Background
Smouldering inflammation is a component of the tumor microenvironment, has recently been considered a hallmark of cancer with an important role in tumor initiation and progression[
1,
2]. In tumor microenvironment, innate immune cells are highly represented, and among the most abundant of these are macrophages. Although the original hypotheses proposed that macrophages are involved in antitumor immunity, there is substantial clinical and experimental evidence that in the majority of cases these tumor-associated macrophages (TAMs) enhance tumor progression to malignancy[
3]. TAMs are heterogeneous in response to environmental signals and generally exhibit similarities with prototypic polarized M2 macrophages, contributing to tumor growth, invasion and angiogenesis[
3‐
5]. Human clinical studies have shown a role of TAMs as tumor promoters based on the association of increased density of TAMs with tumor vascularization, metastasis, and poor prognosis[
6‐
10], which have served as a paradigm for cancer-related inflammation. The link between TAMs and tumor development is well established. However, the mechanisms of TAMs moving toward a tumor-promoting phenotype are not fully understood.
Autophagy is an evolutionarily conserved, catabolic process that involves the entrapment of cytoplasmic components within characteristic vesicles for their delivery to and degradation within lysosomes[
11‐
13]. The role of autophagy extends beyond the general homeostatic removal, degradation, and recycling of damaged proteins and organelles to many specific physiological and pathological processes such as development, immunity, energy homeostasis, cell death, tumorigenesis, among others[
14‐
16]. Autophagy is a multifaceted process, and alterations in autophagic signaling pathways are frequently found in cancer[
17‐
19]. While the involvement of autophagy in tumor development is widely accepted, it remains incompletely understood.
It is generally suggested that endolysosomal proteases play important roles in the degradation and regulation events of autophagic processes[
11]. Cathepsins are cysteine lysosomal proteases that are essential for the turnover of intracellular and extracellular proteins internalized by endocytosis; cathepsins are now recognized that cysteine proteases play pivotal roles in cancer progression[
20]. Of the cysteine cathepsins, B, L and S have been implicated most in serving as prognostic markers in cancer associated with poor outcome[
21‐
25]. Cathepsin S (Cat S), unlike the ubiquitous cathepsin B (Cat B) and cathepsin L (Cat L), exhibits a restricted tissue expression. It is found predominantly in lymphatic tissue, macrophages, and other antigen-presenting cells[
26]. There is an increasing body of data highlighting the upregulation of Cat S in a spectrum of tumors[
23‐
25,
27], where levels of the protease increase with the grade and aggressiveness of disease[
28,
29]. Studies previously demonstrated that macrophage-secreted Cat S plays a key role in tumor progression[
30]. Recently, we further elucidated that Cat S deficiency results in abnormal accumulation of autophagosomes in macrophages and enhances Angiotensin II–induced cardiac inflammation[
31]. However, whether Cat S-mediated autophagic flux promotes tumor development via the induction of TAMs polarization remains unclear. Thus, understanding the complex catabolic reactions that occur in the endolysosomal compartment is crucial for elucidating the mechanisms underlying TAMs-mediated tumor-promoting effects.
In this study, we identified that the expression of Cat S by TAMs is critical for promoting tumor growth and metastasis in vivo. We showed that Cat S deletion significantly blocked polarizing macrophages to the M2 phenotype within the tumor microenvironment. Moreover, we observed that Cat S deficiency led to the accumulation of autophagosomes and attenuation of autophagic flux in macrophages within the tumor microenvironment. Finally, we demonstrated that Cat S-mediated autophagic flux was pivotal to maintain polarization of the M2 TAMs phenotype.
Discussion
TAMs represent a dominant myeloid population in many solid tumors, and their accumulation correlates with poor prognosis[
7,
33]. TAMs-secreted cathepsin protease activity has been implicated in the pathogenesis of cancer[
30], yet the mechanisms by it promotes tumor development are incompletely understood. In this study, we identified Cat S as the most abundantly expressed cysteine protease of the cathepsins family in tumor-infiltrating macrophages, and its level was associated with poor prognosis in human colon carcinoma. We showed that Cat S functions as a potentiator of tumor development through maintaining macrophage cellular homeostasis and mediating the switch to an M2 phenotype in TAMs. Moreover, we gained further insights into the potential mechanisms underlying Cat S as a critical mediator of autophagic flux, the cancer-promoting function of TAMs.
Clinical studies have shown that Cat S is expressed in 95% of cases of primary colorectal tumors and their related metastatic tissue, with significantly higher expression in tumors compared with matched normal colonic mucosa[
39].
In vitro findings demonstrated that specific inhibition of Cat S by an antibody, Fsn0503, could attenuate colorectal cancer cell invasion[
25]. Furthermore, these effects were confirmed in a murine model of sporadic pancreatic carcinogenesis (RIP1-Tag2), in which the genetic ablation of Cat S severally inhibited tumorigenesis through attenuation of tumor invasion and angiogenesis[
23]. However, the mechanism by which stroma-derived Cat S promotes tumor development remains unclear. Therefore, in present investigation, we aimed to investigate the role and the mechanism of Cat S in regulating tumor microenvironment. In this study, we observed that colon carcinoma patients at more advanced clinical stages, those with lymph node or liver involvement, or those who have had recurrence within 3 years displayed markedly higher Cat S expression levels (Figure
1B). These results indicate a potential of Cat S as an independent or supplementary biomarker in the prediction of tumor prognosis. Furthermore, we found development of metastasized tumors in the liver and tumor growth of PancO
2 cells were greatly inhibited in Cat S
-/- mice compared to WT littermates (Figure
2), suggesting a critical role of TAMs Cat S expression in contributing to tumor development.
Besides of colon cancer tissues, previous studies have shown that significantly high levels of Cat S have been reported in a range of tumors including glioma[
28], astrocytoma[
24], lung cancer[
40], prostate cancer[
41], hepatocellular[
27], and pancreatic carcinomas[
42], revealing a possible role for this enzyme in tumour growth and progression. Cat S activity is also present in astrocytoma cells in vitro and the extracellular levels of activity were highest in cultures derived from grade IV tumors[
24]. Furthermore, tumor cells such as breast cancer cells, pancreatic or colon cancer cells, express functional Cat S which has been proposed to involve in invasiveness of cancer cells[
43,
44].
In addition to the production of Cat S by the tumor cells, further analysis confirmed that TAMs produced the protease as previously shown in the mouse models for pancreatic cancer and breast cancer[
23,
30,
45,
46]. Several studies have shown that expression of Cat S by tumour-infiltrating macrophages, could be an important contributor during prostate cancer progression[
41]. Recent studies have demonstrated that both tumor cells and tumor-associated macrophages can produce Cat S within the microenvironment to promote neovascularization and tumor growth[
47]. In the present study, the data showed that Cat S deficiency impaired tumor angiogenesis in metastatic foci. Consistent with our results, Cat S null mice exhibited impaired endothelial microvessel development, suggesting a key role for this protease in angiogenesis[
25]. Similarly, Cat S is markedly up-regulated by endothelial cells during tumour angiogenesis[
27,
42] and importantly, in a murine pancreatic islet carcinoma model (RIP1-Tag2), Cat S knockout mice have been shown to a significant reduction in tumour-associated angiogenic switching and neovascularisation[
23]. Furthermore, the Cat S inhibitory antibody, Fsn0503, blocks extracellular proteolysis, inhibiting endothelial invasion and tube formation into developing tumors[
48]. Taken together, reduced angiogenesis in Cat S deficient mice could also partially contribute to impaired tumor develpoment.
Lysosomal cysteine proteinases are known to mediate intracellular protein turnover, regulating the half-life of proteins critical for normal cell function. Cat S is restricted to lymphatic tissues and cells such as macrophages, indicating more specific roles in cell physiology[
40,
49,
50]. Given that macrophages are major participants in host inflammatory responses, dysregulation of macrophage function may lead to diseases including autoimmune disease and cancer[
3]. In the present study, we provide both
in vitro and
in vivo evidence that Cat S is required for M2 polarization and cancer-promoting functions of macrophages in the tumor microenvironment.
Autophagy is upregulated in response to stress, including growth factor and nutrient limitation, energy depletion, and hypoxia[
12,
51]. As a cell-refreshing and metabolism-supporting pathway, autophagy is required for the normal operation of cellular and organismal physiology. Therefore, through selective autophagy, the central metabolism may be supported by different substrates to restore metabolic and energy homeostasis, redox balance, and biomass production. Thus, autophagy can enable stress adaptation, maintenance of cellular fitness, and survival under different stress conditions[
14]. In advanced cancers, autophagy can promote tumor progression by providing nutrients during starvation[
14,
16,
52]. These findings suggest that autophagy inhibition, rather than stimulation, might be beneficial in the treatment of advanced cancer. Given the known importance of TAMs in cancer development and progression, the response of tumors to autophagy inhibition may also involve the immune system. One of the most important functions of autophagy is to maintain cellular energy under conditions of nutrient deprivation and other forms of stress[
15]. The present study established a novel role of macrophage-secreted Cat S in the proper execution of autophagy. We showed that Cat S was required for efficient autophagic flux, based on the facts that Cat S-deficient macrophages showed more accumulation of autophagic vacuole-like structures and multivesicles than WT macrophages by serum starvation during SL4 cells coculture (Figure
4C). Moreover, significant higher levels of LC3-II were detected in Cat S-deficient macrophages than WT macrophages (Figure
4D). Autophagy is a vacuolar lysosomal degradation pathway for long-lived proteins and damaged organelles that are critical for maintaining cell function under stress conditions. During the late stage of autophagy, autophagosomes are presented to lysosomes to degrade damaged organelles (i.e., mitochondria)[
12,
53]. In the present study, Cat S deficiency increased autophagosome formation in macrophages in the tumor microenvironment (Figure
5A). Furthermore, we also showed that Cat S was required for the fusion processes of autophagosomes and lysosomes, as there were significantly defects in the dequenching process of DQ-BSA in Cat S inhibitor-treated macrophages and Cat S-deficient macrophages cocultured with SL4 cells (Figure
5C and D). These results suggest that the fusion of the autophagic vacuole with the lysosome and the subsequent degradation of its content by Cat S are crucial for proper execution of autophagy in TAMs.
Macrophages are highly heterogeneous cells that can be activated to a functional status between M1 and M2 phenotypes in response to environmental signals. In a simplified view, macrophages activated by LPS and IFN-γ are referred to as M1 macrophages, which are capable of killing pathogens and tumor cells[
4,
5]. Macrophages activated by IL-4, IL-13, and IL-10 are referred to as M2 macrophages, which can suppress inflammation, induce angiogenesis, promote tissue repair, and enhance tumor growth[
4,
5]. We demonstrated that the autophagy inhibitor CQ impairs the transition toward the M2 phenotype in macrophages within the tumor microenvironment. Importantly, lack of Cat S and CQ treatment did not produce additive suppression of M2 macrophage transition within the tumor microenvironment, further highlighting the biological necessity of Cat S-mediated autophagy in mediating M2-type polarization of TAMs.
Cat S, a cysteine protease of the papain family, is expressed in lysosomal/endosomal compartments of antigen-presenting cells, such as B cells, macrophages and dendritic cells (DCs)[
54]. Inside the B cells and DCs, Cat S is the single enzyme during the assembly of the MHC class II-α and II-β chains with the antigenic peptide in the lysosomal/endosomal compartments[
55]. This process contributes to antigen-induced adaptive immunity. Moreover, Cat S has been shown to dominate autoantigen processing in human thymic dendritic cells, which would be an important factor that influences selection of autoreactive T cells[
56]. These findings suggest the role of Cat S in other immune cells could potentially also invovle in changes of our observed tumor development in this study.
Materials and methods
Antibodies and reagents
The antibodies for CD68, Ki-67, CD31, Mac-2, GAPDH, and IgG were from Santa Cruz Biotechnology (Santa Cruz, CA); the antibodies for Cat S, F4/80 and LC-3 was from Abcam (Cambridge, MA); and ChemMate TM EnVision System/DAB Detection Kits were from Dako (Glostrup, Denmark). Antibodies for PerCP/Cy5.5-conjugated CD45.2, phycoerythrin (PE)-conjugated F4/80, fluorescein isothiocyanate (FITC)-conjugated CD206 and isotype control were from Biolegend (San Diego, CA). Cat S inhibitor, Z-FL-COCHO was purchased from Calbiochem (San Diego, CA). Autophagy inhibitor, Chloroquine were purchased from Sigma (St. Louis, MO).
Animals
The Cat S-/- mouse strain was backcrossed onto the genetic background of C57BL/6 for more than 10 generations. Mice were 8–12 weeks old at the beginning of the experiments, matched for age and sex with wild-type (WT) mice, and kept under specific pathogen-free (SPF) conditions at the Beijing Anzhen Hospital affiliated to the Capital University of Medical Science, China. All animals received humane care in compliance with the Animal Management Rule of the Ministry of Health, People’s Republic of China (documentation no. 55, 2001) and the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996) and approved by the Institutional Animal Care and Use and Committee of the Capital University of Medical Science (Beijing, China).
Tumor model
PancO2[
57] and SL4[
32] cells are pancreatic and colon cancer cells, respectively, derived from C57BL/6 mice on the same background as the Cat S
-/- mice and WT control mice. PancO2 and SL4 cells were maintained in DMEM/F12 culture medium, supplemented with 10% FBS in a humidified 37°C incubator under 5% CO
2.
For in vivo subcutaneous tumor model, PancO2 cells were harvested and single-cell suspensions of 1.0 × 106 cells in 200 μl medium were injected subcutaneously into the right flank. Mice were sacrificed 28 days after subcutaneous injection. Tumor volume was measured with a caliper using the formula: V = π × [d2× D]/6, where d is the minor tumor axis and D is the major tumor axis. For in vivo hepatic metastasis model, after anaesthetizing mice, a transverse incision in the left flank was made, exposing the spleen, 1.0 × 106 SL4 tumor cells in 100 μl DMEM/F12 medium were intrasplenically injected with use of a 26-gauge needle. 14 days after inoculation, mice were sacrificed, and the tissues were processed as described below. The spleen and liver were removed, wet spleen and liver weights were measured, and the incidence of liver metastasis was examined.
Human colon carcinoma specimens
The specimens from 30 cases of human colon carcinoma tissue/adjacent normal colon tissues and the clinicopathologic data were obtained from the Second Affiliated Hospital to Nanchang University gastrointestinal tumor bank. The specimens were isolated at the time of surgery, formalin-fixed and paraffin-embedded, and stained with hematoxylin and eosin, then examined by 2 experienced pathologists. The clinicopathologic stage was determined according to the TNM classification system of the International Union against Cancer. RNA was extracted for fresh tissue specimens (from 15 colon carcinoma specimens and 15 adjacent normal colon tissues) with identifiable tumor in the tissue specimen. Human specimens use for research had been approved by the Second Affiliated Hospital to Nanchang University Research Ethics Committee.
Histology and immunohistochemistry
Specimens were fixed for 24 hrs with 10% buffered formalin before embedding in paraffin. Serial sections of 5 μm thick were obtained for histologic analysis. Hematoxylin&eosin (HE) staining involved standard procedures.
For immunohistochemistry, sections were incubated with the primary antibodies for Cat S (1:200), CD68 (1:200), Ki-67 (1:200), CD31 (1:200), Mac-2 (1:200), then incubated with the Dako ChemMateTM EnVision System (Dako, Glostrup, Denmark) for 30 min. Staining was visualized with use of diaminobenzidine and counterstaining with hematoxylin. Negative controls were omission of the primary antibody, non-immune IgG or secondary antibody only; in all cases, negative controls showed insignificant staining. The expressions of Cat S, CD68, Ki-67, CD31, Mac-2 were calculated as proportion of positive area to total tissue area for all measurements of the section.
For double immunofluorescence, 7 μm frozen tissue sections were permeabilized and blocked with 0.1% Triton X-100, 0.2% bovine serum albumin, and 5% normal donkey serum in PBS, then incubated with the primary antibodies F4/80 (1:100), LC-3 (1:200) and Cat S (1:200) overnight at 4°C, then FITC or TRITC-conjugated secondary antibody (Jackson Immuno Research Laboratories, West Grove, PA, USA) at 4°C for 1 hr in the dark, and coverslipped with DAPI-containing mounting medium.
RNA analysis
Total RNA were extracted by use of TRIZOL (Invitrogen, Carlsbad, CA, US). For reverse transcription, 1 μg of total RNA was used to generate first strand cDNA with Oligo-dT primer. Real-time PCR was performed using the SYBR Green Mix (Bio-Rad) on the CFX96 Real-time System (Bio-Rad). The Table
1 shows the primer used.
Table 1
Primers used for qRT-PCR
Human cathepsin B | 5′gtttgcattgctggtcagga3′ | 5′tggcaggacagtggaatgat3′ |
Human cathepsin D | 5′gcgagtacatgatcccctgt3′ | 5′ctctggggacagcttgtagc3′ |
Human cathepsin F | 5′ tggcaacaagatgaagcaag3′ | 5′ttttgtgacagcccccttac3′ |
Human cathepsin H | 5′actggctgttgggtatggag3′ | 5′ aggccacacatgttctttcc3′ |
Human cathepsin K | 5′ccgcagtaatgacacccttt3′ | 5′gcacccacagagctaaaagc3′ |
Human cathepsin L | 5′acagtggaccaagtggaagg3′ | 5′cttctcccacactgctctcc3′ |
Human cathepsin S | 5′tcatacgatctgggcatgaa3′ | 5′aggttctgggcactgagaga3′ |
Human cathepsin Z | 5′aagggggtaatgacctgtcc3′ | 5′ttcattgcatgtcccacatt3′ |
Human GAPDH | 5′acagtcagccgcatcttctt3′ | 5′acgaccaaatccgttgactc3′ |
Mouse Arg-1 | 5′aaagctggtctgctggaaaa3′ | 5′ acagaccgtgggttcttcac3′ |
Mouse FIZZ1 | 5′ttgcaactgcctgtgcttac3′ | 5′ctgggttctccacctcttca3′ |
Mouse IL-10 | 5′ccaagccttatcggaaatga3′ | 5′ttttcacaggggagaaatcg3′ |
Mouse iNOS | 5′gggctgtcacggagatca3′ | 5′ccatgatggtcacattctgc3′ |
Mouse IL-1β | 5′gcccatcctctgtgactcat3′ | 5′aggccacaggtattttgtcg3′ |
Mouse TNF-α | 5′tcttctcattcctgcttgtgg3′ | 5′ggtctgggccatagaactga3′ |
Mouse Tubulin | 5′tctaacccgttgctatcatgc3′ | 5′gccatgttccaggcagtag3′ |
Western blotting
Protein extracts were diluted with loading buffer and separated by electrophoresis on 8%-10% SDS-polyacrylamide gels before transfer to nitrocellulose membranes (Bio-Rad). The membranes were blocked in Odyssey blocking buffer (LI-COR Bioscience, Lincoln, NE) at room temperature for 1 hr, then incubated at 4°C overnight with primary antibodies: Cat S (1:1000), LC-3 (1:1000), GAPDH (1:3000). The membranes were washed 3 times in TBST and incubated with fluorescent secondary antibodies (Alexa Fluor 680 or IRDye 800, Rockland Immunochemicals, Gilbertsville, PA, US) for 1 hr at room temperature at 1:5000, blots were analyzed with the Odyssey infrared imaging system and Odyssey software.
Flow Cytometry
The content of inflammatory cells was quantified by flow cytometry as described[
58]. Briefly, tumor tissues were cut into multiple small cubes and digested in an enzyme mixture for 45 mins at 37°C. The cell suspension was centrifuged and pre-incubated with Fc-γ block antibody (anti-mouse CD16/32; Pharmingen, San Diego, CA, USA) to prevent nonspecific binding. Cell staining involved different combinations of fluorochrome-coupled antibodies to CD45, F4/80, CD206 for 30 mins at 4°C in the dark. Fluorescence data were collected by use of an EPICS XL flow cytometer (Beckman Coulter) and analyzed by use of Cellquest (Beckman). Fluorescence minus one (FMO) controls were included to determine the level of nonspecific staining and auto-fluorescence associated with subsets of cells in each fluorescence channel.
Co-cultures of bone marrow-derived macrophages (BMDMs) and tumor cells
Co-culture systems were established by using transwell inserts (0.4 mm pore, polycarbonate membrane; Costar, Cambridge, USA) and transferred to 6-well culture plates. Bone marrow-derived macrophages (BMDMs) were isolated from tibias and femurs of 8-week-old wild-type C57BL/6 or Cat S
-/- mice as previously described[
58]. SL4 cell suspensions (1 ml, 1 × 10
6 cells) were loaded in the upper inserts, and WT or Cat S
-/- BMDMs suspensions (3 ml, 3 × 10
6 cells) were put into the lower compartment of the culture well for 48 hrs and serum-starved for at least 12 hrs before co-culture. Serum-free DMEM/F12 without SL4 cell inserts was used as a control in the lower compartment of well. BMDMs were preincubated for 1 hr with 10 μmol/L Cat S inhibitor Z-FL-COCHO or 50 μmol/L autophagy inhibitor Chloroquine, immediately after SL4 cells plating and controls received equivalent dilution with DMSO vehicle alone.
Transmission electron microscopy
For transmission electron microscopy (TEM), cells were fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.0) for 2 hrs, post-fixed in 2% osmium tetroxide for 2 hrs and then rinsed with 0.1 M cacodylate buffer. Cells were enrobed in 5% Noble Agar and washed with distilled water 5 times, further fixing with 2% uranyl acetate for 2 hrs, followed by dehydration in 50% (15 min), 70% (16 h), 85% (15 min), 95% (15 min), and 2 changes of 100% ethanol each 15 min. They were then cleared by 2 changes of propylene oxide, each 15 min, and infiltrated with epon resin:propylene oxide (1:1) for 3 hrs, epon resin:propylene oxide (3:1) for 16 hrs, and 2 changes with pure epon resin for total 6 hrs. Thin sections were mounted on grids and examined under the electron microscope (Philips EM410).
Autophagy flux assays
mCherry-GFP-LC3 transfection
Cat S
-/- or WT BMDMs were transfected with mCherry-GFP-LC3 reporter construct by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) as previously described in our lab[
31]. After transfection, Cat S
-/- or WT BMDMs cocultured with SL4 cells without serum for 48 hrs. Cells were fixed with 4% paraformaldehyde and microphotographs of mCherry-GFP-LC3 fluorescence were obtained with the confocal laser-scanning microscope (TCS 4D; Leica, Heidelberg, Germany). Treatment-induced changes in GFP intensity in mCherry-positive puncta were calculated to depict autophagic flux.
High-throughput image analysis
Cat S-/- or WT BMDMs expressing GFP-LC3 were seeded in 96-well plates, then cocultured with SL4 cells in 100 μl of medium/well without serum for 48 hrs. Images data were collected with an ArrayScan HCS 4.0 Reader with a 20× objective (Cellomics) for hoechst-labeled nuclei and GFP-tagged LC3. The detection of punctated staining of GFP-LC3 from the diffuse staining indicated the formation of autophagosomes. Images of 1,000 cells for each well were analyzed to obtain the average cell number per field, fluorescence spot number, area, and intensity per cell.
DQ-BSA degradation assays
To quantify protein degradation induced by stimuli, autophagy flux was analyzed by flow cytometry and confocal microscopy using the self-quenched substrate DQ-Green BSA (Molecular Probes, Eugene, OR) as described in previous studies[
59]. Briefly, WT or Cat S
-/- BMDMs were cocultured with SL4 cells for 48 hrs and serum-starved for at least 12 hrs before co-culture. WT BMDMs were in the presence or absence of Cat S inhibitor Z-FL-COCHO (10 μmol/L) cocultured with SL4 cells. WT or Cat S
-/- BMDMs into the lower compartment were loaded with 10 μg/ml DQ-Green BSA for 15 min at 37°C. The cells were washed twice with PBS to remove excess label, then harvested at indicated time points. Cells were harvested, and Green-fluorescent of DQ-BSA was analyzed by flow cytometry using a FACSCalibur flow cytometer (BD Biosciences) and CellQuest (BD Biosciences) and FlowJo (Treestar) software. For confocal images analysis, cells were placed on coverslips after treatment as above and fixed with 4% formaldehyde. The fluorescent degradation products of DQ-BSA in lysosomes were imaged using a confocal laser-scanning microscope (TCS 4D; Leica, Heidelberg, Germany).
Statistics
Data were analyzed with GraphPad Prism v5.00 for Windows. Results are expressed as mean ± SEM. Differences were analyzed by t test or ANOVA, and results were considered significant at a P<0.05.
Competing interests
The authors declare that there are no conflicts of interest.
Authors’ contributions
MY and JL equally participated in the design and execution of the overall study. JS carried out the pathologic analysis. YQ provided Cat S knockout mice and performed mice genotyping. QJ, XZ participated in its coordination and provided technical support. JD was involved in the conception and design of the study as well as in revising the manuscript. All authors read and approved the final manuscript.