White Paper Syngeneic mouse models as a tool to study...
Transcript of White Paper Syngeneic mouse models as a tool to study...
White Paper
Syngeneic mouse models as a tool to study immune modulatory effects of cancer therapeutics
Immuno-Oncology Platform
Flow CytometryCell SortingImmuno-HistochemistryTarget Validation
Subcutaneous ModelsOrthotopic ModelsMetastasis Models
The Solution Provider for Drug Discovery in Oncology
Recombinant Proteins Biochemical Assay Services Cellular Assay Services In Vivo Testing Services
www.proqinase.com 1
Imm
uno-Oncology P
latformIn Vivo Testing Services
T A R G E T I N G C A N C E R
IntroductionIn a healthy body, the immune system puts consistent pressure on emerging tumor cells by efficiently elimina-ting cells that acquire malignant properties. However, if tumor cells defeat the body’s immune surveillance, they still face persistent immune pressure, which they oppose by immune evasion or even active immune suppression. Cancer treatment disrupts the pre-existing equilibrium between immune pressure and immune evasion in a patient’s tumor – irrespective of the choice of therapy. It has become increasingly evident that the success of many conventional drugs relies on their side effects to activate immune responses for tumor tissue eradication (L. Gal-luzzi et al. 2012). In contrast, immune therapeutics are designed to support the immune system in its struggle to eliminate tumor cells by breaking tolerance or conver-ting the tumor‘s environment from immune-suppressive to immune-activating. To evaluate potential treatments, drug testing in animals with a functional immune system is crucial and mandatory to meet current drug develop-ment standards (JP Hegmans et al. 2014).In this white paper we will introduce four syngeneic tu-mor models to evaluate the immune modulatory capa-cities of new cancer treatments. As an example, we will demonstrate how treatment against immune checkpoint
PD-1/PD-L1 (programmed cell death protein 1/ligand 1) impacts tumor growth and alters immune cell populati-ons.
Flow cytometry analysis of syngeneic tumor modelsThe four syngeneic models characterized in this white pa-per are the 4T1 breast tumor model, the B16-F10 melanoma model, the CT26.wt colorectal tumor model, and the RENCA kidney tumor model.
All models except B16-F10 were implan-ted subcutaneously into the flank of the mice. B16-F10 cells were inserted orthotopically into the der-mis of the mice. Mice were sacrificed approximately three weeks after cell implantation, the solid tumors were isola-ted and dissociated into single cell suspensions, and ana-
Syngeneic mouse models as a tool to study immune modulatory effects of cancer therapeutics
Every cancer treatment has the potential to induce a stimulatory or inhibitory effect on the immune response to a tumor. Scientific knowledge about the significance of the immune system for tumor era-dication during conventional treatment is growing quickly, and an increasing number of immune-modu-lating drugs are entering clinical trials for cancer treatment. This necessitates investigating these drugs in the presence of an intact immune system, and syngeneic tumor models are the ideal tool to achieve this. In addition to our many xenograft mouse models, we established several syngeneic tumor models, which we thoroughly characterized with respect to immune phenotyping and response to immune checkpoint inhibition. Four of them are introduced in this white paper.
Cell line Origin Subcu-taneous
Ortho-topic
Meta-stasis Model
4T1 breast ✓ ✓ ✓*
B16-F10 melanoma ✓ ✓ ✓**
RENCA renal ✓ ✓ -
CT26.wt colorectal ✓ ✓ -
MC38-CEA# colorectal ✓ - -
Clone M3 melanoma ✓ - -
Lewis lung LL/2 lung ✓ - -
*orthotopic implantation **intravenous injection#Carcinoembryonic antigen suited for investigation of tumor-specific T cell response
Syngeneic tumor models available at ProQinase (August 2015)
www.proqinase.com 2
Imm
uno-Oncology P
latformIn Vivo Testing Services
T A R G E T I N G C A N C E R
lysis of T cells, macrophages, and myeloid cell subsets was performed using flow cytometry (Figure 1). The goals of immune phenotyping are to gain information about
a) which immune cells are present in the tumors, b) how many of those immune cells are present, and c) whether the amount of the immune cells correlates with the size of the tumors.
Our four tumor models differed in the total numbers of tumor-infiltrating immune cells (leukocyte populations)
as well as the composition of individual immune cell subsets (Table 1). For example, 4T1 tumors showed the
highest amount of infiltrating leukocytes and the immu-ne cells were mainly PMN-MDSCs and neutrophils (ple-ase see the insert for information about the immune cell subsets). B16-F10 and CT26.wt tumors on the other hand were very poorly infiltrated by leukocytes. All tumors had a similar relative amount of M1 and M2 macrophages. The correlation between intratumoral immune cells and tumor size gives us valuable information about whether
Figure 1. Gating strategy for flow cytometric analysis of tumor-infiltrating immune cell subsets. (A) Immune cell populations in CT26.wt tumors were investigated three weeks after tumor implantation in the flank of the mouse. Tumors were isolated and dissociated into single-cell suspensions. Cells of one tumor that were stained with three antibody panels and analysed by flow cytometry are shown. (B) For comparison, a picture of a 4T1 tumor is shown with a substantial amount of infiltrating PMN-MDSCs.
4T1
Ly6G
Ly6C
PMN-MDSC
M-MDSC
Neutrophils
Macrophage antibody panel
CD
206
(MM
R)
SS
C
CD45
CD
11b
F4/80 MHC II
M2
M1
Myeloid-derived suppressor cell (MDSC) antibody panel
PMN-MDSC
Neutrophils
M-MDSC
SS
C
CD45
SS
C
CD11b
Ly6G
Ly6C
T lymphocyte antibody panel
SS
C
FSC
CD
45
CD3
CD
4
CD8
Regulatory T cells
CD
25
FoxP3
CD
44
CD25
CD
44
CD25
A
B
www.proqinase.com 3
Imm
uno-Oncology P
latformIn Vivo Testing Services
T A R G E T I N G C A N C E R
a particular subset holds an immune-activating (tumor growth suppressing) or immune-suppressive (tumor growth supporting) role. Interestingly, the frequency of some immune cell subsets did indeed significantly corre-late with tumor volume (Table 2). For example, we can assume that the growth of CT26.wt tumors is dependent on the presence and absence of M1 and M2 macropha-ges. Furthermore, B16-F10 tumors seem to rely on the presence of activated CD8+ T cells to overcome the sup-pressive function of regulatory T cells. Although B16-F10 tumors were infiltrated by M-MDSCs with the highest fre-quency, the correlation data suggest no important role of these cells for tumor growth.
Immune-checkpoint blockade via anti-PD-L1 antibodyA very attractive target for immune therapy against can-cer is the immune checkpoint PD-1/PD-L1 since it is a me-chanism tumor cells employ to suppress the activity and proliferation of tumor-infiltrating immune cells (A. Buqué et al. 2015). In addition, targeting this pathway interferes
with communication between immune cells in lymphatic tissues and “takes the brake off” immu-ne-activating processes. All four tumor models were tes-ted with an anti-PD-L1 antibody treatment. Of all tumors, only CT26.wt tumors responded to therapy with significant tumor growth inhibition (Figure 2). We evaluated the tumor infiltrating immune cells of treated and un-treated mice via flow cytomet-ry at the endpoint of the study, which was three weeks after tu-mor implantation.
Impact of anti-PD-L1 treat-ment on macrophages As an example of examining the mechanism of action of a drug based on flow cytometry data, we show the analysis of the im-pact of PD-L1 blockade on the different immune cell subsets in the CT26.wt model. Investigati-on of the macrophage subsets shows that tumors in vehicle-treated mice contained more M2 than M1 macrophages, helping these tumors uphold an immu-ne-suppressive environment (Figure 3A). During anti-PD-L1 treatment, however, M1 cells proliferated and grew to exceed the amount of M2 cells, which
suggests that anti-PD-L1 reprograms the polarization and proliferation of macrophages (Figure 3B). As more immu-ne-activating M1 macrophages are present in the tumors, there is likely increased interference with the immune-suppressive environment, due to cytokine secretion (D.I. Gabrilovich et al. 2012). Furthermore, M1 macrophages
Syngeneic Model 4T1 B16-F10 CT26.wt RENCA
Leukocytes (number of cells per mm3 tumor tissue) 27191 9752 13596 27869
Table 1. Frequencies of tumor-infiltrating leukocyte populations.(A) Leukocytes comprise the absolute amount of CD45+ cells in the tumors. (B) The individual immune cell subsets are shown relative to each other with font sizes reflecting their frequencies.
4T1 B16-F10 CT26.wt RENCA
M1 macrophages M1 macrophages M1 macrophages M1 macrophages
M2 macrophages M2 macrophages M2 macrophages M2 macrophages
M-MDSC M-MDSC M-MDSC M-MDSC
PMN-MDSC PMN-MDSC PMN-MDSC PMN-MDSC
Neutrophils Neutrophils Neutrophils Neutrophils
CD8+ T cells CD8+ T cells CD8+ T cells CD8+ T cells
CD8+CD44+ T cells CD8+CD44+ T cells CD8+CD44+ T cells CD8+CD44+ T cells
CD8+CD44+CD25+ T cells CD8+CD44+CD25+ T cells CD8+CD44+CD25+ T cells CD8+CD44+CD25+ T cells
CD4+ T cells CD4+ T cells CD4+ T cells CD4+ T cells
CD4+CD44+ T cells CD4+CD44+ T cells CD4+CD44+ T cells CD4+CD44+ T cells
Regulatory T cells Regulatory T cells Regulatory T cells Regulatory T cells
A
B
www.proqinase.com 4
Imm
uno-Oncology P
latformIn Vivo Testing Services
T A R G E T I N G C A N C E R
clear tumor cells, thereby helping decrease the size of the tumor. The importance of the M2/M1 ratio is demonst-rated by the significant correlations to the tumor volume (Figure 3C).
Impact of anti-PD-L1 treatment on MDSCsM-MDSCs were the dominant myeloid subset found in the CT26.wt tumors (Figure 4A). The number of M-MDSCs in-
creased in response to PD-L1 treatment. This subset was more commonly present in small tumors, both with and without anti-PD-L1 treatment (not shown). PMN-MDSCs, on the other hand, were present in larger tumors, sugges-ting a role to support tumor growth, although only very few PMN-MDSCs were present in CT26.wt tumors. Inte-restingly, under PD-L1 blockade, this effect disappeared and the PMN-MDSC counts no longer correlated with
Figure 2. Impact of anti-PD-L1 antibody treatment on growth of syngeneic tumors after subcutaneous (4T1, CT26.wt, RENCA) or orthotopic (B16-F10) implantation. PD-L1 blockade alone rarely led to tumor growth inhibition. Only CT26.wt tumors responded significantly with decreased growth. Arrows indicate time points of intraperitoneal injections of 10mg/kg anti-mouse PD-L1 antibody (BioXCell) into 8 mice per group. * p=0.02 (Mann-Whitney test)
days after tumor implantation
tum
or v
olum
e (m
m3 )
0 2 4 6 8 10 12 14 16 180
200
400
6004T1
days after tumor implantation
tum
or v
olum
e [m
m3 ] B16-F10
days after tumor implantation
tum
or v
olum
e (m
m3 )
0 2 4 6 8 10 12 14 16 18 20 220
500
1000
1500
RENCA
days after tumor implantation
tum
or v
olum
e (m
m3 )
0 2 4 6 8 10 12 14 16 18 20 22 240
200
400
600
800
CT26.wt
days after tumor implantation
tum
or v
olum
e (m
m3 )
0 2 4 6 8 10 12 14 16 18 200
500
1000
1500
*
anti-PD-L1vehicle
Syngeneic Model
Immune Cell Subset
4T1 B16-F10 CT26.wt RENCA
Total Leukocytes -0.2 0 -0.5 0
M2/M1 ratio 0.1 0.1 0.8 0.5
M1 Macrophages 0.1 -0.6 -0.6 -0.3
M2 Macrophages 0 -0.1 0.3 -0.1
M-MDSC 0 -0.2 -0.2 -0.1
PMN-MDSC 0.2 few cells few cells 0
Neutrophils 0.3 -0.1 0 0.1
CD8+CD44+ T cells 0 0 -0.2 0.1
CD8+CD44+CD25+ T cells 0 0.4 -0.8 0.1
CD4+CD44+ T cells -0.7 0 -0.8 -0.7
Regulatory T cells -0.6 0 -0.7 -0.7
CD8+CD44+ / Regulatory T cell ratio
0.4 0.7 0.4 -0.1
CD8+CD44+CD25+ / Regulatory T cell ratio
0 0.7 0 0.3
CD4+CD44+ / Regulatory T cell ratio
0.4 0.1 -0.1 -0.1
Table 2. Correlation analysis between tumor volume and immune cell subset frequencies. For each tumor, the number of a certain immune cell subset was correlated with the size of the tumor, indicating activating or suppressing roles of the individual immune cell subsets. Negative R²-values (-1 to 0) represent negative correlations, e.g. high numbers of immune cells found mainly in small tumors. Positive R²-values (0 to 1) represent positive correlations, e.g. high numbers of immune cells found mainly in large tumors. Statistically significant values are shown in bolded red font (students t test).
www.proqinase.com 5
Imm
uno-Oncology P
latformIn Vivo Testing Services
T A R G E T I N G C A N C E R
ratio
M2/
M1
mac
roph
ages
vehicle anti-PD-L1
0
1
2
3
4
5
6
ratio
M2/
M1
mac
roph
ages
vehicle anti-PD-L1
B
# ce
lls p
er
1 M
io le
ukoc
ytes
M1 M2macrophages
A
# ce
llspe
r 1 M
io. l
euko
cyte
s
M1 M20
50000
100000
150000
200000
macrophages
vehicle
anti-PD-L1
Figure 3. Mechanism of action analysis of anti-PD-L1 antibody based on the macrophages subsets infiltrating CT26.wt tumors. (A) Anti-PD-L1 treatment increased the proliferation of “good” M1 macrophages and suppressed the proliferation of “bad” M2 macrophages. (B) The ratio of M2 to M1 macrophages switched from greater than 1 to less than 1 after PD-L1 blockade (students t test). (C) The ratio of M2 to M1 macrophages negatively correlated with and without anti-PD-L1 treatment, which suggests no impact of the PD-L1 blockade on macrophage function. * p≤0.05; ** p≤0.01
tum
or v
olum
e [m
m3 ]
M2/M1 macrophage ratio
tum
or
volu
me
[mm
³]
0 1 2 3 4 50
500
1000
1500
2000
2500
3000
R2=0.8 **
M2/M1 macrophage ratio
tum
or
volu
me
[mm
³]
0.0 0.5 1.0 1.5 2.00
500
1000
R2=0.7 *
tum
or v
olum
e [m
m3 ]
M2/M1 macrophage ratio
C
M2/M1 macrophage ratio
tum
or
volu
me
[mm
³]
0 1 2 3 4 50
500
1000
1500
2000
2500
3000
R2=0.8 **
M2/M1 macrophage ratio
tum
or
volu
me
[mm
³]
0.0 0.5 1.0 1.5 2.00
500
1000
R2=0.7 *
M2/M1 macrophage ratio
tumor growth (Figure 4B). Neutrophils proliferated under PD-L1 treatment, but this increase was not significant. A trend emerged following PD-L1 treatment, which sug-gests a detrimental role of neutrophils on tumor growth (not shown) indicating that PD-L1 has to be blocked for neutrophils to effectively kill tumor cells.
Impact of anti-PD-L1 treatment on T lymphocytesInvestigation of the T lymphocyte compartment showed a significant increase in the number of all T cell subsets with anti-PD-L1 treatment (Figure 5A). There was a negative correlation between activated CD4+ and CD8+ T cells and tumor growth, suggesting that the effector T cells were actively fighting the tumor (Table 2). However, the same result applied to regulatory T cells, which belong to the
immune-suppressive subsets. PD-1 is highly expressed on regulatory T cells. A blockade of this pathway was shown to interfere with their proliferation and to inhibit their im-mune-suppressive capacities (L.M. Francisco et al. 2009). In accordance with these data, PD-L1 blockade in CT26.wt tumors seems to diminish the suppressive nature of regu-latory T cells, as demonstrated by the complete shift in cor-relation between tumor volume and the ratio of activated CD8+ T cells to regulatory T cells (Figure 5B). Taken together, we can nicely follow the action of anti-PD-L1 treatment using flow cytometric analysis with our three antibody pa-nels to detect macrophages, MDSCs and T cells. Our findings included repolarization of M2 to M1 macro-phages, a loss of function of the PMN-MDSC subset, and proliferation and gain of function of CD8+ effector T cells.
www.proqinase.com 6
Imm
uno-Oncology P
latformIn Vivo Testing Services
T A R G E T I N G C A N C E R
We conclude that PD-L1 blockade works on different im-mune cell subsets to enable the immune system to effec-tively kill tumor cells and suppress tumor growth in the CT26.wt model.
SummaryIn this white paper we used anti-PD-L1 treatment as an example to demonstrate that proper flow cytometry analysis is a valuable tool for mechanism of action analy-sis. The characterization and validation of our syngeneic tumor models helps our customers to choose the right model for their drug development approach. We provide flow cytometry data and analysis for all of our models, which will be made available upon request, and we con-stantly expand the number of available cell lines and im-mune cell markers.
References1. Galluzzi L et al.: The secret ally: immunostimulation by anticancer drugs; Nat Rev Drug Discov; 2012; PMID: 223017982. Hegmans JP et al.: Immunomodulation in Cancer; Curr Opin Pharmacol; 2014; PMID: 250111123. Buqué A et al.: Trial Watch: Immunomodulatory mono-clonal antibodies for oncologcial indications; Oncoimmu-nology; 2015; PMID: 261374034. Gabrilovich DI et al.: Coordinated regulation of mye-loid cells by tumours; Nat Rev Immunol; 2012; PMID: 224379385. Francisco LM et al.: PD-L1 regulates the development, maintenance, and function of induced regulatory T cells; J Exp Med; 2009; PMID: 20008522
Figure 4: Mechanism of action analysis of anti-PD-L1 antibody based on the myeloid cells subsets infiltrating CT26.wt tumors.(A) The numbers of tumor-infiltrating M-MDSCs, PMN-MDSCs, and neutrophils are shown. ** p≤0.01(B) Analysis of the correlations between PMN-MDSCs and tumor volume suggests a loss of function of the PMN-MDSC subset.
PMN-MDSC cells/1 mio. leukocytes
tum
or v
olum
e [m
m³]
500 1000 1500 20000
500
1000
1500
2000
2500
3000
R2=0.6
PMN-MDSC cells/1 mio. leukocytes
tum
or v
olum
e [m
m³]
0 1000 2000 3000 4000 50000
500
1000
R2=0.0
tum
or v
olum
e [m
m3 ]
PMN-MDSC / 1 Mio leukocytesPMN-MDSC cells/1 mio. leukocytes
tum
or v
olum
e [m
m³]
500 1000 1500 20000
500
1000
1500
2000
2500
3000
R2=0.6
PMN-MDSC cells/1 mio. leukocytes
tum
or v
olum
e [m
m³]
0 1000 2000 3000 4000 50000
500
1000
R2=0.0
tum
or v
olum
e [m
m3 ]
PMN-MDSC / 1 Mio leukocytes
B
# ce
llspe
r 1
Mio
. leu
kocy
tes
M-MDSC PMN-MCSCNeutrophils0
50000
100000
150000
vehicleanti-PD-L1
**
M- PMN- NeutrophilsMDSC MDSC
# ce
lls /
1 M
io le
ukoc
ytes
M-MDSC
A
PMN-MDSC
Neutro-phils
Figure 5. Mechanism of action analysis of anti-PD-L1 antibody based on the T cell subsets infiltrating CT26.wt tumors. (A) The graph shows the frequencies of T cell subsets infiltrating CT26.wt tumors. CD44+ is an activation marker, thus activated CD4+ and CD8+ are shown next to regulatory T cells. ** p≤0.01(B) Upon anti-PD-L1 treatment, activated CD8+ T cells gained an advantage in their tumor-killing function over the immune-suppressive regulatory T cells.
# ce
lls/1
Mio
leuk
ocyt
es
A
# ce
llspe
r 1 M
io. l
euko
cyte
s
CD4+CD44+ CD8+CD44+ regulatory 0
5000
10000
15000
20000
25000vehicleanti-PD-L1
**
**
**
T cellsT cells T cellsCD4+CD44+
T cellsCD8+CD44+
T cellsRegulatory
T cells
tum
or v
olum
e [m
m3 ]
tum
or v
olum
e [m
m3 ]
B
CD8+CD44+ / regulatory T cell ratio
tum
or v
olum
e [m
m³]
0 2 4 60
500
1000
1500
2000
2500
3000
R2=0.4
CD8+CD44+ / regulatory T cell ratio
tum
or v
olum
e [m
m³]
0 2 4 60
500
1000
R2= - 0.2
CD8+CD44+ / regulatory T cell ratio
tum
or v
olum
e [m
m³]
0 2 4 60
500
1000
1500
2000
2500
3000
R2=0.4
CD8+CD44+ / regulatory T cell ratio
tum
or v
olum
e [m
m³]
0 2 4 60
500
1000
R2= - 0.2
CD8+CD44+ /Regulatory T cell ratio CD8+CD44+ /Regulatory T cell ratio
The Solution Provider for Drug Discovery in Oncology
Recombinant Proteins Active Protein Kinases Active Lipid Kinases Additional Kinase Variants Recombinant Kinase SubstratesKinaseSampler & KinaseSamplerPlus Custom-Tailored Protein Production
Biochemical Assay Services Free Choice Kinase Assay Projects Kinase Wildtype-Profiler & Kinase 410-ProfilerKinase MutantProfiler KinaseFinder Kinase SubstrateFinder HitDetector Custom-Tailored Services
Cellular Assay Services Phosphorylation Assays Proliferation Assays Soft Agar Assays Migration Assays 3D Tumor Spheroid AssaysAngiogenesis AssayCustom-Tailored Services
In Vivo Testing Services Subcutaneous Tumor Models Orthotopic Tumor Models Metastasis Tumor Models In Vivo Angiogenesis Model Immuno-Oncology Platform
Email: [email protected] Phone: +49-761-206-1780 Fax: +49-761-206-1781
ProQinase GmbH Breisacher Str. 117 D-79106 Freiburg