CCL2 is transcriptionally controlled by the lysosomal protease cathepsin S in a CD74-dependent manner

Abstract

Cathepsins S (CatS) has been implicated in numerous tumourigenic processes and here we document for the first time its involvement in CCL2 regulation within the tumour microenvironment. Analysis of syngeneic tumours highlighted reduced infiltrating macrophages in CatS depleted tumours. Interrogation of tumours and serum revealed genetic ablation of CatS leads to the depletion of several pro-inflammatory chemokines, most notably, CCL2. This observation was validated in vitro, where shRNA depletion of CatS resulted in reduced CCL2 expression. This regulation is transcriptionally mediated, as evident from RT-PCR analysis and CCL2 promoter studies. We revealed that CatS regulation of CCL2 is modulated through CD74 (also known as the invariant chain), a known substrate of CatS and a mediator of NFkB activity. Furthermore, CatS and CCL2 show a strong clinical correlation in brain, breast and colon tumours. In summary, these results highlight a novel mechanism by which CatS controls CCL2, which may present a useful pharmacodynamic marker for CatS inhibition.

Keywords: cancer; chemokine; inflammation; microenvironment; protease.

Figure 1. CatS ablation reduces tumour growth and macrophage recruitment

MC38 and LLC cells were stably transduced with lentiviral vectors expressing non-targeting (NT) control or CatS targeting shRNA sequences. a. MC38 NT control and CatS shRNA expressing cells were implanted into wild-type mice, with tumour growth was monitored and recorded as indicated. b. LLC NT control and CatS shRNA expressing cells were implanted into wild-type mice, with tumour growth monitored and recorded as indicated. c. LLC tumours were harvested 10 days post-implantation and subjected to flow cytometry to ascertain the number of F4/80⁺ cells. All in vivo experiments are indicative of 12 tumours, with data representative of the mean ± SEM. d. Raw264.7 cell migration was examined in response to tumour conditioned media from MC38 NT control and CatS shRNA cells (n = 3). e. Raw264.7 cell migration was also examined in response to tumour conditioned media from LLC NT control and CatS shRNA cells (n = 3). f. NIH3T3 fibroblast migration was examined in response to conditioned media from MC38 NT control and CatS shRNA expressing cells (n = 3). All data is representative of the mean ± SEM.

Figure 2. Pro-inflammatory chemokine expression levels are altered in the absence of CatS

MC38 NT control cells were implanted in wild-type mice, with CatS shRNA expressing cells implanted into CatS^−/− mice in order to identify changes in protein expression due to diminished CatS expression. Tumours and serum were extracted 12 days after implantation. a. Interrogation of tumour lysates by antibody array, identified altered expression of CXCL10, CXCL1 and CCL2 (highlighted by the dashed boxes from left to right). b. Densitometry analysis of antibody arrays enabled quantification of expression changes of CXCL10, CXCL1 and CCL2. c. Serum samples were also interrogated by antibody array to determine changes based on CatS expression status. d. Densitometry analysis enabled quantification of the changes in CCL2 expression, made relative to mice bearing the control tumours.

Figure 3. CCL2 expression is correlated with CatS

CCL2 expression was measured by ELISA in a. MC38 cell lysates, b. MC38 supernatants, c. LLC cell lysates and d. LLC supernatants. e. Transient overexpression of CatS in MC38 cells led to the significant induction of CCL2 expression, whereas overexpression of catalytically inactive CatS (CatS C/S) had no impact on CCL2 when measured by ELISA. All experiments are representative of n = 3 and data is representative of the mean ± SEM.

Figure 4. CatS regulation of CCL2 is responsible for altered macrophage recruitment

a. Raw264.7 cell migration was examined using tumour cell conditioned media from MC38 NT control (NT) and CatS shRNA (CatS) expressing cells. NT control conditioned media was also incubated with an antagonistic CCL2 antibody to verify the impact of CCL2 on cell migration. b. The lack of CCL2 expression in B16-F10 cells, was verified by ELISA, in comparison to other cell lines utilized in this work. c. B16-F10 cells were engineered to stably express the NT control and CatS shRNA constructs and knockdown of CatS expression was verified by RT-PCR. GAPDH amplification was used as an internal control to ensure equal loading. d. Raw264.7 cell migration using conditioned media harvested from B16-F10 NT control and CatS shRNA expressing cells was ascertained. All experiments are representative of n = 3 and data is representative of the mean ± SEM.

Figure 5. CatS can transcriptionally regulate CCL2 expression

RT-PCR analysis of CCL2 expression in a. MC38 and b. LLC cells expressing CatS shRNA constructs, with GAPDH amplification used as an internal control. c. MC38 cells were transiently transfected with a CCL2 promoter construct containing a luciferase reporter element. Luciferase activity was measured 24 hrs following transfection using the Dual-Glo^® luciferase assay system. All experiments are representative of n = 3 and data is representative of the mean ± SEM.

Figure 6. CatS regulation of CCL2 is mediated by CD74

MTT assay was used to measure cell viability, following treatment of MC38 cells with a. CD74 agonistic antibody, or b. CD74 ligand, MIF. Absorbance was measured at 570 nm and results expressed relative to the untreated control. c. Transient transfection of a NFkB luciferase construct in MC38 cells. Luciferase activity was measured 24 hrs following transfection using the Dual-Glo^® luciferase assay system. d. The impact of transient overexpression of full length CD74 (CD74-FL) and CD74 intracellular domain (CD74-ICD) on CCL2 expression in MC38 cells was determined by ELISA. e. MC38 cells were examined for alterations in CCL2 production by ELISA due to expression of the CatS shRNA, following treatment with DAPT, or both in combination. All experiments are representative of n = 3 and data is representative of the mean ± SEM.

Figure 7. CatS and CCL2 expression are correlated in human tumour samples

Gene array analysis examining the correlation between CatS and CCL2 in human tumour datasets representing a. colorectal, b. breast and c. brain tumours. d. MC38 cells were treated with an increasing concentration of CatS inhibitor 6 and the impact on CCL2 expression in the supernatants was measured by ELISA. Experiments are representative of n = 3 and data is representative of the mean ± SEM. PC: Pearson coefficient, SR: Spearman rank.

Figure 8. Proposed mechanism of action for CatS mediated regulation of CCL2 via CD74

CD74 is internalized within endocytic vesicles, where the extracellular domain resides within the vesicle and the intracellular domain (CD74-ICD) extends into the cytoplasm. CatS within the endocytic vesicle has been proposed as a key protease responsible for the cleavage of the CD74 within the endosome (green scissors), whereas gamma secretase has been implicated in regulated intramembrane proteolysis of the transmembrane domain of CD74 resulting in liberation of CD74-ICD (blue scissors). Released CD74-ICD can translocate and interact with NFkB in the nucleus where it regulates the expression of chemokines such as CCL2. Adapted from Martín-Ventura, J.L. et al [20].