Nuclear FAK Controls Chemokine Transcription, … show that immune evasion requires nuclear FAK’s...
1
SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 5 10 15 20 25 CD4 + FoxP3 + CD25 + %CD4 + **** **** SCC FAK-wt SCC FAK-/- SCC FAK-kd SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 5 10 15 CD45 + CD11b + F4/80 - %CD45 + G-MDSC M-MDSC * **** 0 3 7 10 14 17 21 0 50 100 150 200 250 300 Day Volume (mm 3 ) SCC FAK-wt CD25 Depletion SCC FAK-/- SCC FAK-wt SCC FAK-wt Isotype Ctrl SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 15 30 45 60 %CD45 + CD45 + CD11b + F4/80 hi Ly6C hi Ly6C lo A E C SCC FAK-wt SCC FAK-/- SCC FAK-kd 0.0 0.2 0.4 0.6 CD8 + : Treg Ratio D B F SCC FAK-wt Thymus Helios Control CD4 + CD25 + FoxP3 + Nuclear FAK Controls Chemokine Transcription, Tregs, and Evasion of Anti-tumor Immunity Alan Serrels, 1,2 Tom Lund, 1 Bryan Serrels, 1 Adam Byron, 1 Rhoanne C. McPherson, 2 Alexander von Kriegsheim, 1 Laura Gómez-Cuadrado, 1 Marta Canel, 1 Morwenna Muir, 1 Jennifer E. Ring, 3 Eleni Maniati, 4 Andrew H Sims, 1 Jonathan A. Pachter, 3 Valerie G. Brunton, 1 Nick Gilbert, 5 Stephen M. Anderton, 2 Robert J.B. Nibbs, 6 and Margaret C. Frame 1 Introduction Here we report a completely new function of the integrin effector signaling protein Focal Adhesion Kinase (FAK) in driving anti-tumor immune evasion. Specifically, the activity of nuclear- targeted FAK in cancer cells drives recruitment and retention of intra-tumoral regulatory T-cells (Tregs) by transcriptionally regulating chemokine and cytokine ligand-receptor networks, crucially including transcription of Ccl5 and Tgfb2. In turn, these changes inhibit antigen-primed cytotoxic CD8+ T-cell activity in the tumor microenvironment, permitting survival and growth of FAK- expressing tumors. We show that immune evasion requires nuclear FAK’s catalytic activity and nuclear localization. Mechanistically, nuclear FAK is associated with chromatin and exists in complex with transcription factors and their upstream transcriptional regulators known to control Ccl5 expression. Moreover, FAK’s nuclear immune-modulatory activities are likely specific to cancer cells, as normal keratinocyte counterparts of SCC cells do not have nuclear FAK. Finally, we show that a small molecule FAK kinase inhibitor, VS-4718, which is currently in clinical development, drives depletion of Tregs and promotes CD8+ T-cell-mediated tumor clearance. It is therefore likely that FAK inhibitors may trigger immune-mediated tumor regression and provide previously unrecognized therapeutic benefit. Email [email protected] www.ecrc.ed.ac.uk www.igmm.ac.uk 0 7 14 21 28 35 42 49 0 200 400 SCC FAK-wt Ctrl SCC FAK-/- Ctrl 0 50 100 150 0 50 100 150 A B 0 3 7 10 14 17 0 200 400 600 800 Days Volume (mm 3 ) SCC FAK-wt SCC FAK-/- **** * CD1-Nude (Immune-deficient) FVB (Immune-competent) 0 3 7 10 14 17 21 0 100 200 300 400 Days Volume (mm 3 ) SCC FAK-wt SCC FAK-/- **** * 0 3 7 10 14 17 21 24 28 0 100 200 300 400 Days Volume (mm 3 ) SCC FAK-wt SCC FAK-/- SCC FAK-kd *** **** 0 3 7 10 14 17 21 0 100 200 300 400 Days Volume (mm 3 ) SCC FAK-wt SCC FAK-/- SCC FAK-/- CD4 Depletion SCC FAK-/- CD8 Depletion SCC FAK-/- CD4 + CD8 Depletion SCC FAK-/- Isotype Ctrl **** **** **** E C 0 3 7 10 14 17 21 0 100 200 300 400 Days Volume (mm 3 ) SCC FAK-wt SCC FAK-/- SCC FAK-wt CD4 Depleted SCC FAK-wt CD8 Depleted SCC FAK-wt CD4 + CD8 Depleted SCC FAK-wt Isotype Ctrl **** **** **** **** D F Volume (mm 3 ) Days Tumor Free Re-challenge: SCC FAK-/- Pre-challenge: SCC FAK-/- Re-challenge: SCC FAK-wt Tumor Free Pre-challenge: SCC FAK-/- Figure 1. Loss of FAK or FAK kinase activity results in CD8 + T-cell dependent SCC tumor clearance. (A) SCC FAK-wt and SCC FAK-/- subcutaneous tumor growth in immune-deficient CD-1 nude mice. *p<0.05, ****p<0.0001 (Sidak corrected 2way ANOVA). (B) SCC FAK-wt and SCC FAK-/- subcutaneous tumor growth in immune-competent FVB mice. *p<0.05, ****p<0.0001 (Sidak corrected 2way ANOVA). (C) SCC FAK-/- tumor growth in FVB mice treated with T-cell depleting antibodies. p<0.0001 (Turkey corrected 2way ANOVA, comparison to SCC FAK-/-). (D) SCC FAK-wt tumor growth in FVB mice treated with T-cell depleting antibodies. p<0.0001 (Turkey corrected 2way ANOVA, comparison to SCC FAK-wt). (E) Top - Secondary tumor re- challenge with SCC FAK-/- cells following a pre-challenge with SCC FAK-/- cells and a 7 day tumor free period. Middle - Secondary tumor re-challenge with SCC FAK-wt cells following a pre-challenge with SCC FAK-/- cells and a 7 day tumor free period. Bottom – Subcutaneous growth of SCC FAK-wt and SCC FAK-/- tumors injected at 28 into a cohort of mice in which no pre-challenge had been performed. (F) Tumor growth measurements in FVB mice following subcutaneous injection of SCC FAK-wt, SCC FAK-/-, and SCC FAK-kd cells. Statistics represent FAK-wt vs FAK-/-. Data are represented as mean ± SEM, N = 5 – 6 tumors / group. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001. Figure 1. Key points: • FAK expression is required for the survival, and growth, of SCC tumors in FVB host mice with a functional adaptive immune system. • CD8+ T-cells are responsible for regression of FAK-/- tumors • FAK-wt SCC tumors have similar immunogenicity to FAK-/- SCC tumors, and that their distinct fates are not due to differences in intrinsic ability to prime an adaptive immune response. • Tumor survival depends on FAK kinase activity. SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 10 30 40 50 60 % CD8 + %Ki-67 + * * %CD8 + CD44 hi SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 10 20 30 40 50 PD-1 + Tim-3 + SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 5 10 15 % CD8 + CD44 hi PD-1 + LAG-3 + * ** SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 5 10 15 %CD8 + CD44 hi PD-1 + Tim-3 + LAG-3 + SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 20 40 60 80 100 120 CD45 + CD3 + CD4 - CD8 + %CD8 + CD44 hi CD62L low CD44 hi CD62L hi CD44 low CD62L low SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 20 40 60 80 100 120 %CD4 + CD45 + CD3 + CD4 + CD8 - CD44 hi CD62L low CD44 hi CD62L hi CD44 low CD62L low SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 1 2 3 Fold Change (normalized to CD8+ T-cell No.) effCD8 + T-cell SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 5 10 15 20 25 %CD45 + CD45 + CD3 + CD4 + CD8 - SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 1 2 3 4 %CD45 + * * CD45 + CD3 + CD4 - CD8 + %CD8 + SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 20 40 60 80 CD69 + SCC FAK-wt SCC FAK-/- SCC FAK-kd 0 20 40 60 80 %CD4 + CD69 + A C B E F L SCC FAK-wt SCC FAK-/- H I J SCC FAK-kd D G K Figure 2. FAK-depleted tumors exhibit a heightened CD8 + T- cell response. (A) FACS quantification of total intra-tumoral CD4 + T-cells. (B) FACS quantification of CD69 + cells as a percentage of CD4 + T-cells. (C) FACS quantification of CD4 + CD44 hi CD62L low , CD4 + CD44 hi CD62L hi , CD4 + CD44 low CD62L hi T-cell subpopulations. (D) FACS quantification of total intra-tumoral CD8 + T-cells. (E) FACS quantification of CD69 + cells as a percentage of CD8 + T- cells. ( F ) Quantification of CD8 + CD44 hi CD62L low , CD8 + CD44 hi CD62L hi , CD8 + CD44 low CD62L hi T-cell subpopulations. (G) Fold-change in effector (CD8 + CD44 hi CD62L low ) CD8 + T-cells when normalized to total CD8 + T-cell percentage. (H) FACS quantification of PD-1 + LAG-3 + T-cells as a percentage of CD8 + CD44 hi tumor infiltrating T-cells from 6 independent tumor samples 7 days post implantation. (I) FACS quantification of PD-1 + Tim-3 + T-cells as a percentage of CD8 + CD44 hi tumor infiltrating T-cells from 3 independent tumor samples 7 days post implantation. (J) FACS quantification of PD-1 + Tim-3 + LAG-3 + T-cells as a percentage of CD8 + CD44 hi tumor infiltrating T-cells from 3 independent tumor samples 7 days post implantation. (K) FACS quantification of Ki-67 + cells as a percentage of tumor infiltrating CD8 + T-cells from 3 independent tumor samples 7 days post implantation. (L) Representative images of histological staining of frozen sections from SCC FAK-wt, SCC FAK-/-, and SCC FAK-kd tumors 7 days post implantation using anti-CD8 antibody clone 53-6.7. Dashed white lines demark tumor boundary. Scale bar = 500 μm. Data are represented as mean ± SEM. N = 5 unless stated otherwise. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001. Figure 3 . FAK regulates the immuno-suppressive microenvironment. (A) FACS quantification of Ly6C hi and Ly6C low macrophage populations expressed as a percentage of tumor infiltrating CD45 + leukocytes. ( B ) FACS quantification of Ly6C hi Gr1 low (M-MDSC) and Ly6C int Gr1 hi (G-MDSC) populations expressed as a percentage of tumor infiltrating CD45 + leukocytes. (C) Quantification of CD4 + CD25 + FoxP3 + Tregs expressed as a percentage of tumor infiltrating CD4 + T-cells. (D) CD8 + T-cell to Treg ratio calculated using mean values from Figs. 2D and 3C. (E) SCC FAK-wt tumor growth in FVB mice treated with anti-CD25 clone PC61.5 T-cell depleting antibodies. Values represent the mean ± SEM from 6 tumors. (F) FACS analysis of Helios expression in CD4 + CD25 + FOXP3 + tumor infiltrating Tregs and Tregs isolated from the thymus of tumor bearing mice. Control represents background signal from a sample stained with CD4, CD25, and FoxP3 conjugated antibodies but not Helios. Representative replicates are shown in different colors for thymus and SCC FAK-wt samples. Data are represented as mean ± SEM unless stated otherwise. FACS analysis represents the mean from N = 5. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001. Figure 4. FAK regulates transcription of cytokines implicated in Treg recruitment and expansion. (A) Transcriptomic profiling of SCC FAK-wt and SCC FAK-/- cells. (B) Functional enrichment analysis of genes upregulated in SCC FAK-wt cells (bottom gray bar in A). Overrepresented biological processes are displayed as a heat map (log 10 -transformed color scale) (top); asterisks indicate presence of cytokine-related genes. Overrepresented gene families are displayed as a bar chart (bottom); dashed line indicates P = 0.001. Displayed terms satisfy P < 0.05 (Benjamini–Hochberg- corrected hypergeometric tests). (C) qRT-PCR array analysis of cytokine and chemokine expression in SCC FAK-wt and SCC FAK-/- cells. Gray bar indicates cluster of genes upregulated in SCC FAK-wt cells; cytokine and chemokine gene names are listed. Green arrowheads indicate reported roles in Treg recruitment; red arrowhead indicates reported role in peripheral Treg induction. (D) qRT-PCR array analysis of chemokine and receptor expression in tumor- and thymus-derived Tregs. Gray bar indicates cluster of genes upregulated in tumor-derived Tregs; receptor gene names are listed. (E) Interaction network analysis of chemokine ligand gene expression detected in SCC cells (circles; left) and corresponding receptor gene expression detected in Tregs (squares; right). Genes are ordered vertically by fold change. Light gray lines connect receptor–ligand pairs; green lines indicate pairs upregulated at least two-fold in SCC FAK-wt cells and tumor- derived Tregs. (F) qRT-PCR analysis of selected cytokine and chemokine gene expression in SCC cells. ****P < 0.0001 (Sidak- corrected one-way ANOVA). Data are represented as mean ± SEM. FAK FAK GAPDH GAPDH PARP PARP SCC FAK-WT Cytoplasm Nucleus SCC FAK-/- SCC FAK-KD FAK Tubulin Total SCC FAK-wt SCC FAK-/- SCC FAK-NLS 0.0 0.5 1.0 Rel. Intensity *** *** SCC FAK-wt pLKO SCC FAK-/- SCC FAK-wt shRNA-Ccl5 P1 SCC FAK-wt shRNA-Ccl5 P2 0.0 0.5 1.0 Rel. Intensity 0 3 7 10 14 17 21 24 28 0 100 200 300 400 500 Days Volume (mm 3 ) SCC FAK-wt pLKO SCC FAK-/- SCC FAK-NLS A B C D F **** **** FAK FAK GAPDH GAPDH PARP PARP SCC FAK-WT Cytoplasm Nucleus SCC FAK-/- SCC FAK-NLS FAK GAPDH Total FAK (60 Sec) FAK (10 min exp.) PARP WC Nuc GAPDH WC Nuc SCC FAK-wt Keratinocyte H G 0 3 7 10 14 17 21 24 27 0 100 200 300 400 500 Days Volume (mm 3 ) SCC FAK-wt pLKO SCC FAK-/- SCC FAK-wt shRNA-CCL5 P1 SCC FAK-wt shRNA-CCL5 P2 **** SCC FAK-wt pLKO SCC FAK-/- SCC FAK-wt shRNA-Ccl5 P1 SCC FAK-wt shRNA-Ccl5 P2 0 20 40 60 No. of Tregs / mg of tumor E Ccl5 CD4 + CD25 + FoxP3 + Ccl5 Figure 5. Nuclear FAK regulates transcription of Ccl5 which is required for Treg recruitment and tumor growth. (A) qRT-PCR analysis of Ccl5 gene expression knockdown in SCC FAK-wt cells stably expressing two independent shRNA constructs targeting Ccl5 (P1 and P2). (B) SCC FAK-wt shRNA-Ccl5 tumor growth in FVB mice. N = 6 tumors / group. (C) FACS quantitation of tumor infiltrating Treg absolute numbers from SCC FAK-wt shRNA-Ccl5 tumors. Data represents a single value from 6 pulled tumor samples. (D) Anti-FAK western blot of cytoplasmic, nuclear, and total protein fractions from SCC FAK-wt, SCC FAK-/-, and SCC FAK-NLS cells. Anti-GAPDH (cytoplasmic) and anti-PARP (nuclear) antibodies were used as controls to check the quality of cellular fractionation. ( E) qRT-PCR analysis of Ccl5 gene expression in SCC FAK-NLS mutant cells. (F) Tumor growth of SCC FAK-NLS expressing cells in FVB mice. (G) Anti-FAK western blot from whole cell (WC) and nuclear (Nuc) lysates extracted from SCC FAK-wt tumor cells and primary skin keratinocytes. 60 second exposure time is shown for all samples and additional 10 minute exposure time is shown for keratinocyte samples. Anti-GAPDH (cytoplasmic) and anti-PARP (nuclear) antibodies were used as controls to check the quality of cellular fractionation. Data are represented as mean ± SEM except when detailed otherwise. p- value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001. Figure 2. Key points: • SCC FAK-/- and FAK-kd (kinase-deficient) tumors have elevated effector CD8 T-cell numbers. • CD8 T-cells infiltrating SCC FAK-wt tumors have elevated expression of markers indicative of a heightened state of dysfunction. Figure 3. Key points: • Regulatory T-cell (Treg) numbers are increased in FAK-expressing tumors. • Loss of FAK or FAK kinase activity alters the CD8 T-cell : Treg ratio in tumors, shifting the balance from tolerance to immunity. • Tregs are required for survival of FAK-expressing SCC tumors. • Tregs are likely recruited into FAK-expressing tumors. Helios positivity implies Tregs are of thymic origin. Figure 4. Key points: • FAK drives expression of chemokines and cytokines in SCC cancer cells. • Network analysis of the relationship between FAK-dependent chemokine ligand expression in SCC cells and tumor-infiltrating Treg chemokine receptor expression shows the existence of a FAK-driven paracrine signaling axis between cancer cells and intra-tumoral Tregs based on chemokine ligand-receptor interactions. • Ccl5, Cxcl10, and TGFb2 expression are dependent on FAK kinase activity. Figure 5. Key points: • Depletion of Ccl5 expression in FAK-expressing SCC cells results in tumor regression associated with decreased intra-tumoral Treg levels. • FAK-dependent expression of Ccl5 requires its nuclear localization and kinase activity. • FAK nuclear accumulation is specific to SCC cancer cells. Figure 6. Key points: • FAK is associated with chromatin, and exists in complex with transcription factors and transcriptional regulators that are linked to regulation of Ccl5 expression. • FAK interacts with core transcriptional machinery to influence gene transcription and promote tumor immune escape. Figure 6. Nuclear FAK interacts with regulators of Ccl5 transcription. (A) Sucrose fractionation of soluble chromatin preparation from nuclei isolated from SCC FAK-wt cells. Protein preparations recovered from each fraction were analyzed by western blotting with anti-FAK, HP1α, and histone H3 antibodies (top three panels). DNA recovered from each fraction was run down an agarose gel alongside a 1 kilobase (kb) and 100 base pair (bp) ladder (bottom panel). Fraction 7 (black arrowhead) represents the chromatin-containing fraction. (B) Schematic detailing the workflow used for proteomic analysis of the nuclear FAK interactome in the context of Ccl5 transcription factors (TFs). (C) Interaction network analysis of proteins that bind FAK in the nucleus of SCC cells. Predicted Ccl5 TFs (squares; bottom) and respective TF binders (circles; top) enriched by at least four-fold in nuclear FAK immunoprecipitations (SCC FAK-wt over SCC FAK-/- controls; p < 0.05) are shown (stringent network). Ccl5 TFs not detected (ND) are shown as gray squares. TF complexes or groups are indicated; proteins are labeled with gene names for clarity. TF binders are aligned above TF groups with which there are the greatest number of reported interactions. For full network, see Supplementary Figure 8. For full protein interaction list, see Supplementary Table 1. (D) Isolation of the TFIID component TAF9 by FAK immunoprecipitation (IP) from SCC FAK-wt cell nuclear extracts. 0 3 7 10 14 17 21 0 200 400 600 800 1000 Days Tumour Volume (mm 3 ) SCC FAK-wt CD8+ Vehicle SCC FAK-wt Isotype Ctrl + Vehicle SCC FAK-wt Isotype Ctrl +VS-4718 SCC FAK-wt CD8+ VS-4718 0 3 7 10 14 17 21 24 28 0 100 200 300 400 Days Volume (mm 3 ) SCC FAK-wt Vehicle Treated SCC FAK-/- Vehicle Treated SCC FAK-wt VS-4718 Treated SCC FAK-/- VS-4718 Treated Start Treatment 0 3 7 10 14 17 21 24 0 100 200 300 400 Day Volume (mm 3 ) SCC FAK-wt Vehicle Treated SCC FAK-wt VS-4718 Treated SCC FAK-/- Vehicle Treated SCC FAK-/- VS-4718 Treated A C B Vehicle VS-4718 Vehicle VS-4718 0 10 20 30 %CD4 + CD4 + FoxP3 + CD25 + SCC FAK-wt SCC FAK-/- **** *** *** D Figure 7. The FAK kinase inhibitor VS-4718 leads to immune mediated SCC clearance. (A) SCC FAK-wt and SCC FAK-/- tumor growth in FVB animals treated with either vehicle alone or VS-4718. Treatment started 24hrs pre-tumor cell inoculation and continued for the duration of the experiment. N = 6 tumors / group. (B) FACS analysis of tumor infiltrating CD4 + CD25 + FoxP3 + Tregs expressed as a percentage of tumor infiltrating CD4 + T-cells. ***p<0.001, ****p<0.0001 (Tukey-corrected one-way ANOVA). (C) SCC FAK-wt tumor growth in FVB animals treated with either vehicle or VS-4718 and either isotype control or CD8 depleting antibodies. N = 6 tumors / group. (D) SCC FAK-wt and SCC FAK-/- tumor growth in FVB animals treated with either vehicle or VS-4718. Treatment started 5 days post-tumor cell inoculation (grey dashed arrow / line) and continued for the duration of the experiment. N = 6 tumors / group. Data are represented as mean ± SEM. Figure 7. Key points: • VS-4718 treatment drives immune-mediated regression of SCC FAK-wt tumors. • VS-4718 modifies the immuno-suppressive tumor environment to promote tumor regression. • FAK inhibitors may trigger immune-mediated tumor regression and provide previously unrecognized therapeutic benefit. 1 Edinburgh Cancer Research UK Centre, Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom. 2 MRC Centre for Inflammation Research, The Queens Medical Research Institute, University of Edinburgh, Edinburgh, United Kingdom. 3 Verastem Inc, 215 First Street, Suite 440, Cambridge, Massachusetts, United States of America. 4 Queen Mary, University of London, Centre for Cancer and Inflammation, Charterhouse Square, London, United Kingdom. 5 MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom. 6 Institute of Infection, Immunity, and Inflammation, University of Glasgow, Glasgow, United Kingdom.
Transcript of Nuclear FAK Controls Chemokine Transcription, … show that immune evasion requires nuclear FAK’s...
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
5
10
15
20
25CD4+ FoxP3+ CD25+
%C
D4+
**** ****
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
5
10
15CD45+ CD11b+ F4/80-
%C
D45
+
G-MDSCM-MDSC* ****
0 3 7 10 14 17 210
50
100
150
200
250
300
Day
Volu
me
(mm
3 )
SCC FAK-wt CD25 DepletionSCC FAK-/-SCC FAK-wt
SCC FAK-wt Isotype Ctrl
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
15
30
45
60
%C
D45
+
CD45+ CD11b+ F4/80hi
Ly6C
hi!
Ly6C
lo!
A!
E!
C!
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0.0
0.2
0.4
0.6
CD
8+ : T
reg
Rat
io
D!B!
F!
SCC FAK-wt!Thymus!
Helios!
Control!CD4+CD25+FoxP3+!
Nuclear FAK Controls Chemokine Transcription, Tregs, and Evasion of Anti-tumor Immunity Alan Serrels,1,2 Tom Lund,1 Bryan Serrels,1 Adam Byron,1 Rhoanne C. McPherson,2 Alexander von Kriegsheim,1 Laura Gómez-Cuadrado,1 Marta Canel,1 Morwenna Muir,1 Jennifer E. Ring,3 Eleni Maniati,4 Andrew H Sims,1 Jonathan A. Pachter,3 Valerie G. Brunton,1 Nick Gilbert,5 Stephen M. Anderton,2 Robert J.B. Nibbs,6 and Margaret C. Frame1
Introduction Here we report a completely new function of the integrin effector signaling protein Focal Adhesion Kinase (FAK) in driving anti-tumor immune evasion. Specifically, the activity of nuclear-targeted FAK in cancer cells drives recruitment and retention of intra-tumoral regulatory T-cells (Tregs) by transcriptionally regulating chemokine and cytokine ligand-receptor networks, crucially including transcription of Ccl5 and Tgfb2. In turn, these changes inhibit antigen-primed cytotoxic CD8+ T-cell activity in the tumor microenvironment, permitting survival and growth of FAK-expressing tumors. We show that immune evasion requires nuclear FAK’s catalytic activity and nuclear localization. Mechanistically, nuclear FAK is associated with chromatin and exists in complex with transcription factors and their upstream transcriptional regulators known to control Ccl5 expression. Moreover, FAK’s nuclear immune-modulatory activities are likely specific to cancer cells, as normal keratinocyte counterparts of SCC cells do not have nuclear FAK. Finally, we show that a small molecule FAK kinase inhibitor, VS-4718, which is currently in clinical development, drives depletion of Tregs and promotes CD8+ T-cell-mediated tumor clearance. It is therefore likely that FAK inhibitors may trigger immune-mediated tumor regression and provide previously unrecognized therapeutic benefit.
Email [email protected] www.ecrc.ed.ac.uk www.igmm.ac.uk
0 7 14 21 28 35 42 490
200400 SCC FAK-wt Ctrl
SCC FAK-/- Ctrl
050100150050100150
A! B!
0 3 7 10 14 170
200
400
600
800
Days
Volu
me
(mm
3 )
SCC FAK-wtSCC FAK-/-
****
*
CD1-Nude (Immune-deficient)! FVB (Immune-competent)!
0 3 7 10 14 17 210
100
200
300
400
Days
Volu
me
(mm
3 )
SCC FAK-wtSCC FAK-/-
****
*
0 3 7 10 14 17 21 24 280
100
200
300
400
Days
Volu
me
(mm
3 )
SCC FAK-wtSCC FAK-/- SCC FAK-kd
***
****
0 3 7 10 14 17 210
100
200
300
400
Days
Volu
me
(mm
3 )
SCC FAK-wtSCC FAK-/-SCC FAK-/- CD4 DepletionSCC FAK-/- CD8 DepletionSCC FAK-/- CD4 + CD8 DepletionSCC FAK-/- Isotype Ctrl
****
********
E!
C!
0 3 7 10 14 17 210
100
200
300
400
Days
Volu
me
(mm
3 )
SCC FAK-wtSCC FAK-/- SCC FAK-wt CD4 DepletedSCC FAK-wt CD8 DepletedSCC FAK-wt CD4 + CD8 DepletedSCC FAK-wt Isotype Ctrl
********
****
****
D!
F!
Volu
me
(mm
3 )!
Days!
Tumor!Free!
Re-challenge:!SCC FAK-/-!
Pre-challenge:!SCC FAK-/-!
Re-challenge:!SCC FAK-wt!
Tumor!Free!
Pre-challenge:!SCC FAK-/-!
Figure 1. Loss of FAK or FAK kinase activity results in CD8+ T-cell dependent SCC tumor clearance. (A) SCC FAK-wt and SCC FAK-/- subcutaneous tumor growth in immune-deficient CD-1 nude mice. *p<0.05, ****p<0.0001 (Sidak corrected 2way ANOVA). (B) SCC FAK-wt and SCC FAK-/- subcutaneous tumor growth in immune-competent FVB mice. *p<0.05, ****p<0.0001 (Sidak corrected 2way ANOVA). (C) SCC FAK-/- tumor growth in FVB mice treated with T-cell depleting antibodies. p<0.0001 (Turkey corrected 2way ANOVA, comparison to SCC FAK-/-). (D) SCC FAK-wt tumor growth in FVB mice treated with T-cell depleting antibodies. p<0.0001 (Turkey corrected 2way ANOVA, comparison to SCC FAK-wt). (E) Top - Secondary tumor re-challenge with SCC FAK-/- cells following a pre-challenge with SCC FAK-/- cells and a 7 day tumor free period. Middle - Secondary tumor re-challenge with SCC FAK-wt cells following a pre-challenge with SCC FAK-/- cells and a 7 day tumor free period. Bottom – Subcutaneous growth of SCC FAK-wt and SCC FAK-/- tumors injected at 28 into a cohort of mice in which no pre-challenge had been performed. (F) Tumor growth measurements in FVB mice following subcutaneous injection of SCC FAK-wt, SCC FAK-/-, and SCC FAK-kd cells. Statistics represent FAK-wt vs FAK-/-. Data are represented as mean ± SEM, N = 5 – 6 tumors / group. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001.!
Figure 1. Key points: • FAK expression is required for the survival, and growth, of SCC tumors in FVB host mice with a functional
adaptive immune system. • CD8+ T-cells are responsible for regression of FAK-/- tumors • FAK-wt SCC tumors have similar immunogenicity to FAK-/- SCC tumors, and that their distinct fates are not
due to differences in intrinsic ability to prime an adaptive immune response. • Tumor survival depends on FAK kinase activity.
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
1030
40
50
60
% C
D8+
%Ki-67+
* *
%CD8+CD44
hi
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
10
20
30
40
50PD-1+Tim-3+
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
5
10
15
% C
D8+ C
D44
hi
PD-1+LAG-3+
***
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
5
10
15
%CD8+CD44
hi
PD-1+Tim-3+LAG-3+
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
20
40
60
80
100
120
CD45+ CD3+ CD4- CD8+
%C
D8+
CD
44hiC
D62
Llow!
CD
44hiC
D62
Lhi!
CD
44lo
wC
D62
Llow!
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
20
40
60
80
100
120
%C
D4+
CD45+ CD3+ CD4+ CD8-
CD
44hiC
D62
Llow!
CD
44hiC
D62
Lhi!
CD
44lo
wC
D62
Llow!
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
1
2
3
Fold
Cha
nge
(nor
mal
ized
to C
D8+
T-c
ell N
o.) effCD8+ T-cell
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
5
10
15
20
25
%CD45
+
CD45+CD3+
CD4+CD8-
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
1
2
3
4
%CD45
+
**
CD45+CD3+
CD4-CD8+
%C
D8+
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
20
40
60
80
CD69+
SCC FAK-wt
SCC FAK-/-
SCC FAK-kd0
20
40
60
80
%C
D4+
CD69+A! C!B! E! F!
L!SCC FAK-wt! SCC FAK-/-!
H! I! J!
SCC FAK-kd!
D!
G! K!
Figure 2. FAK-depleted tumors exhibit a heightened CD8+ T-cell response. (A) FACS quantification of total intra-tumoral CD4+ T-cells. (B) FACS quantification of CD69+ cells as a percentage of CD4+ T-cells. (C) FACS quantification of CD4+CD44hiCD62Llow, CD4+CD44hiCD62Lhi, CD4+CD44lowCD62Lhi T-cell subpopulations. (D) FACS quantification of total intra-tumoral CD8+ T-cells. (E) FACS quantification of CD69+ cells as a percentage of CD8+ T-c e l l s . ( F ) Q u a n t i fi c a t i o n o f C D 8 + C D 4 4 h i C D 6 2 L l o w , CD8+CD44hiCD62Lhi, CD8+CD44lowCD62Lhi T-cell subpopulations. (G) Fold-change in effector (CD8+CD44hiCD62Llow) CD8+ T-cells when normalized to total CD8+ T-cell percentage. (H) FACS quantification of PD-1+LAG-3+ T-cells as a percentage of CD8+CD44hi tumor infiltrating T-cells from 6 independent tumor samples 7 days post implantation. (I) FACS quantification of PD-1+Tim-3+ T-cells as a percentage of CD8+CD44hi tumor infiltrating T-cells from 3 independent tumor samples 7 days post implantation. (J) FACS quantification of PD-1+Tim-3+LAG-3+ T-cells as a percentage of CD8+CD44hi tumor infiltrating T-cells from 3 independent tumor samples 7 days post implantation. (K) FACS quantification of Ki-67+ cells as a percentage of tumor infiltrating CD8+ T-cells from 3 independent tumor samples 7 days post implantation. (L) Representative images of histological staining of frozen sections from SCC FAK-wt, SCC FAK-/-, and SCC FAK-kd tumors 7 days post implantation using anti-CD8 antibody clone 53-6.7. Dashed white lines demark tumor boundary. Scale bar = 500 μm. Data are represented as mean ± SEM. N = 5 unless stated otherwise. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001.#!
Figure 3 . FAK regulates the immuno-suppressive microenvironment. (A) FACS quantification of Ly6Chi and Ly6Clow macrophage populations expressed as a percentage of tumor infiltrating CD45+ leukocytes. (B) FACS quantification of Ly6ChiGr1low (M-MDSC) and Ly6CintGr1hi (G-MDSC) populations expressed as a percentage of tumor infiltrating CD45+ leukocytes. (C) Quantification of CD4+CD25+FoxP3+ Tregs expressed as a percentage of tumor infiltrating CD4+ T-cells. (D) CD8+ T-cell to Treg ratio calculated using mean values from Figs. 2D and 3C. (E) SCC FAK-wt tumor growth in FVB mice treated with anti-CD25 clone PC61.5 T-cell depleting antibodies. Values represent the mean ± SEM from 6 tumors. (F) FACS analysis of Helios expression in CD4+CD25+FOXP3+ tumor infiltrating Tregs and Tregs isolated from the thymus of tumor bearing mice. Control represents background signal from a sample stained with CD4, CD25, and FoxP3 conjugated antibodies but not Helios. Representative replicates are shown in different colors for thymus and SCC FAK-wt samples. Data are represented as mean ± SEM unless stated otherwise. FACS analysis represents the mean from N = 5. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001.!
Figure 4. FAK regulates transcription of cytokines implicated in Treg recruitment and expansion. (A) Transcriptomic profiling of SCC FAK-wt and SCC FAK-/- cells. (B) Functional enrichment analysis of genes upregulated in SCC FAK-wt cells (bottom gray bar in A). Overrepresented biological processes are displayed as a heat map (log10-transformed color scale) (top); asterisks indicate presence of cytokine-related genes. Overrepresented gene families are displayed as a bar chart (bottom); dashed line indicates P = 0.001. Displayed terms satisfy P < 0.05 (Benjamini–Hochberg-corrected hypergeometric tests). (C) qRT-PCR array analysis of cytokine and chemokine expression in SCC FAK-wt and SCC FAK-/- cells. Gray bar indicates cluster of genes upregulated in SCC FAK-wt cells; cytokine and chemokine gene names are listed. Green arrowheads indicate reported roles in Treg recruitment; red arrowhead indicates reported role in peripheral Treg induction. (D) qRT-PCR array analysis of chemokine and receptor expression in tumor- and thymus-derived Tregs. Gray bar indicates cluster of genes upregulated in tumor-derived Tregs; receptor gene names are listed. (E) Interaction network analysis of chemokine ligand gene expression detected in SCC cells (circles; left) and corresponding receptor gene expression detected in Tregs (squares; right). Genes are ordered vertically by fold change. Light gray lines connect receptor–ligand pairs; green lines indicate pairs upregulated at least two-fold in SCC FAK-wt cells and tumor-derived Tregs. (F) qRT-PCR analysis of selected cytokine and chemokine gene expression in SCC cells. ****P < 0.0001 (Sidak-corrected one-way ANOVA). Data are represented as mean ± SEM.#
FAK!
FAK!
GAPDH!
GAPDH!
PARP!
PARP!
SCC FAK-W
T!
Cytoplasm
!Nucleus!
SCC FAK-/-!
SCC FAK-K
D!
FAK!
Tubulin!Total!
SCC FAK-wt
SCC FAK-/-
SCC FAK-NLS
0.0
0.5
1.0
Rel
. Int
ensi
ty
***! ***!
SCC FAK-wt p
LKO
SCC FAK-/-
SCC FAK-wt s
hRNA-Ccl5
P1
SCC FAK-wt s
hRNA-Ccl5
P20.0
0.5
1.0
Rel
. Int
ensi
ty
0 3 7 10 14 17 21 24 280
100
200
300
400
500
Days
Volu
me
(mm
3 )
SCC FAK-wt pLKOSCC FAK-/-SCC FAK-NLS
A! B! C! D!
F!
****!****!FAK!
FAK!
GAPDH!
GAPDH!
PARP!
PARP!
SCC FAK-W
T!
Cytoplasm
!Nucleus!
SCC FAK-/-!
SCC FAK-N
LS!
FAK!
GAPDH!Total!
FAK (60 Sec)!
FAK (10 min exp.)!
PARP!
WC! Nuc!
GAPDH!
WC! Nuc!
SCC FAK-w
t!
Keratin
ocyte!H!
G!
0 3 7 10 14 17 21 24 270
100
200
300
400
500
Days
Volu
me
(mm
3 )
SCC FAK-wt pLKOSCC FAK-/-SCC FAK-wt shRNA-CCL5 P1SCC FAK-wt shRNA-CCL5 P2
****!
SCC FAK-wt p
LKO
SCC FAK-/-
SCC FAK-wt s
hRNA-Ccl5
P1
SCC FAK-wt s
hRNA-Ccl5
P20
20
40
60
No.
of T
regs
/ m
g of
tum
or
E!
Ccl5!CD4+CD25+FoxP3+!
Ccl5!
Figure 5. Nuclear FAK regulates transcription of Ccl5 which is required for Treg recruitment and tumor growth. (A) qRT-PCR analysis of Ccl5 gene expression knockdown in SCC FAK-wt cells stably expressing two independent shRNA constructs targeting Ccl5 (P1 and P2). (B) SCC FAK-wt shRNA-Ccl5 tumor growth in FVB mice. N = 6 tumors / group. (C) FACS quantitation of tumor infiltrating Treg absolute numbers from SCC FAK-wt shRNA-Ccl5 tumors. Data represents a single value from 6 pulled tumor samples. (D) Anti-FAK western blot of cytoplasmic, nuclear, and total protein fractions from SCC FAK-wt, SCC FAK-/-, and SCC FAK-NLS cells. Anti-GAPDH (cytoplasmic) and anti-PARP (nuclear) antibodies were used as controls to check the quality of cellular fractionation. (E) qRT-PCR analysis of Ccl5 gene expression in SCC FAK-NLS mutant cells. (F) Tumor growth of SCC FAK-NLS expressing cells in FVB mice. (G) Anti-FAK western blot from whole cell (WC) and nuclear (Nuc) lysates extracted from SCC FAK-wt tumor cells and primary skin keratinocytes. 60 second exposure time is shown for all samples and additional 10 minute exposure time is shown for keratinocyte samples. Anti-GAPDH (cytoplasmic) and anti-PARP (nuclear) antibodies were used as controls to check the quality of cellular fractionation. Data are represented as mean ± SEM except when detailed otherwise. p-value = Not significant >0.05, * <0.05, ** <0.01, *** <0.001, **** <0.0001. !!
Figure 2. Key points: • SCC FAK-/- and FAK-kd (kinase-deficient) tumors have elevated effector CD8 T-cell numbers. • CD8 T-cells infiltrating SCC FAK-wt tumors have elevated expression of markers indicative of a heightened
state of dysfunction.
Figure 3. Key points: • Regulatory T-cell (Treg) numbers are increased in FAK-expressing tumors. • Loss of FAK or FAK kinase activity alters the CD8 T-cell : Treg ratio in tumors, shifting the balance from
tolerance to immunity. • Tregs are required for survival of FAK-expressing SCC tumors. • Tregs are likely recruited into FAK-expressing tumors. Helios positivity implies Tregs are of thymic origin.
Figure 4. Key points: • FAK drives expression of chemokines and cytokines in SCC cancer cells. • Network analysis of the relationship between FAK-dependent chemokine ligand expression in SCC cells
and tumor-infiltrating Treg chemokine receptor expression shows the existence of a FAK-driven paracrine signaling axis between cancer cells and intra-tumoral Tregs based on chemokine ligand-receptor interactions.
• Ccl5, Cxcl10, and TGFb2 expression are dependent on FAK kinase activity.
Figure 5. Key points: • Depletion of Ccl5 expression in FAK-expressing SCC cells results in tumor regression associated with
decreased intra-tumoral Treg levels. • FAK-dependent expression of Ccl5 requires its nuclear localization and kinase activity. • FAK nuclear accumulation is specific to SCC cancer cells.
Figure 6. Key points: • FAK is associated with chromatin, and exists in complex with transcription factors and transcriptional
regulators that are linked to regulation of Ccl5 expression. • FAK interacts with core transcriptional machinery to influence gene transcription and promote tumor
immune escape.
Figure 6. Nuclear FAK interacts with regulators of Ccl5 transcription. (A) Sucrose fractionation of soluble chromatin preparation from nuclei isolated from SCC FAK-wt cells. Protein preparations recovered from each fraction were analyzed by western blotting with anti-FAK, HP1α, and histone H3 antibodies (top three panels). DNA recovered from each fraction was run down an agarose gel alongside a 1 kilobase (kb) and 100 base pair (bp) ladder (bottom panel). Fraction 7 (black arrowhead) represents the chromatin-containing fraction. (B) Schematic detailing the workflow used for proteomic analysis of the nuclear FAK interactome in the context of Ccl5 transcription factors (TFs). (C) Interaction network analysis of proteins that bind FAK in the nucleus of SCC cells. Predicted Ccl5 TFs (squares; bottom) and respective TF binders (circles; top) enriched by at least four-fold in nuclear FAK immunoprecipitations (SCC FAK-wt over SCC FAK-/- controls; p < 0.05) are shown (stringent network). Ccl5 TFs not detected (ND) are shown as gray squares. TF complexes or groups are indicated; proteins are labeled with gene names for clarity. TF binders are aligned above TF groups with which there are the greatest number of reported interactions. For full network, see Supplementary Figure 8. For full protein interaction list, see Supplementary Table 1. (D) Isolation of the TFIID component TAF9 by FAK immunoprecipitation (IP) from SCC FAK-wt cell nuclear extracts. #
0 3 7 10 14 17 210
200
400
600
800
1000
Days
Tum
our V
olum
e (m
m3 ) SCC FAK-wt CD8+ Vehicle
SCC FAK-wt Isotype Ctrl + Vehicle
SCC FAK-wt Isotype Ctrl +VS-4718SCC FAK-wt CD8+ VS-4718
0 3 7 10 14 17 21 24 28 310
100
200
300
400
Days
Volu
me
(mm
3 )
SCC FAK-wt Vehicle TreatedSCC FAK-/- Vehicle TreatedSCC FAK-wt VS-4718 TreatedSCC FAK-/- VS-4718 Treated
Start Treatment
0 3 7 10 14 17 21 240
100
200
300
400
Day
Volu
me
(mm
3 )
SCC FAK-wt Vehicle Treated
SCC FAK-wt VS-4718 TreatedSCC FAK-/- Vehicle Treated
SCC FAK-/- VS-4718 Treated
A!
C!
B!
Vehicl
e
VS-4718
Vehicl
e
VS-4718
0
10
20
30
%C
D4+
CD4+ FoxP3+ CD25+
SCC FAK-wt SCC FAK-/-
**** *** ***
D!
Figure 7. The FAK kinase inhibitor VS-4718 leads to immune mediated SCC clearance. (A) SCC FAK-wt and SCC FAK-/- tumor growth in FVB animals treated with either vehicle alone or VS-4718. Treatment started 24hrs pre-tumor cell inoculation and continued for the duration of the experiment. N = 6 tumors / group. (B) FACS analysis of tumor infiltrating CD4+CD25+FoxP3+ Tregs expressed as a percentage of tumor infiltrating CD4+ T-cells. ***p<0.001, ****p<0.0001 (Tukey-corrected one-way ANOVA). (C) SCC FAK-wt tumor growth in FVB animals treated with either vehicle or VS-4718 and either isotype control or CD8 depleting antibodies. N = 6 tumors / group. (D) SCC FAK-wt and SCC FAK-/- tumor growth in FVB animals treated with either vehicle or VS-4718. Treatment started 5 days post-tumor cell inoculation (grey dashed arrow / line) and continued for the duration of the experiment. N = 6 tumors / group. Data are represented as mean ± SEM. !!
Figure 7. Key points: • VS-4718 treatment drives immune-mediated regression of SCC FAK-wt tumors. • VS-4718 modifies the immuno-suppressive tumor environment to promote tumor regression. • FAK inhibitors may trigger immune-mediated tumor regression and provide previously unrecognized
therapeutic benefit.
1Edinburgh Cancer Research UK Centre, Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom. 2MRC Centre for Inflammation Research, The Queens Medical Research Institute, University of Edinburgh, Edinburgh, United Kingdom.3Verastem Inc, 215 First Street, Suite 440, Cambridge, Massachusetts, United States of America. 4Queen Mary, University of London, Centre for Cancer and Inflammation, Charterhouse Square, London, United Kingdom. 5MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom. 6Institute of Infection, Immunity, and Inflammation, University of Glasgow, Glasgow, United Kingdom.
