This work shows that sphingosine kinase 1 (SphK1) expression in peritoneal macrophage induces expression of cyclooxygenase-2 and tumor necrosis factor-α, leading to colon carcinogenesis. The results strongly support that targeting SphK1 represents a novel approach for inflammatory colon cancer chemoprevention and/or treatment.
Abstract
Accumulating evidence suggests that the sphingosine kinase 1 (SphK1)/sphingosine 1-phosphate (S1P) pathway plays a pivotal role in colon carcinogenesis. Our previous studies indicate that the SphK1/S1P pathway mediates colon carcinogenesis at least by regulating cyclooxygenase 2 (COX-2) expression and prostaglandin E2 (PGE2) production. However, the mechanisms by which this pathway regulates colon carcinogenesis are still unclear. First, we show that SphK1 deficient mice significantly attenuated azoxymethane (AOM)-induced colon carcinogenesis as measured by colon tumor incidence, multiplicity, and volume. We found that AOM activates peritoneal macrophages to induce SphK1, COX-2, and tumor necrosis factor (TNF)-α expression in WT mice. Interestingly, SphK1 knockout (KO) mice revealed significant reduction of COX-2 and TNF-α expression from AOM-activated peritoneal macrophages, suggesting that SphK1 regulates COX-2 and TNF-α expression in peritoneal macrophages. We found that inoculation of WT peritoneal macrophages restored the carcinogenic effect of AOM in Sphk1 KO mice as measured by aberrant crypt foci (ACF) formation, preneoplastic lesions of colon cancer. In addition, downregulation of SphK1 only in peritoneal macrophage by short hairpin RNA (shRNA) reduced the number of ACF per colon induced by AOM. Intraperitoneal injection of sphingolipids demonstrates that S1P enhanced AOM-induced ACF formation, while ceramide inhibited. Finally, we show that SphK inhibitor SKI-II significantly reduced the number of ACF per colon. These results suggest that SphK1 expression plays a pivotal role in the early stages of colon carcinogenesis through regulating COX-2 and TNF-α expression from activated peritoneal macrophages.
Introduction
Colorectal cancer is the second leading cause of all cancer-related deaths, with 49190 estimated annual deaths in the USA in 2016 (1). Despite aggressive screening guidelines for early detection and detailed knowledge of several critical events underlying the pathogenesis of colorectal cancer, it continues to be a major health concern in developed countries.
Sphingolipids, especially ceramide, sphingosine, ceramide-1-phosphate and sphingosine-1-phosphate (S1P), play key roles as regulatory molecules in cancer development (2). S1P promotes cell proliferation and survival, and regulates angiogenesis, whereas sphingosine and ceramide inhibit cell proliferation and stimulate apoptosis. Sphingosine kinase (SphK) phosphorylates sphingosine to form S1P and is a critical regulator of sphingolipid-mediated functions. Two mammalian isozymes, SphK1 and SphK2, have been identified (3). An inducible SphK1 is activated by numerous growth factors and cytokines, and recent studies have revealed that the SphK1/S1P pathway regulates the cyclooxygenase-2 (COX-2)/prostaglandin E2 (PGE2) pathway (4–8). Furthermore, SphK1 knockout (KO) mice showed lower colon cancer development in an inflammation-related colon carcinogenesis mouse model (5). Thus, the SphK1/S1P pathway may play a crucial role in colon carcinogenesis. However, the mechanisms by which this pathway mediates colon carcinogenesis are still unclear.
Numerous studies reveal that the arachidonic acid pathway, especially the COX-2/PGE2 pathway, is involved in colon carcinogenesis. For example, COX-2 inhibitors significantly reduced colon carcinogenesis in rodents (9,10) and inhibited intestinal polyp formation in familial adenomatous polyposis patients (11). PGE2 administration enhanced colon carcinogenesis (12) and overcame indomethacin-reduced intestinal polyp formation in ApcMin mice (13).
Although COX-2 is expressed in colon cancer cells (14), several studies show that COX-2 expression is in macrophages localized in the subepithelium adjacent to the adenoma in human colon (15). Activated macrophages are known to produce tumor necrosis factor (TNF)-α, IL-1β and IL-6, which not only induce COX-2 expression but also stimulate proliferation and migration of human colon cancer cells (16). Together, the evidence suggests that macrophages may play an important role in colon carcinogenesis.
In this study, with azoxymethane (AOM)-induced aberrant crypt foci (ACF; preneoplastic lesions of colon cancer) formation model, we provide evidence that SphK1 expression in peritoneal macrophages may accelerate initiation of colon carcinogenesis. We also show that SphK1 induced COX-2 and TNF-α expression in peritoneal macrophages in response to AOM injection. We then discuss the implications of these results for the role of the SphK1/S1P pathway in macrophage and the potential that blockage of macrophage activation is one of the inhibitory mechanisms by which SphK1 deficiency prevents colon cancer.
Material and methods
Mice
Mice were housed and handled in the division of laboratory animal resources facility under Medical University of South Carolina (MUSC) and University of Hawaii (UH) guidelines. Mice were maintained under controlled conditions of humidity (50 ± 10%), light (12/12 h light/dark cycle) and temperature (23 ± 2°C). All mouse experiments were approved by the institutional animal care and use committee at MUSC. SphK1 homozygous KO mice of the 129SV-C57BL/6 background, a kind gift from Dr. Richard L. Proia [National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)/NIH, Bethesda, MD], were backcrossed to C57BL/6 wild-type (WT) mice (purchased from Charles River Laboratories, Wilmington, MA) at least 10 times (17). Genotypes of SphK1 KO mice were determined by PCR analysis of genomic DNA isolated from tail biopsies (17).
Isolation of peritoneal macrophage and real-time PCR analysis
SphK1 KO and C57BL/6 WT mice at 8 weeks of age (n = 5 for each group) were intraperitoneally injected with either vehicle or AOM (10 mg/kg). The peritoneal macrophages were harvested by washing with 10 ml ice-cold RPMI 1640 (Thermo Fisher Scientific, Waltham, MA) with 1% fetal bovine serum (FBS) and antibiotics (100 U/ml of penicillin, 100 µg/ml of streptomycin) as described previously (18), and placed in 35-mm plates. The macrophages were allowed to adhere for 2 h in the growth media at 37°C and 5% CO2 in an incubator. Peritoneal cells except for macrophages were removed by washing with phosphate-buffered saline (PBS). RNA from the macrophages was extracted using RNeasy Mini Kit (Qiagen, Valencia, CA) according to manufacture’s instructions. cDNA was synthesized using qScript cDNA SuperMix (Quanta Biosciences, Gaithersburg, MD). Real-time PCR was performed with MyiQ2 Two-Color Real-Time PCR Detection System (Bio-Rad). The standard real-time PCR reaction volume was 20 µl, and consisted of 10 µl of PerfeCTa SYBR Green FastMix (Quanta Biosciences), 7 µl RNAse-free H2O, 1 µl forward primer (final concentration of 1 µM), 1µl reverse primer (1 µM) and 1 µl cDNA (0.5 ng/µl). All reactions were performed in triplicate. Primer sets can be found in Supplementary Table S1, available at Carcinogenesis Online.
The data were analyzed using Q-Gene software (19) and expressed as mean normalized expression (MNE). MNE is directly proportional to the amount of RNA of the target gene (SphK1, SphK2, COX-2 and TNF-α) relative to the amount of RNA of the reference gene β-actin.
Cell culture and AOM treatment
Peritoneal cells were harvested as described above. Peritoneal cells including macrophages and macrophages purified from peritoneal cells were cultured in RPMI 1640 plus 10% FBS and antibiotics (100 U/ml of penicillin, 100 µg/ml of streptomycin) with/without AOM (1 mg/ml) for 3 h. Macrophages were washed with Dulbecco's PBS and RNA from the macrophages was extracted as described above. IC-21 mouse peritoneal macrophages were purchased from American Type Culture Collection (ATCC, Manassas, VA). IC-21 cells were maintained in RPMI 1640 with 10% FBS and antibiotics (100 U/ml of penicillin, 100 µg/ml of streptomycin). IC-21 cells were cultured in RPMI 1640 media with/without AOM (1 mg/ml) for 3 h.
Transfection of cell line and TaqMan OpenArray
A plasmid with a sequence-verified mouse SphK1 cDNA cloned within pCMV6-Entry vector and plasmid with vector alone (Origene Technologies, Rockville, MD) were transiently transfected into IC-21 cells using Fugene HD transfection reagent (Roche Diagnostics, Indianapolis, IN). RNA was extracted as described above, reverse transcribed into cDNA and loaded into TaqMan® OpenArray® Mouse Inflammation Panel plate (Thermo Fisher Scientific, Ref. 4475393) consisting of 632 gene targets selected for their involvement in inflammatory response. Real-time PCR and quantitative gene expression were performed on a QuantStudio 12K Flex Real-Time PCR System (Thermo Fisher Scientific).
ACF formation assay
AOM was obtained from the NCI Chemical Carcinogen Reference Standard Repository (Bethesda, MD). The ACF formation assay was performed as described previously (5). Briefly, mice received an AOM injection (10 mg/kg, i.p.) once a week for 3 weeks, starting at 6 weeks of age. All animals were killed at 2 weeks after the last AOM administration. After laparotomy, the entire colon was resected and filled with 10% neutral buffered formalin and then opened longitudinally from the anus to the cecum. Each colon was fixed flat between sheets of filter paper in 10% neutral buffered formalin for 24 h. All colons were stained with 0.1% methylene blue (Sigma-Aldrich, St. Louis, MO) in saline and assessed under a light microscope for the number of ACF per colon according to the procedure of Bird (20).
Preparation of resident peritoneal macrophages
Resident peritoneal cells were harvested from untreated WT or SphK1 KO mice (6–8 weeks old) by washing with 10-ml ice-cold PBS containing antibiotics and 2 mM EDTA as described previously (18) and washed with the same buffer. Resident peritoneal cells placed in 35-mm plates and the macrophages were allowed to adhere for 2 h in the growth media at 37°C and 5% CO2 in an incubator. Peritoneal cells except for macrophages were removed by washing with PBS. The macrophages were detached using enzyme-free Cell Dissociation Buffer (Thermo Fisher Scientific, Waltham, MA) and adjusted to 1 × 107 cells/ml and kept on ice until inoculation. The cell suspension (0.1 ml; 1 × 106 cells) was intraperitoneally inoculated into SphK1 KO mice 1 day before AOM injection. Six of WT mice received AOM alone, nine of SphK1 KO mice received AOM and SphK1 KO macrophage, and eight of SphK1 KO mice received AOM and WT macrophage. ACF formation was analyzed as described above.
Knockdown of SphK1 in peritoneal macrophage by short hairpin RNA in vivo
To knockdown SphK1 only in peritoneal macrophages, we employed in vivo-jetPEI-Man (Polyplus Transfection, New York NY), which is a mannose-conjugated linear polyethylenimine derivative that binds to cells expressing mannose-specific membrane receptors, such as macrophages. Short hairpin RNA (shRNA) for SphK1 and scramble negative control (SCR) were purchased from Qiagen (KM26123H). The in vivo-jetPEI®-Man/shRNA complexes were prepared and transfected according to manufacture’s instruction. Briefly, mice received i.p. injection of shRNA (2 mg/kg) 1 day before AOM injection, once a week for 3 weeks. Twelve of the WT mice received SCR shRNA injection and 13 of the WT mice received SphK1 shRNA injection. ACF formation was analyzed as described above.
Exposure of peritoneal cells to sphingolipids in vivo
To determine the effect of exposure of peritoneal cells to sphingolipid on AOM-induced ACF, mice were separated into four groups and injected with C2-ceramide, sphingosine, sphingosine-1-phosphate or 10% DMSO (control) (n = 12/group). Then, 9 of the 12 mice in each group were injected with AOM 1 h after the sphingolipid injection. The remaining three mice in each group were injected with PBS as a control. ACF formation was analyzed as described above.
SphK inhibitor SKI-II treatment
Male C57BL/6 WT mice received oral administration (P.O.) of SKI-II (n = 10, 100 mg/kg, ChemBridge Corporation, San Diego, CA) or vehicle (n = 10, polyethylene glycol 400, PEG400, Sigma-Aldrich, St. Louis, MO) by gavage 1 h before AOM injection once a week for 3 week. ACF formation was analyzed as described above.
Statistical analysis
Experimental data were expressed as mean with standard error. All statistical analyses were conducted using the Student t test for comparing means of two groups, and one-way or two-way analysis of variance with post-hoc Dunnett’s multiple comparisons test or Tukey test when comparing more than two groups. A P value of <0.05 was considered significant. All statistical analyses and figures were carried out using GraphPad Prism software 6.0 (GraphPad Software, Inc.).
Results
AOM activates peritoneal macrophages to induce SphK1
First, we examined whether AOM can activate peritoneal macrophages to induce SphK1 and/or SphK2 (protocol summarized in Figure 1A). We show that SphK1 expression in peritoneal macrophages was significantly up-regulated by AOM treatment (P < 0.05, Figure 1B). Interestingly, the levels of SphK2 were not changed in response to AOM (Figure 1C). One day after AOM exposure, macrophage SphK1’s expression returned to control levels (data not shown). These results indicate that AOM up-regulates SphK1 but not SphK2 in peritoneal macrophages.
Figure 1.
Effect of AOM injection on gene expression in macrophage. (A) A protocol for isolation of peritoneal macrophage and real-time PCR analysis. AOM injection induces SphK1 (B), not SphK2 (C) expression in WT macrophages. SphK1 deficiency abolished COX-2 (D) and TNF-α (E) expression in macrophages in response to AOM injection. Data were expressed as MNE, which is normalized with β-actin as the reference gene. **P < 0.01, ***P < 0.001.
SphK1 deficiency suppresses COX-2 and TNF-α expression induced by AOM in peritoneal macrophages
Since previous results implicated SphK1 in the induction of COX-2 (4–8) and having demonstrated the acute effects of AOM on macrophages and AOM’s induction of SphK1, it became important to determine if AOM induces COX-2 and TNF-α in peritoneal macrophages, which are important inflammatory factors in colon carcinogenesis. In addition, we sought to determine if SphK1 plays a role in regulation of these factors in peritoneal macrophages. Thus, we analyzed levels of COX-2 and TNF-α expression in peritoneal macrophages isolated from SphK1 KO mice. The results clearly show that AOM-induced COX-2 and TNF-α expression in peritoneal macrophages in WT mice (P < 0.01, Figure 1D and E, black bars). Thus, AOM is capable of acutely activating peritoneal macrophages, a previously unsuspected target for the action of AOM.
Interestingly, the levels of COX-2 and TNF-α expression in peritoneal macrophages isolated from SphK1 KO mice were significantly suppressed when compared with WT mice (P < 0.01, Figure 1D and E, gray bars), demonstrating that SphK1 plays a pivotal role in regulating these important inflammatory factors induced by AOM in peritoneal macrophages.
AOM does not induce SphK1 expression in vitro
To investigate whether AOM directly induces SphK1 expression, we harvested peritoneal cells including macrophages and treated whole peritoneal cells (co-culture) and purified macrophages (mono-culture) with AOM (1 mg/ml) in vitro. The result showed that AOM treatment did not increase SphK1 expression in macrophages cultured with and without other peritoneal cells (Supplementary Figure S1A is available at Carcinogenesis Online), suggesting that AOM is not capable of activating SphK1 in vitro. We also employed IC-21 mouse peritoneal macrophage cell line. We confirm that treatment of 1 mg/ml AOM for 3 h did not induce SphK1 expression (Supplementary Figure S1B is available at Carcinogenesis Online).
SphK1 overexpression in macrophages activates inflammation pathways
To explore the mechanism by which SphK1 in macrophages influences colon carcinogenesis, we overexpressed SphK1 by plasmid transfection and assessed the alteration of global gene expression in SphK1-overexpressing IC-21 mouse peritoneal macrophages (Supplementary Table S2 is available at Carcinogenesis Online). We first confirmed that SphK1 was highly expressed in SphK1 plasmid-transfected cells (Figure 2A). The heatmap analysis of gene expression revealed that SphK1 expression was associated with a change in global gene expression levels in peritoneal macrophages (Figure 2B). We found that a total of 186 genes had significant changes in their gene expression, including 54 upregulated and 132 downregulated genes. Although the OpenArray is an inflammation panel, the genes on the panel have additional functions. Therefore, the genes were further classified into various cellular signaling pathways using Molecule Annotation System software (Figure 2C). We identified 16 pathways including Signaling by Interleukins, Fc epsilon receptor (FCERI) signaling, Toll-Like receptors Cascades and MyD88-related pathways, suggesting that SphK1 in macrophages is capable of activating those pathways.
Figure 2.
Effect of SphK1 overexpression on gene expression in macrophages. (A) Transfection of SphK1 plasmid significantly increased SphK1 expression in IC-21 cells. (B) Heat map from two-way hierarchical clustering analysis based on the 281 common shared probe sets and 4 samples (2 per plasmid). The genes with at least one NaN (Not a Number) measurements are eliminated. (C) Functional annotation of the genes regulated by SphK1. Pie chart represents the percentages of genes annotated to each functional category. Functional categories were assigned by the Panther tools (http://www.pantherdb.org/tools/).
Macrophages isolated from WT mice partially rescue AOM-induced ACF formation in SphK1 KO mice
We examined the possibility that SphK1 expression in peritoneal macrophages functions as a key player in the initiation of colon carcinogenesis. We sought to determine the ability of resident peritoneal macrophages isolated from WT or SphK1 KO mice to induce ACF formation in SphK1 KO mice. SphK1 KO mice were inoculated (i.p.) with WT or KO macrophages. The protocol for this experiment is summarized in Figure 3A. To confirm active macrophages persisted in the peritoneal cavity after inoculation, mice were injected AOM 1 day after macrophage inoculation, and then macrophages were isolated 3 h post AOM injection followed by SphK1 expression analysis using real-time RT-PCR. As expected, SphK1 KO macrophages were void of SphK1 while WT macrophages showed expression of SphK1, indicating WT macrophages indeed remained with SphK1 expression in the peritoneal cavity of SphK1 KO mice (Figure 3B, P < 0.0001). Also, SphK1 KO mice inoculated with WT macrophages showed trends that COX-2 and TNF-α expression levels were higher than SphK1 KO mice inoculated SphK1 KO macrophages (Figure 3B, P = 0.1848 and 0.1430, respectively). The most intriguing finding is that SphK1 KO mice, which normally would have reduced AOM-induced ACF when inoculated with WT peritoneal macrophages, developed significantly greater AOM-induced ACF compared with SphK1 KO mice inoculated with SphK1 KO macrophages (Figure 3C, P < 0.05). The WT macrophages restored the carcinogenic effect of AOM in SphK1 KO mice, suggesting that SphK1 is necessary for macrophages to activate and play an integral role in the early stages of colon carcinogenesis. While the introduction of WT macrophages into SphK1 KO mice augmented ACF formation, ACF formation remained lower compared with WT mice given AOM (P = 0.1412). This observation indicates that other mechanisms are also involved in ACF formation while “SphK1 expression in peritoneal macrophages” accounts for partial ACF formation.
Figure 3.
Effect of macrophage inoculation into SphK1 KO mice. (A) A protocol for macrophage inoculation and ACF formation. White rhombus, black rhombus, black triangle and arrow indicate SphK1 KO macrophage inoculation, WT macrophage inoculation, AOM injection and euthanize, respectively. Group 1: AOM alone in WT mice, Group 2: AOM and SphK1 macrophages in SphK1 KO mice B, Group 3: AOM and WT macrophages in SphK1 KO mice. (B) Expression levels of SphK1, COX-2 and TNF-α in peritoneal macrophages isolated from SphK1 KO mice inoculated with SphK1 KO or WT macrophages. Data were expressed as MNE, which is normalized with β-actin as the reference gene. P values are compared with SphK1 KO macrophages. (C) WT macrophages inoculation into SphK1 KO mice ‘rescued’ AOM-induced ACF formation. Bars represent group mean and standard error of mean; symbols represent ACF number of each animal. *P < 0.05, ***P < 0.001 versus SphK1 KO treated with AOM and SphK1 KO macrophages.
Targeting macrophage SphK1 inhibits AOM-induced ACF formation
To confirm that SphK1 expression in peritoneal macrophages plays a role in colon carcinogenesis, we selectively silenced macrophage SphK1 with shRNA and examined the effects on ACF formation induced by AOM (Figure 4A). First, we confirmed that in vivo shRNA transfection significantly downregulated expression levels of SphK1, COX-2 and TNF-α in peritoneal macrophage (Figure 4B, P < 0.0001, P < 0.05, P < 0.05, respectively), but not other peritoneal cells (data not shown). Next, we examined the effect of targeting macrophage SphK1 on AOM-induced ACF formation. The results from ACF formation assay demonstrate that knockdown of macrophage SphK1 significantly reduced the number of ACF induced by AOM (Figure 4C, P < 0.0001), supporting that SphK1 expression in peritoneal macrophages plays an important role in ACF formation.
Figure 4.
Effect of SphK1 shRNA on SphK1 expression in peritoneal macrophages and AOM-induced ACF formation. (A) A protocol for SphK1 shRNA injection and ACF formation. Black rhombus, white rhombus, black triangle and arrow indicate SCR shRNA injection, SphK1 shRNA injection, AOM injection and euthanize, respectively. SphK1 shRNA downregulated expression levels of SphK1, COX-2 and TNF-α in peritoneal macrophages (B) and reduced the number of ACF per colon induced by AOM (C). SphK1 expression levels were expressed as MNE, which is normalized with β-actin as the reference gene. *P < 0.05 and ****P < 0.0001 versus SCR.
Effects of Exposure of peritoneal cells to sphingolipids ACF formation induced by AOM
Given that SphK1 regulates the metabolism of ceramide, sphingosine and S1P, it became important to determine if any of these lipids modulate ACF formation. C2-ceramide, sphingosine and S1P were intraperitoneally administered to WT mice, and then the mice were injected with AOM to induce ACF formation (Figure 5A). The results showed that ceramide reduced the number of ACF per colon (Figure 5B, square, P < 0.05), whereas S1P increased the number of ACF per colon induced by AOM (Figure 5B, rhombus, P < 0.01). The results suggest that ceramide may induce apoptosis or inactivation of peritoneal macrophages, whereas S1P may activate peritoneal macrophages to induce inflammatory factors, such as COX-2 and TNF-α in response to AOM injection. However, sphingosine injection demonstrated varied responses (Figure 5B, triangle). We analyzed sphingolipid levels in blood (Figure 5C–E). The injection of C2-ceramide and S1P did not change blood levels of ceramide, sphingosine and S1P. However, sphingosine-injected animals demonstrate higher blood levels of ceramide, sphingosine and S1P. The results suggest that the ceramide and S1P may stay in peritoneal cavity and directly influence peritoneal macrophages, while sphingosine may be immediately absorbed and removed from peritoneal cavity.
Figure 5.
Effect of sphingolipids injection on AOM-induced ACF formation and sphingolipid levels in blood. (A) A protocol for SphK1 sphingolipids injection and ACF formation. White rhombus, light gray rhombus, dark gray rhombus, black triangle and arrow indicate control (vehicle), C2-ceramide, sphingosine, S1P, AOM and euthanize, respectively. (B) ACF formation. S1P significantly increased the number of ACF per colon, while ceramide reduced. Sphingosine injection demonstrated varied results. Data indicates the number of ACF per colon. (C) Total ceramide levels in blood. (D) Sphingosine levels in blood. (E) S1P levels in blood. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 versus WT mice. N.S., not significant.
SphK inhibitor SKI-II prevents ACF formation induced by AOM
Finally, we examined the effect of SphK inhibition by SKI-II on AOM-induced ACF formation. SKI-II was administered by oral gavage to mice 1 h prior to AOM injection, leading to low SphK activity at the time of AOM injection (Figure 6A). First, we confirmed that the vehicle for SKI-II (PEG400) did not affect the number of ACF per colon induced by AOM. As shown in Figure 6B, SKI-II significantly reduced the number of ACF per colon induced by AOM (P < 0.001). The result suggests that SKI-II inhibited the initiation of colon carcinogenesis triggered by AOM.
Figure 6.
Effect of SKI-II on AOM-induced ACF formation. (A) A protocol for SphK1 SKI-II and ACF formation. Black rhombus, white rhombus, black triangle and arrow indicate PEG400 (vehicle as a control), SKI-II, AOM injection and euthanize, respectively. (B) ACF formation. SKI-II significantly reduced the number of ACF per colon induced by AOM. Data indicates the number of ACF per colon. ***P < 0.001 versus PEG400.
Discussion
The results from this study provide strong evidence that SphK1 expression in peritoneal macrophage is involved in the early stage of colon carcinogenesis. We first demonstrate that AOM injection activates peritoneal macrophages to induce SphK1, COX-2 and TNF-α in vivo, while AOM was not able to induce SphK1 expression in vitro. Although macrophages were cultured with other peritoneal cells (including T and B cells), AOM did not induce SphK1 expression, suggesting that the effect of AOM in macrophages may be due to systemic and non-cell-autonomous regulations. To our knowledge, this is the first report showing that AOM can acutely activate macrophages through i.p. injection. One of the known mechanisms by which AOM induces tumorigenesis is through base pair miscoding (21). In this scenario, AOM is metabolized by CYP2E1 in the liver to form metabolite methylazoxymethanol (MAM) which is then decomposed to a methyldiazonium ion, leading to base modifications. If the base modifications are not repaired prior to DNA replication, this can represent initiation of tumorigenesis. This process takes time and represents a chronic or long-term effect. AOM is also known to induce acute apoptosis in murine colon (22,23). Since previous studies have shown that DNA adduct in colon were found 6 h after exposure of mice to AOM (24), AOM-induced acute apoptosis may be due to MAM-induced DNA damage. We showed that AOM was capable of activating macrophages in 3 h, suggesting that peritoneal macrophages may be acutely activated indirectly by AOM through a MAM-independent mechanism. In addition, OpenArray analysis demonstrated that SphK1-overexpressing macrophages induce inflammation pathways, such as Signaling by Interleukins, FCERI signaling, Toll-Like receptors Cascades and MyD88-related pathways. Taken together, the results suggest that SphK1 enhances AOM-triggered colon carcinogenesis by induction of inflammation pathways.
We then provide evidence for a critical role for the activated peritoneal macrophages in colon carcinogenesis. Previously, we reported that SphK1 deficiency significantly inhibits AOM-induced colon carcinogenesis as well as inflammation-related colon carcinogenesis induced by AOM and DSS by inhibiting COX-2 expression and PGE2 production (5,6). Several studies have shown that COX-2 expression in colon tumors was limited to activated macrophages and their surroundings, suggesting that activated macrophages are involved in COX-2 related colon carcinogenesis (15,16,25). The results from this study show that inoculation of peritoneal macrophages isolated from WT mice rescued the lowered ACF formation induced by AOM in SphK1 KO mice. In addition, using shRNA in vivo transfection technology, we found that downregulation of SphK1 only in peritoneal macrophage by shRNA reduced the number of ACF per colon induced by AOM. Although it is still unclear how activated peritoneal macrophages affect colon carcinogenesis, we believe that activated peritoneal macrophages via the SphK1/S1P pathway trigger cancer initiation through a different mechanism from MAM-related pathway. We found that SphK1 may regulate COX-2 and TNF-α expression in macrophages. Numerous studies have revealed the roles of inflammatory factors, such as COX-2/PGE2 pathway (9–13) and TNF-α (26) in colon carcinogenesis. In addition, our previous work has demonstrated that the SphK1/S1P pathway regulates the COX-2/PGE2 pathway in colon cancer cells (6), and that SphK1 deficiency inhibits inflammation-related colon carcinogenesis (5). Several lines of evidence suggest that COX-2 plays a crucial role in the growth and progression of ACF formation and colorectal adenoma. In addition, COX-2 is expressed in infiltrated macrophages or in surrounding colorectal lamina propria, indicating that macrophages may be necessary in COX-2 related colon carcinogenesis. Moreover, TNF-α expression in human macrophages was discovered in the colonic tissues in both patients with Crohn’s disease and ulcerative colitis (27), and clinical studies have shown a dramatic improvement in Crohn’s disease patients treated with anti-TNF-α therapy (28). These results suggest that COX-2 and TNF-α expression in macrophages play a pivotal role in the early stages of colon carcinogenesis before developing adenomas.
We also demonstrate that intraperitoneal injection of S1P enhanced AOM-induced ACF formation. Previous studies have shown that exogenous S1P administration induces COX-2 and TNF-α expression in murine macrophages (29,30), suggesting that S1P activated peritoneal macrophages to induce COX-2 and TNF-α expression leading to the initiation of colon carcinogenesis. On the other hand, ceramide injection reduced ACF formation induced by AOM. Taken together, since SphK1 is a key regulator of the balance of ceramide and S1P (31), it is suggested that the inhibition of SphK1 decreases S1P (macrophage activator) and increases ceramide (macrophage inactivator) leading to the prevention of colon cancer.
Finally, we examined whether SphK inhibitor has an inhibitory effect on colon carcinogenesis, specifically in the early stage. We showed that SKI-II significantly reduced the number of ACF formation. Previously, we reported that SphK1 KO mice developed less number of ACF formation induced by AOM, while SphK2 KO mice developed almost same number of ACF as compared with WT mice, suggesting that SphK1 is more important than SphK2 in colon carcinogenesis (5). Thus, although SKI-II is a pan-SphK inhibitor, the result from this study indicates that inhibitory effect of SKI-II on ACF formation is most likely due to the inhibition of SphK1 in peritoneal macrophage as well as colon.
In conclusion, this study shows that the SphK1/S1P pathway plays an important role in the early stages of colon carcinogenesis via activation of peritoneal macrophages. SphK1 is necessary to signal the up-regulation of COX-2 and TNF-α when induced by AOM. Series of animal experiments demonstrate that activation of peritoneal macrophages is associated with the initiation of colon carcinogenesis. Finally, we demonstrate that inhibition of SphK1 by inhibitor attenuates colon carcinogenesis. Although continued investigation of the mechanisms involved in the results is required, these findings suggest that the SphK1/S1P pathway regulates activation of peritoneal macrophages and SphK1-related activation of macrophages may linked to the early stages of colon carcinogenesis. Taken together, SphK1 can be a potent target for colon cancer chemoprevention.
Supplementary Material
Supplementary data are available at Carcinogenesis online.
Funding
University of Hawaii Cancer Center Start-up fund to H.F., NIH grants (R01CA124687 to T.K. and P01CA97132 to L.M.O., Y.A.H. and T.K.) and a Veterans Affairs Merit Award to L.M.O.
Ethics approval and consent to participate
All mouse experiments were approved by the institutional animal care and use committee at UH (IACUC# 10–945).
Consent for publication
The manuscript does not contain any individual person’s data in any form.
Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional file.
Authors contributions
H.F. performed all experiments, conceived and was involved in all aspects of the study, P.M.Y., Y.Sh. and K.I. performed all experiments, R.P. performed additional experiments, R.C. and Y.Su. performed bioinformatics analysis for OpenArray, L.M.O., Y.A.H. and T.K. conceived and were involved in all aspects of the study.
Supplementary Material
Acknowledgement
We thank the Animal Carcinogenesis Shared Resource, University of Hawaii Cancer Center.
Conflict of Interest Statement: None declared.
Abbreviations
- ACF
aberrant crypt foci
- AOM
azoxymethane
- COX
cyclooxygenase
- FBS
fetal bovine serum
- MAM
methylazoxymethanol
- MNE
mean normalized expression
- PBS
phosphate-buffered saline
- PGE2
prostaglandin E2
- S1P
sphingosine 1-phosphate
- SCR
scramble negative control
- shRNA
short hairpin RNA
- SphK1
sphingosine kinase 1
- TNF
tumor necrosis factor
References
- 1. Siegel R.L., et al. (2016) Cancer statistics, 2016. CA. Cancer J. Clin., 66, 7–30. [DOI] [PubMed] [Google Scholar]
- 2. Ogretmen B., et al. (2004) Biologically active sphingolipids in cancer pathogenesis and treatment. Nat. Rev. Cancer, 4, 604–616. [DOI] [PubMed] [Google Scholar]
- 3. Liu H., et al. (2002) Sphingosine kinases: a novel family of lipid kinases. Prog. Nucleic Acid Res. Mol. Biol., 71, 493–511. [DOI] [PubMed] [Google Scholar]
- 4. Billich A., et al. (2005) Basal and induced sphingosine kinase 1 activity in A549 carcinoma cells: function in cell survival and IL-1beta and TNF-alpha induced production of inflammatory mediators. Cell. Signal., 17, 1203–1217. [DOI] [PubMed] [Google Scholar]
- 5. Kawamori T., et al. (2009) Role for sphingosine kinase 1 in colon carcinogenesis. FASEB J., 23, 405–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kawamori T., et al. (2006) Sphingosine kinase 1 is up-regulated in colon carcinogenesis. FASEB J., 20, 386–388. [DOI] [PubMed] [Google Scholar]
- 7. Pettus B.J., et al. (2003) The sphingosine kinase 1/sphingosine-1-phosphate pathway mediates COX-2 induction and PGE2 production in response to TNF-alpha. FASEB J., 17, 1411–1421. [DOI] [PubMed] [Google Scholar]
- 8. Pettus B.J., et al. (2005) The coordination of prostaglandin E2 production by sphingosine-1-phosphate and ceramide-1-phosphate. Mol. Pharmacol., 68, 330–335. [DOI] [PubMed] [Google Scholar]
- 9. Kawamori T., et al. (1998) Chemopreventive activity of celecoxib, a specific cyclooxygenase-2 inhibitor, against colon carcinogenesis. Cancer Res., 58, 409–412. [PubMed] [Google Scholar]
- 10. Nakatsugi S., et al. (1997) Suppression of intestinal polyp development by nimesulide, a selective cyclooxygenase-2 inhibitor, in Min mice. Jpn. J. Cancer Res., 88, 1117–1120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Steinbach G., et al. (2000) The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N. Engl. J. Med., 342, 1946–1952. [DOI] [PubMed] [Google Scholar]
- 12. Kawamori T., et al. (2003) Enhancement of colon carcinogenesis by prostaglandin E2 administration. Carcinogenesis, 24, 985–990. [DOI] [PubMed] [Google Scholar]
- 13. Hansen-Petrik M.B., et al. (2002) Prostaglandin E(2) protects intestinal tumors from nonsteroidal anti-inflammatory drug-induced regression in Apc(Min/+) mice. Cancer Res., 62, 403–408. [PubMed] [Google Scholar]
- 14. Eberhart C.E., et al. (1994) Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology, 107, 1183–1188. [DOI] [PubMed] [Google Scholar]
- 15. Bamba H., et al. (1999) High expression of cyclooxygenase-2 in macrophages of human colonic adenoma. Int. J. Cancer, 83, 470–475. [DOI] [PubMed] [Google Scholar]
- 16. Jedinak A., et al. (2010) Activated macrophages induce metastatic behavior of colon cancer cells. Immunobiology, 215, 242–249. [DOI] [PubMed] [Google Scholar]
- 17. Allende M.L., et al. (2004) Mice deficient in sphingosine kinase 1 are rendered lymphopenic by FTY720. J. Biol. Chem., 279, 52487–52492. [DOI] [PubMed] [Google Scholar]
- 18. Zhang X., et al. (2008) The isolation and characterization of murine macrophages. Curr. Protoc. Immunol., Chapter 14, Unit 14.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Muller P.Y., et al. (2002) Processing of gene expression data generated by quantitative real-time RT-PCR. Biotechniques, 32, 1372–1374, 1376, 1378. [PubMed] [Google Scholar]
- 20. Bird R.P. (1987) Observation and quantification of aberrant crypts in the murine colon treated with a colon carcinogen: preliminary findings. Cancer Lett., 37, 147–151. [DOI] [PubMed] [Google Scholar]
- 21. Nagasawa H.T., et al. (1972) Decomposition of methylazoxymethanol, the aglycone of cycasin, in D 2 O. Nature, 236, 234–235. [DOI] [PubMed] [Google Scholar]
- 22. Westley C.B., et al. (2010) Enhanced acute apoptotic response to azoxymethane-induced DNA damage in the rodent colonic epithelium by Tyrian purple precursors: a potential colorectal cancer chemopreventative. Cancer Biol. Ther., 9, 371–379. [DOI] [PubMed] [Google Scholar]
- 23. Kwon Y., et al. (2007) Aging alters acute apoptotic response to azoxymethane in the colon of rats. Exp. Gerontol., 42, 1154–1161. [DOI] [PubMed] [Google Scholar]
- 24. Rosenberg D.W., et al. (2009) Mouse models for the study of colon carcinogenesis. Carcinogenesis, 30, 183–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Tanaka S., et al. (2006) Monocyte chemoattractant protein 1 and macrophage cyclooxygenase 2 expression in colonic adenoma. Gut, 55, 54–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Popivanova B.K., et al. (2008) Blocking TNF-alpha in mice reduces colorectal carcinogenesis associated with chronic colitis. J. Clin. Invest., 118, 560–570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Stucchi A., et al. (2006) A new transcription factor that regulates TNF-alpha gene expression, LITAF, is increased in intestinal tissues from patients with CD and UC. Inflamm. Bowel Dis., 12, 581–587. [DOI] [PubMed] [Google Scholar]
- 28. Yamamoto-Furusho J.K. (2007) Innovative therapeutics for inflammatory bowel disease. World J. Gastroenterol., 13, 1893–1896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Hammad S.M., et al. (2008) Dual and distinct roles for sphingosine kinase 1 and sphingosine 1 phosphate in the response to inflammatory stimuli in RAW macrophages. Prostaglandins Other Lipid Mediat., 85, 107–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lee H., et al. (2002) Lysophospholipid regulation of mononuclear phagocytes. Biochim. Biophys. Acta, 1582, 175–177. [DOI] [PubMed] [Google Scholar]
- 31. Hannun Y.A., et al. (2008) Principles of bioactive lipid signalling: lessons from sphingolipids. Nat. Rev. Mol. Cell Biol., 9, 139–150. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets supporting the conclusions of this article are included within the article and its additional file.