Natural selection of tumor vari-ants in the generation of “tumor escape” phenotypes. Nat Immu-nol 3:999–1005. doi:10.1038/ni1102-999

48. O’Hagan DT, Singh M, Ulmer JB (2006) Microparticle-based technologies for vaccines. Methods 40:10–19. doi:10.1016/j.ymeth.2006.05.017

49. Bryceson YT, March ME, Ljunggren H-G, Long EO (2006) Syn-ergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion. Blood 107:159–166. doi:10.1182/blood-2005-04-1351

50. Weighardt H, Jusek G, Mages J et al (2004) Identification of a TLR4- and TRIF-dependent activation program of dendritic cells. Eur J Immunol 34:558–564. doi:10.1002/eji.200324714

51. Hemmi H, Kaisho T, Takeuchi O et al (2002) Small anti-viral compounds activate immune cells via the TLR7 MyD88-depend-ent signaling pathway. Nat Immunol 3:196–200. doi:10.1038/ni758

52. Guo R-F, Riedemann NC, Ward PA (2004) Role of C5a–C5aR interaction in sepsis. Shock 21:1–7. doi:10.1097/01.shk.0000105502.75189.5e

53. Lee H, Whitfeld PL, Mackay CR (2008) Receptors for com-plement C5a. The importance of C5aR and the enigmatic role of C5L2. Immunol Cell Biol 86:153–160. doi:10.1038/sj.icb.7100166

54. Ricklin D, Hajishengallis G, Yang K, Lambris JD (2010) Com-plement: a key system for immune surveillance and homeostasis. Nat Immunol 11:785–797. doi:10.1038/ni.1923

55. Rosenberg SA, Sportès C, Ahmadzadeh M et al (2006) IL-7 administration to humans leads to expansion of CD8+ and CD4+ cells but a relative decrease of CD4+ T-regulatory cells. J Immunother 29:313–319. doi:10.1097/01.cji.0000210386.55951.c2

56. Petermann F, Rothhammer V, Claussen MC et al (2010) γδ T cells enhance autoimmunity by restraining regulatory T cell responses via an interleukin-23-dependent mechanism. Immu-nity 33:351–363. doi:10.1016/j.immuni.2010.08.013

57. Liu M, Guo S, Stiles JK (2011) The emerging role of CXCL10 in cancer (Review). Oncol Lett 2:583–589. doi:10.3892/ol.2011.300

58. Silva A, Mount A, Krstevska K et al (2015) The combination of ISCOMATRIX adjuvant and TLR agonists induces regression of established solid tumors in vivo. J Immunol 194:2199–2207. doi:10.4049/jimmunol.1402228

59. Ahonen CL, Wasiuk A, Fuse S et al (2008) Enhanced efficacy and reduced toxicity of multifactorial adjuvants compared with unitary adjuvants as cancer vaccines. Blood 111:3116–3125. doi:10.1182/blood-2007-09-114371

60. Wells JW, Cowled CJ, Farzaneh F, Noble A (2008) Combined triggering of dendritic cell receptors results in synergistic activa-tion and potent cytotoxic immunity. J Immunol 181:3422–3431. doi:10.4049/jimmunol.181.5.3422