Comments
Abstract
Rationale: Metabolic alterations contribute to cancer development and progression. However, the molecular mechanisms relating metabolism to cancer metastasis remain largely unknown.
Objectives: To identify a key metabolic enzyme that is aberrantly overexpressed in invasive lung cancer cells and to investigate its functional role and prognostic value in lung cancer.
Methods: The differential expression of metabolic enzymes in noninvasive CL1-0 cells and invasive CL1-5 cells was analyzed by a gene expression microarray. The expression of target genes in clinical specimens from patients with lung cancer was examined by immunohistochemistry. Pharmacologic and gene knockdown/overexpression approaches were used to investigate the function of the target gene during invasion and metastasis in vitro and in vivo. The association between the target gene expression and clinicopathologic parameters was further analyzed. Bioinformatic analyses were used to discover the signaling pathways involved in target gene-regulated invasion and migration.
Measurements and Main Results: Squalene synthase (SQS) was up-regulated in CL1-5 cells and in the tumor regions of the lung cancer specimens. Loss of function or knockdown of SQS significantly inhibited invasion/migration and metastasis in cell and animal models and vice versa. High expression of SQS was significantly associated with poor prognosis among patients with lung cancer. Mechanistically, SQS contributed to a lipid-raft-localized enrichment of tumor necrosis factor receptor 1 in a cholesterol-dependent manner, which resulted in the enhancement of nuclear factor-κB activation leading to matrix metallopeptidase 1 up-regulation.
Conclusions: Up-regulation of SQS promotes metastasis of lung cancer by enhancing tumor necrosis factor-α receptor 1 and nuclear factor-κB activation and matrix metallopeptidase 1 expression. Targeting SQS may have considerable potential as a novel therapeutic strategy to treat metastatic lung cancer.
Cholesterol metabolic reprogramming is now considered a hallmark of cancer. However, little is known about the metabolic alterations necessary for supporting cancer progression.
In this study, we identify the importance of squalene synthase (SQS), an enzyme of cholesterol biosynthesis pathway, in lung cancer and show that it may promote lung cancer invasion and metastasis. Our data not only identify the prognostic value of SQS in patients with lung cancer but also highlight the potential of targeting SQS as a novel therapeutic strategy for inhibiting lung cancer invasion and metastasis.
Lung cancer is the leading cause of cancer-related death worldwide, and non–small cell lung cancer (NSCLC) accounts for up to 80% of those cases. Patients with lung cancer often have poor prognoses and low 5-year survival rates (1, 2). The high mortality rate of this disease is mostly caused by its high metastatic rate (3). Recently, metabolic reprogramming (4) has been identified as a major hallmark of cancer that plays an important role in the malignant properties of cancer cells (5). In addition to driving the malignant phenotype of cancer cells, metabolism contributes to signal transduction, tumorigenesis, and metastasis (6–9). However, how cancer cells take advantage of bioenergetics to regulate the malignant phenotype has not been clearly elucidated.
In this context, cancer cells demonstrate increased concentrations of cholesterol metabolites (8, 9), and targeting cholesterol has been beneficial in reducing cancer-related mortality in various types of cancers, including lung cancer (10, 11). Cholesterol is a major structural component of lipid rafts (rafts) and functions to modulate membrane fluidity. Cells obtain cholesterol from the circulatory system and by de novo synthesis via the mevalonate (MVA) pathway (12, 13). MVA pathway genes are frequently up-regulated in cancer, and the pharmacologic depletion of cholesterol suppresses invasion, indicating that cholesterol plays an important role in cancer progression and prolongs survival in patients with lung cancer (10, 14). Rafts are cholesterol-enriched membrane microdomains within the plasma membrane. Their functions include signaling molecule assembly, membrane fluidity regulation, and membrane protein or receptor exchange (15–17). In signal transduction processes, rafts serve as physical platforms for the coordination of diverse molecules; for example, rafts can concentrate receptors for ligand binding (18, 19). Furthermore, the abnormal accumulation of cholesterol is associated with tumor growth and survival (20, 21).
In the present study, we investigated whether cholesterol biosynthesis affected the mechanisms governing cancer cell migration/invasion and metastasis in vivo. We focused on squalene synthase (SQS), the enzyme responsible for the committed step of the MVA pathway for cholesterol biosynthesis, because the inhibition of SQS decreases the synthesis of cholesterol in rafts (21). Some of the results of these studies have been previously reported in the form of an abstract (22, 23).
Immunohistochemistry (IHC) staining by Ventana IHC staining system (Ventana, Tucson, AZ) was used to visualize the IHC staining signal. IHC was performed using the primary antibodies (see Table E10 in the online supplement). Other materials and methods are provided in the online supplement.
To identify metabolic enzymes associated with lung cancer progression, we analyzed genes that were differentially expressed (twofold change) between the highly invasive CL1-5 lung cancer cells and their relatively noninvasive counterpart, CL1-0 (GSE7670). By filtering the gene signature for enzyme annotation, we identified 372 probes putatively associated with lung cancer invasion (see Table E1). Based on an ingenuity pathways analysis, several biologic pathways involved in cholesterol biosynthesis were significantly enriched in the CL1-5 cells (Figure 1A). Eight genes that encode enzymes for de novo cholesterol biosynthetic pathways were up-regulated in CL1-5 cells: (1) hydroxymethylglutaryl-CoA synthase 1 (HMGCS1); (2) farnesyl-diphosphate synthase (FDPS); (3) farnesyl-diphosphate farnesyltransferase (FDFT1, also known as SQS); (4) squalene epoxidase (SQLE); (5) methylsterol monooxygenase 1 (MSMO1); (6) cytochrome P-450, family 51, subfamily A (CYP51A1); (7) 7-dehydrocholesterol reductase (DHCR7); and (8) 24-dehydrocholesterol reductase (DHCR24) (Figure 1B).
Figure 1. Identification of squalene synthase (SQS) as a potential target for inhibiting lung cancer invasion/migration. (A) Pathway analysis of genes differentially expressed (twofold change) between CL1-5 and CL1-0 from the GSE7670 dataset filtered by enzyme annotation. (B) Schematic diagram of up-regulated (red) and inhibited (blue) cholesterol biosynthetic enzymes in CL1-5. (C) Kaplan-Meier analysis of overall survival using publicly available lung cancer microarray datasets, stratified according to HMGCS1, HMGCR, FDPS, FDFT1 (SQS), FNTA, and PGGT1B expression. HR = hazard ratio. Effects of mevastatin (D) and zaragozic acid A (E) on the invasion/migration capabilities of lung cancer cells. A549 and CL1-5 cells were treated with mevastatin or zaragozic acid A at the indicated concentrations and subjected to invasion and migration assays. DMSO = dimethyl sulfoxide. (F) A549 and CL1-5 cells were infected with the indicated virus in culture plates. Reverse-transcriptase polymerase chain reaction analysis of HMGCS1 and HMGCR expression in A549 and CL1-5 cells after virus infection. GAPDH = glyceraldehyde phosphate dehydrogenase. (G) Effects of HMGCS1 and HMGCR knockdown on the invasion/migration capabilities of A549 and CL1-5 cells. Data are presented as the means ± SD; **P < 0.01.
[More] [Minimize]By contrast, FNTA and PGGT1B, which encode the nonsterol branch enzymes farnesyltransferase, CAAX Box, α and protein geranylgeranyltransferase type I, β subunit were not significantly up-regulated in CL1-5 cells (Figure 1B; see Table E1). Moreover, Kaplan-Meier survival analyses using publicly available lung cancer microarray datasets (24) demonstrated that high expression levels of HMGCS1, SQS, SQLE, CYP51A1, and DHCR7 were significantly associated with worse overall survival, whereas FDPS, MSMO1, DHCR24, FNTA, and PGGT1B expression levels showed little or no significant association (Figure 1C; see Figure E1A). These results indicated that the up-regulation of HMGCS1, FDPS, SQS, SQLE, CYP51A1, MSMO1, DHCR7, and DHCR24 in CL1-5 cells might contribute to the enhancement of cellular cholesterol synthesis and cancer metastasis.
To elucidate the functional roles of these upstream enzymes on the invasive capability of lung cancer cells, highly invasive CL1-5 and A549 cells were treated with mevastatin, zaragozic acid A, or shRNA to inhibit hydroxymethylglutaryl coenzyme A reductase (HMGCR), SQS, and HMGCS1, respectively. Inhibition of SQS activity, but not HMGCS1 or HMGCR, significantly inhibited the mobility and invasiveness of A549 and CL1-5 cells (Figures 1D–1G). We next determined whether cholesterol correlated with the invasive capability of lung cancer cells. Inhibitors or shRNA targeting the downstream enzymes SQLE, CYP51A1, MSMO1, and DHCR7 significantly inhibited the mobility and invasiveness of A549 and CL1-5 cells (see Figures E1B–E1D). These results suggested that cholesterol is potentially involved in the process of lung cancer progression. Furthermore, SQS is the first enzyme of the sterol branch, and targeting SQS by zaragozic acid A significantly inhibited lung cancer cell invasiveness. These results suggested that SQS might play an important role in the invasion and metastasis of lung cancer.
To examine the expression levels of SQS in human NSCLC tissues, tissue microarrays containing samples from 53 patients with stage I-II lung cancer and 82 patients with stage III-IV lung cancer were immunostained for SQS. Detailed clinicopathologic information including demographics, tumor histology, pathologic stage, and survival times was obtained for all patients (see Table E2). In 27 of the 30 patients, SQS was significantly up-regulated in the tumor compared with corresponding normal tissue from the same patient (Figures 2A and 2B; P < 0.01). As shown in Figure 2C, the expression of SQS was scored according to its cytoplasmic staining intensity on a scale from 0 to 3. Using the defined scoring criteria, we stratified patients into an SQS-high group (scores 2 and 3) and an SQS-low group (scores 0 and 1) and found that SQS expression level was significantly associated with advanced stage (stage III-IV; P < 0.01), tumor size (pT; P = 0.003), regional lymph node metastasis (pN; P = 0.02), distal metastasis (pM; P = 0.043), and recurrence (P = 0.035) (Figure 2D; see Table E3).
Figure 2. Squalene synthase (SQS) is overexpressed in lung cancer and correlated with poor survival. (A) Representative images of SQS expression in paired specimens of lung cancer and corresponding normal tissues. Scale bar: 100 μm. (B) Quantification of SQS in 30 NT-paired lung cancer specimens; **P < 0.01. (C) Representative images of SQS expression scores in lung cancer tissues. The scores are calculated as the staining intensity × the percentage of stained cells. Scale bar: 100 μm. (D) Quantification of SQS expression by immunohistochemistry analysis of lung cancer specimens. The number (n) of samples for each stage is indicated at the top of each column; *P < 0.05. (E) Kaplan-Meier analysis of disease-free survival and overall survival of 135 patients with lung cancer stratified by SQS level. (F) Multivariate Cox regression hazard ratio for risk of death in patients with lung cancer. DFS = disease-free survival; HR = hazard ratio; OS = overall survival.
[More] [Minimize]Furthermore, a survival analysis revealed that a high level of SQS was significantly associated with a shorter disease-free survival and overall survival compared with patients with low SQS expression (Figure 2E; P < 0.01). Moreover, this result was consistent with survival analyses conducted using publicly available lung cancer microarray datasets, which demonstrated that high expression of SQS mRNA predicted worse overall and first-progression survival (Figure 1C; see Figure E2). Univariate and multivariate analyses of overall survival and disease-free survival revealed that SQS expression level was a significant, independent predictor of survival for patients with lung cancer (Figure 2F; see Table E4).
We next investigated the effects of SQS perturbation on the invasion, migration, and proliferation capabilities of lung cancer cells. Western blot analysis showed that endogenous SQS was up-regulated in highly invasive lung cancer cells (Figures 3A and 3B). Knockdown of SQS in the highly invasive A549 and CL1-5 lung cancer cells significantly reduced their invasion/migration capabilities compared with nonsilencing control cells (Figure 3C; see Figures E3A and E3B). Complementarily, enforced expression of SQS in the poorly invasive CL1-0 and H1355 cells significantly enhanced their migration/invasion capabilities (Figure 3D; see Figures E3C and E3D). Neither overexpression nor knockdown of SQS in lung cancer cells affected their proliferation potential (see Figures E3E and E3F).
Figure 3. Squalene synthase (SQS) enhances invasion/migration and metastasis in cell and animal models. (A) Western blot analysis of endogenous SQS in human lung cancer cell lines. Bottom, Invasive capabilities of lung cancer cell lines. (B) Correlation between SQS expression and invasion capability in lung cancer cells. (C) SQS knockdown and the invasion/migration capabilities of A549 and CL1-5 lung cancer cells. Top, Western blot analysis of SQS after lentiviral-mediated RNAi. Bottom, Invasion (open) and migration capabilities (filled) of A549 and CL1-5 cells infected with shSQS or nonsilencing (NS) shRNA. Data are presented as the means ± SD; **P < 0.01. (D) SQS overexpression and invasion/migration potential of CL1-0 and H1355 cells. Top, Western blot analysis of SQS expression of CL1-0 and H1355 cells overexpressing SQS. Bottom, Invasion (open) and migration (filled) capabilities of CL1-0 and H1355 cells overexpressing SQS. Data are presented as the means ± SD; **P < 0.01. (E) Left, Western blot analysis of SQS in CL1-5 cells stably expressing the NS shRNA or SQS shRNA2. Middle, Representative lung hematoxylin and eosin (H&E) images of mice intravenously injected with CL1-5/NS shRNA cells or CL1-5/SQS-shRNA2 (five mice per group); the arrows indicate the metastatic nodules. Right, Quantification of metastatic lung nodules in individual mice 4 weeks after intravenous injection of CL1-5 cells infected with NS shRNA (NS) or SQS shRNA2 (S2). Data are presented as the means ± SD; *P < 0.05. (F) Intravenous model, SQS overexpression and metastasis in vivo of CL1-0 (left) and H1355 (right) cells. Left, Expression of SQS protein as examined by Western blotting in CL1-0 cells stably expressing the vector or SQS. Left microphotographs, Representative lung H&E images of mice intravenously injected with CL1-0/SQS cells or CL1-0/vector (three mice per group). Bottom, Summary numbers of mice bearing lung lesions following injection with empty vector or SQS-expressing CL1-0 cells. Right, Representative H1355 model in vivo. (G) Orthotopic model, SQS overexpression and metastasis in vivo of CL1-0 (left) and H1355 (right) cells. Left, Representative H&E lung images of mice orthotopically injected with empty vector or SQS-expressing CL1-0 cells (three mice per group). Arrows indicate the CL1-0/vector or CL1-0/SQS tumor that was established in the left lung. The arrowhead indicates an intrapulmonary metastatic tumor nodule on the right lung. Scale bars: 50 mm (field ×12.5) and 20 μm (field ×400). Bottom, Summary numbers of mice bearing metastases following injection with empty vector or SQS-expressing CL1-0 cells. Right, Representative H1355 model in vivo.
[More] [Minimize]We next evaluated the effects of SQS on metastasis in vivo. As shown in Figure 3E (left), we stably knocked down SQS in CL1-5 cells and intravenously injected those cells into the lateral tail veins of mice. Four weeks after injection, the lungs of the mice were removed and examined for metastases. SQS knockdown resulted in fewer detectable lung metastases compared with the control cells (Figure 3E, middle). Quantification of the metastatic nodules using dissecting microscopy and histologic analysis of the lungs from each mouse confirmed that the number of lung metastases was significantly reduced in the mice carrying SQS-knockdown tumors (Figure 3E, right). In a comparative experiment, we also examined the effect of SQS overexpression in CL1-0 and H1355 on metastasis after intravenous injection and orthotopic (left lung) implantation into mice. In the intravenous-injection model, SQS overexpression resulted in an increase in the number of detectable metastatic lung foci compared with the control animals (Figure 3F). A similar result was observed in the orthotopic model, in which tumor metastases were increased in the right lung of the mice carrying the primary tumors generated from SQS-overexpressing CL1-0 and H1355 cells (Figure 3G).
To identify the molecular mechanisms potentially regulated by SQS, we used a gene expression microarray to examine control cells (CL1-0/pCMV-SPORT6) and SQS-overexpressing CL1-0 cells. Overexpression of SQS resulted in the up-regulation of 174 genes and the down-regulation of 199 genes, with a 1.5-fold change cut-off value (see Tables E5 and E6). A gene-annotation enrichment analysis predicted that tumor necrosis factor (TNF) signaling was activated on SQS overexpression (see Figures E4A and E4B and Table E7). We therefore validated that SQS-promoted invasion/migration depends on TNF receptor 1 (TNFR1). As shown in Figure 4A, the silencing of TNFR1 caused a significant inhibition of migration and invasion activity in vitro in SQS-overexpressing cells. These data suggested that SQS-promoted invasion was mediated through TNFR1. It has been shown that TNFR1 activation is required for TNF-α–mediated nuclear factor (NF)-κB activation (25). We therefore focused on whether SQS could activate NF-κB. The NF-κB reporter assay demonstrated that NF-κB activity was significantly up-regulated on SQS overexpression (Figure 4B). However, IκB-α expression was markedly decreased in SQS-overexpressing cells, which was similar to the effect of TNF-α treatment in the CL1-0 vector control cells (see Figure E4C). Moreover, knockdown of SQS in CL1-5 cells significantly increased IκB-α expression (see Figure E4D). These data suggested that up-regulation of SQS promoted invasion through TNFR1–NF-κB signaling.
Figure 4. Squalene synthase (SQS) promotes invasion/migration though matrix metallopeptidase 1 (MMP1). (A) Tumor necrosis factor receptor 1 (TNFR1) silencing and the invasion/migration capabilities of CL1-0/SQS lung cancer cells. Top, Western blot analysis of SQS and TNFR1 following transfection with TNFR1 siRNA. Bottom, Invasion (open) and migration (filled) capabilities of CL1-0/SQS cells after transfection with TNFR1 siRNA. Data are presented as the means ± SD; **P < 0.01. (B) Top, Nuclear factor (NF)-κB activity reporter assay in CL1-0 cells. The SQS vector or empty vector was transfected into CL1-0 cells. Bottom, NF-κB activity following TNF-α treatment in CL1-0 cells. The empty vector was transfected as the negative control, and GFP provided an internal control. After 72 hours, green fluorescence was detected in the cells, followed by treatment with firefly luciferase to determine luminescence. Data are presented as the means ± SD; **P < 0.01. (C) Expression of MMP1 and SQS mRNA in A549 cells following SQS knockdown compared with the nonsilencing (NS) shRNA control. (D) Left, MMP1 promoter activity assay in CL1-0/SQS cells. CL1-0/vector cells were used as controls, and GFP was an internal control. After 48 hours, green fluorescence was detected in the cells, followed by treatment with firefly luciferase for luminescence determination. Right, Expression of MMP1 and SQS mRNA in CL1-0/SQS cells. (E) Expression of MMP1 mRNA in cells overexpressing SQS compared with the vector control. Top, Reverse-transcriptase polymerase chain reaction analysis of SQS and MMP1 expression in CL1-0/SQS cells after MMP1 knockdown. Bottom, Invasion and migration capabilities of CL1-0/SQS cells after MMP1 knockdown. Data are presented as the means ± SD; **P < 0.01. (F) Top, Reverse-transcriptase polymerase chain reaction analysis of SQS and MMP1 expression in CL1-0/SQS cells after treatment with the NF-κB inhibitor, Bay 11-7082. Bottom, Invasion and migration capabilities of CL1-0/SQS cells after NF-κB inhibitor treatment. Data are presented as the means ± SD; **P < 0.01. (G) Representative images of SQS and MMP1 immunohistochemistry staining in serial sections of lung cancer patient tissues. (H) Quantification of SQS and MMP1 expression by immunohistochemistry analysis of lung cancer specimens. The number (n) of samples is indicated at the top of each column. Data are presented as the means ± SD; *P < 0.05. GAPDH = glyceraldehyde phosphate dehydrogenase; GFP = green fluorescent protein.
[More] [Minimize]To identify SQS-regulated downstream NF-κB targets involved in the invasion/migration process, we further analyzed the microarray data of the SQS-overexpressing CL1-0 cells. According to the gene ontology analysis, several biologic processes predicted to be involved in invasion and migration were enriched (see Table E8). We focused on genes that had been previously reported as transcriptional targets of NF-κB. Matrix metallopeptidase 1 (MMP1) was identified because of its involvement in the process of metastasis. It has been reported that the expression of MMP1 can be regulated by NF-κB and is associated with the development and progression of lung cancer (see Figure E4B) (26–28). Reverse-transcriptase polymerase chain reaction demonstrated that MMP1 expression was significantly down-regulated on SQS knockdown (Figure 4C). In addition, promoter assay and reverse-transcriptase polymerase chain reaction confirmed that MMP1 was up-regulated in SQS-overexpressing CL1-0 cells compared with the vector control cells (Figure 4D). We also observed a significant inhibition of migration and invasion activity by suppressing MMP1, via shRNA or an NF-κB inhibitor (Bay11–7082), in SQS-overexpressing cells (Figures 4E and 4F).
We next investigated whether SQS expression correlated with MMP1 expression in patients with lung cancer. Representative IHC staining for SQS and MMP1 on serial sections of clinical lung cancer specimens revealed positive correlative staining patterns in lung tumor tissues (Spearman nonparametric correlation test; correlation coefficient = 0.217; P = 0.028; n = 104) (Figures 4G and 4H; see Table E9). Moreover, a correlation analysis using publicly available lung cancer microarray datasets also showed similar results (see Figure E4E).
We next studied the mechanism of SQS-enhanced TNFR1 signaling and determined that overexpression of SQS did not affect the mRNA levels of TNFR1 in CL1-0 cells, indicating that SQS did not transcriptionally regulate TNFR1 expression (Figure 5A). Because SQS is the primary enzyme controlling cellular cholesterol biosynthesis and localization of TNFR1 in the cholesterol-enriched rafts of the plasma membrane contributes to its trimerization, we hypothesized that the up-regulation of SQS might increase the localization of TNFR1 in the rafts. As expected, the knockdown of SQS reduced cholesterol biosynthesis, whereas it was significantly enhanced with the overexpression of SQS, relative to vector controls (Figure 5B). The knockdown of SQS reduced rafts staining intensity, as evaluated using a red fluorescently labeled cholera toxin subunit B, which binds to the pentasaccharide chain of the plasma membrane ganglioside GM1 (Figure 5C). Complementarily, overexpression of SQS in CL1-0 cells significantly induced the clustering of rafts (Figure 5D). Functional studies showed that treatment with methyl-β-cyclodextrin, a detergent that specifically depletes cholesterol, reduced the invasion and migration capabilities of A549 and CL1-5 cells (Figure 5E; see Figures E5A and E5B). Moreover, cholesterol replenishment restored the migratory capability of SQS-knockdown cells (Figure 5F).
Figure 5. Squalene synthase (SQS) facilitates the enrichment of tumor necrosis factor receptor 1 (TNFR1) to lipid rafts to enhance lung cancer invasion/migration. (A) reverse-transcriptase polymerase chain reaction analysis of SQS and TNFR1 expression in CL1-0/SQS cells. (B) Changes in the total cellular cholesterol level of A549 (left) and CL1-5 (middle) on SQS silencing. Right, Total cellular cholesterol level of CL1-0 cells detected on SQS overexpression. NS = nonsilencing. Data are presented as the means ± SD; **P < 0.01; *P < 0.05. (C) Left, A549 and CL1-5 cells were infected with shSQS or NS shRNA followed by staining with the cholera toxin B subunit to identify lipid rafts (red fluorescence). Right, Quantification of fluorescence intensity by LAS AF. Data are presented as the means ± SD; **P < 0.01. (D) Top, CL1-0 cells were transfected with a control (vector) or SQS overexpression plasmid followed by staining with the cholera toxin B subunit to recognize rafts (red). Staining with anti-SQS antibody ensued using a CY3-conjugated anti–rabbit Ig antibody (green). The rafts and SQS areas of colocalization are in yellow. Bottom, Quantification of fluorescence intensities by LAS AF. Scale bars: 10 μm. Data are presented as the means ± SD; **P < 0.01. (E) Treatment with or without methyl-β-cyclodextrin (MβCD) (10 mM) and the invasion/migration potential of A549 and CL1-5 cells. DMSO = dimethyl sulfoxide. Data are presented as the means ± SD; **P < 0.01. (F) Effects of cholesterol replenishment on the migration activity of SQS-silenced A549 and CL1-5 cells. SQS-silenced cells were incubated in the presence or absence of cholesterol and subjected to a migration assay. Data are presented as the means ± SD; **P < 0.01. (G) The relative expression of SQS, TNFR1, caveolin-1 (Cav-1), and Fyn from Triton-soluble (S) and Triton-insoluble (I) protein from cells overexpressing SQS compared with the vector control (Raft/Cav-1 on I-fraction). (H) Empty-vector or SQS-overexpressing CL1-0 cells were incubated with an anti-TNFR1 antibody followed by a CY3-conjugated anti–rabbit Ig antibody (green) and stained with the cholera toxin B subunit to recognize rafts (red). The colocalization of TNFR1 with rafts is in yellow. Scale bar: 10 μm. (I) Effects of cholesterol replenishment on the TNFR1 enrichment of SQS-silenced CL1-5 cells. SQS-silenced cells were incubated in the presence or absence of cholesterol at the designated concentrations for 1 hour and subjected to Western blotting analysis. (J) Nuclear factor (NF)-κB activity reporter assay in CL1-5 cells. The shSQS or NS shRNA was infected into CL1-5 cells. The NS shRNA was infected as the negative control, and renilla signals provided an internal control. After 48 hours, SQS-silenced cells were incubated in the presence or absence of cholesterol at the designated concentrations for 1 hour and subjected to firefly signal was detected in the cells, followed by treatment with StopandGlo reagent to determine renillia signal. Data are presented as the means ± SD; **P < 0.01; *P < 0.05. (K) Models of SQS-induced NF-κB activation via enrichment of TNFR1 to lipid rafts leading to subsequent matrix metallopeptidase 1 (MMP1) up-regulation to promote invasion and metastasis in lung cancer. GAPDH = glyceraldehyde phosphate dehydrogenase.
[More] [Minimize]To demonstrate the localization of TNFR1 in the rafts, total protein from SQS-overexpressing CL1-0 cells was fractionated into soluble and insoluble (raft-associated proteins) fractions. Western blot analysis showed that TNFR1 was enriched in the insoluble cell lysate fraction isolated from SQS-overexpressing CL1-0 cells compared with that from vector control cells (Figures 5G and 5H), whereas TNFR2 was not detected (data not shown). In a complementary experiment, Figure 5I showed knockdown of SQS in CL1-5 cells significantly reduced TNFR1 expression in the insoluble fraction (I) compared with soluble fraction (S), whereas cholesterol replenishment restored TNFR1 expression. In addition, SQS knockdown inhibited NF-κB activity and cholesterol replenishment restored NF-κB activity (Figure 5J). Furthermore, knockdown of SQS also increased IκB-α expression in CL1-5 cells (see Figure E4D). Taken together, our data suggest that SQS overexpression enriches TNFR1 rafts localization, enhances the activation of TNFR1-NF-κB signaling, and increases MMP1 expression, thereby contributing to lung cancer metastasis (Figure 5K).
To evaluate whether the targeting of SQS could be a potential therapeutic strategy to inhibit lung cancer metastasis, the antimetastatic effect of zaragozic acid A was examined in an animal model using GFP- and luciferase-labeled CL1-5 cell (CL1-5-GL) transplantation. The CL1-5-GL cells were intravenously injected into the lateral tail vein of mice, and the mice were subsequently treated with zaragozic acid A (0.1 mg/kg/2 d, 0.3 mg/kg/2 d, or 1 mg/kg/2 d; intraperitoneal injection) (Figure 6A). The mice that received zaragozic acid A at 1 mg/kg displayed the lowest tumor burden (Figure 6A, left). The tumor burden was monitored and quantified according to bioluminescence intensity using an in vivo imaging system (IVIS) (Figure 6A). The lungs were removed and examined for metastases 20 days after injection, and the mice receiving zaragozic acid A at 1 mg/kg had fewer metastatic nodules than those receiving only saline (Figure 6B). The quantitative data confirmed the therapeutic effect of zaragozic acid A at a 1 mg/kg concentration on reducing the level of photon radiance, lung weight, and number of surface lung metastases (Figures 6B and 6C). A similar result was also observed in the A549-GL model (see Figure E6). Taken together, these data suggest that SQS might be a crucial mediator and a potential therapeutic target of lung cancer metastasis.
Figure 6. Zaragozic acid A treatment suppressed metastasis in mouse lung cancer models. (A) CL1-5-GL cells were intravenously injected into NOD-SCID mice that were treated over an interval of 1 day with saline only (S) and zaragozic acid A: 0.1 mg/kg; 0.3 mg/kg; 1 mg/kg. Luminescence was measured using a noninvasive, bioluminescence imaging system (IVIS spectrum) at Days 1 (left) and 20 (middle). Tumor growth is expressed as the bioluminescence intensity (BLI) change (five mice per group) (right). Data are presented as the means ± SE; **P < 0.01. (B) Bright-field imaging, luminescence, fluorescence, bright-field imaging (formalin-fixed), and hematoxylin and eosin staining in the lungs of mice treated with saline or zaragozic acid A at 0.1 mg/kg, 0.3 mg/kg, or 1 mg/kg at Day 20 after intravenous injection. The arrowheads indicate metastatic tumor nodules on the lung. Scale bars: 50 mm (field ×12.5) and 20 mm (field ×40). (C) Representative tumor weight (left), number of metastatic nodules (middle), and BLI (right) in the lung of mice treated with saline or zaragozic acid A at 0.1 mg/kg, 0.3 mg/kg, or 1 mg/kg at Day 20 after intravenous injection. Data are presented as the means ± SD and means ± SE (for BLI); **P < 0.01.
[More] [Minimize]It has been shown that aberrant activation of EGFR, KRAS, and ALK are important driver mutations for lung adenocarcinoma (AdC) development. To elucidate the effects of those oncogene mutations on SQS expression, we analyzed lung AdC microarray datasets (GSE31210) that contain 127 tissues with EGFR mutation, 120 tissues with KRAS mutation, 11 tissues with ALK fusion, and 68 EGFR/KRAS/ALK wild-type tissues. The results showed SQS mRNA was significantly up-regulated in patients with KRAS mutation compared with patients with EGFR/KRAS/ALK wild-type (see Figure E7A). In addition, we also analyzed the integrated genomics data of 230 lung AdC from TCGA using cBio portal (29, 30). The data showed SQS levels have no significant differences in p53 mutation and LKB1 mutation patient populations compared with wild-type population (see Figure E7B).
Here, we identify that enzymes involved in cholesterol biosynthesis are enriched in the highly invasive CL1-5 lung cancer cell line. In particular, SQS, a determinant enzyme in de novo cholesterol biosynthesis, is capable of potently promoting the metastasis of lung cancer cells. A clinicopathologic analysis indicated that the overexpression of SQS is significantly associated with a poorer outcome in clinical patients with lung cancer. Moreover, our data revealed that the enhancement of SQS-induced cholesterol biosynthesis promotes the formation of rafts that are loaded with an enriched TNFR1, which in turn induces the NF-κB–mediated up-regulation of MMP1, a critical protease for metastatic lung cancer (Figure 5K). In metastatic CL1-5 lung cancer cells, either pharmaceutical inhibition or knockdown of SQS dramatically suppressed invasiveness in vitro and lung colonization in vivo. These findings suggest that SQS is likely a viable drug target for combating malignant lung cancer.
Cancer cells display high rates of de novo lipogenesis in order to continually provide building blocks for maintaining cellular functions that contribute to the malignant phenotype of cancer, such as energy homeostasis, membrane synthesis, lipid raft formation, and steroid hormone synthesis (8, 9, 31). Consistent with this notion, MVA pathway, lipid metabolism genes, and atherosclerosis-related genes were frequently up-regulated in cancers and played critical roles in cancer progression (32–34). HMGCR, an enzyme that regulates MVA biosynthesis in lipid metabolism, has been identified as an oncogene and seems to be up-regulated in tumorigenesis and associated with rat sarcoma viral oncogene homolog (RAS) activation (35). Statins, which target HMGCR to reduce cholesterol biosynthesis and protein prenylation, have been used as antitumor agents in various types of malignancies (10, 11, 36, 37). However, side effects including myalgias, abdominal pain, nausea, and hepatotoxicity might limit the use of these statins therapeutically (38–40). Because aberrant increases in cholesterol (8, 9, 41, 42) have been linked to an increased risk of cancer progression, the development of new cholesterol-lowering medications is urgently needed. Here, we find that targeting SQS by zaragozic acid A effectively inhibits lung cancer cell invasion, whereas targeting HMGCR with mevastatin or enzymes downstream of SQS in cholesterol biosynthesis using terbinafine and AY 9944 only slightly or moderately affects the invasive activity of lung cancer cells in vitro (Figures 1D and 1E; see Figure E1B). In addition, the knockdown of SQS inhibits, whereas the overexpression of SQS promotes, the invasive and metastatic activities of lung cancer cells by modulating the de novo biosynthesis of cholesterol. Based on these findings, we suggest the potential of SQS as a good therapeutic target for combating metastatic lung cancer by lowering cholesterol biosynthesis. In contrast to HMGCR, SQS is a highly specialized enzyme for cholesterol biosynthesis; therefore, the targeting of SQS is likely to be relatively safe compared with the targeting of HMGCR.
Lipid raft formation is regulated by cholesterol content (20, 43, 44) and facilitates the membrane trafficking of cholesterol-binding proteins (45). The suppression of SQS in prostate cancer has been shown to reduce the level of raft-associated cholesterol, leading to reduced cell proliferation but elevated cell death (21). Moreover, cholesterol depletion seems to induce CD44 shedding and delocalization of the focal adhesion complex from rafts, thereby suppressing tumor cell migration (46, 47). Here, we show that the enforced expression of SQS facilitates TNFR1 enrichment into rafts, because of an increased level of cholesterol, sequentially leading to NF-κB activation and MMP1 up-regulation in poorly metastatic CL1-0 cells in the absence of an extracellular TNF-α supplement (see Figure E4C).
Conversely, our data revealed that the knockdown of SQS reduces TNFR1 enrichment in rafts because of insufficient cholesterol content rather than the inefficient expression of TNFR1, resulting in poor NF-κB activation and MMP1 up-regulation in the highly metastatic CL1-5 cells. This defect can be recovered by supplying extracellular cholesterol, indicating that in lung cancer cells cholesterol-mediated raft formation modulates TNFR1 enrichment and promotes cancer metastasis through TNFR1-induced NF-κB activation and MMP1 up-regulation in a TNF-α–independent manner. Although we cannot exclude possible contributions from intermediates and metabolic enzymes downstream of SQS in cholesterol biosynthesis and cancer metastasis, this is the first study to demonstrate that SQS overexpression promotes the metastatic activity of lung cancer cells by promoting TNFR1-mediated NF-κB activation and MMP1 up-regulation, because of the deposition of TNFR1 in cholesterol-enriched rafts.
Accumulating evidence has shown EGFR, KRAS, and ALK mutations are frequently detected in lung AdC in a mutually exclusive manner (48). By analyzing microarray datasets, we found SQS levels were significantly up-regulated in KRAS mutation patients compared with wild-type patients. Oncogenic KRAS mutation has been shown to promote metabolic reprogramming toward enhanced glucose uptake, increased glutamine demands for biosynthetic pathways, and decreased Krebs cycle activity (49). This metabolic adaptation provides advantages to shunt glucose-derived acetyl-CoA to either cholesterol biosynthesis or fatty acid synthesis. In an inducible KrasG12D pancreatic ductal AdC mouse model, inactivation of KrasG12D significantly down-regulates the transcripts encoded for enzymes in cholesterol biosynthesis pathway through inhibiting MAPK signaling (50). Further studies are needed to investigate the mechanisms underlying KRAS mutation and the subsequent effects of SQS-increased cholesterol levels in lung cancer.
In conclusion, we identify that SQS might serve not only as a metastasis-associated gene but also as a prognostic marker for lung cancer. Because cholesterol biosynthesis plays an important role in cancer evolution, the targeting of SQS is potentially a specific, effective, and safe strategy for combating clinical malignancies.
The authors thank Miss Tracy Tsai for pathology assistance. They appreciate the kind gift from Dr. Vincenti of Dartmouth College for matrix metallopeptidase 1 prompter. They also thank GRC Affymetrix and Instrument Core for their support in microarray, IVIS spectrum, confocal microscopy, and Aperio digital pathology.
| 1. | Hoffman PC, Mauer AM, Vokes EE. Lung cancer. Lancet 2000;355:479–485. |
| 2. | Spira A, Ettinger DS. Multidisciplinary management of lung cancer. N Engl J Med 2004;350:379–392. |
| 3. | Herbst RS, Heymach JV, Lippman SM. Lung cancer. N Engl J Med 2008;359:1367–1380. |
| 4. | Kroemer G, Pouyssegur J. Tumor cell metabolism: cancer’s Achilles’ heel. Cancer Cell 2008;13:472–482. |
| 5. | Buchakjian MR, Kornbluth S. The engine driving the ship: metabolic steering of cell proliferation and death. Nat Rev Mol Cell Biol 2010;11:715–727. |
| 6. | Gordan JD, Lal P, Dondeti VR, Letrero R, Parekh KN, Oquendo CE, Greenberg RA, Flaherty KT, Rathmell WK, Keith B, et al. HIF-alpha effects on c-Myc distinguish two subtypes of sporadic VHL-deficient clear cell renal carcinoma. Cancer Cell 2008;14:435–446. |
| 7. | Zhang WC, Shyh-Chang N, Yang H, Rai A, Umashankar S, Ma S, Soh BS, Sun LL, Tai BC, Nga ME, et al. Glycine decarboxylase activity drives non-small cell lung cancer tumor-initiating cells and tumorigenesis. Cell 2012;148:259–272. |
| 8. | Nelson ER, Wardell SE, Jasper JS, Park S, Suchindran S, Howe MK, Carver NJ, Pillai RV, Sullivan PM, Sondhi V, et al. 27-Hydroxycholesterol links hypercholesterolemia and breast cancer pathophysiology. Science 2013;342:1094–1098. |
| 9. | Wu Q, Ishikawa T, Sirianni R, Tang H, McDonald JG, Yuhanna IS, Thompson B, Girard L, Mineo C, Brekken RA, et al. 27-Hydroxycholesterol promotes cell-autonomous, ER-positive breast cancer growth. Cell Reports 2013;5:637–645. |
| 10. | Han JY, Lee SH, Yoo NJ, Hyung LS, Moon YJ, Yun T, Kim HT, Lee JS. A randomized phase II study of gefitinib plus simvastatin versus gefitinib alone in previously treated patients with advanced non-small cell lung cancer. Clin Cancer Res 2011;17:1553–1560. |
| 11. | Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related mortality. N Engl J Med 2012;367:1792–1802. |
| 12. | Brown MS, Goldstein JL, Siperstein MD. Regulation of cholesterol synthesis in normal and malignant tissue. Fed Proc 1973;32:2168–2173. |
| 13. | Sato R, Takano T. Regulation of intracellular cholesterol metabolism. Cell Struct Funct 1995;20:421–427. |
| 14. | Zhang Q, Furukawa K, Chen HH, Sakakibara T, Urano T, Furukawa K. Metastatic potential of mouse Lewis lung cancer cells is regulated via ganglioside GM1 by modulating the matrix metalloprotease-9 localization in lipid rafts. J Biol Chem 2006;281:18145–18155. |
| 15. | Silvius JR, del Giudice D, Lafleur M. Cholesterol at different bilayer concentrations can promote or antagonize lateral segregation of phospholipids of differing acyl chain length. Biochemistry 1996;35:15198–15208. |
| 16. | Garman SC, Kinet JP, Jardetzky TS. Crystal structure of the human high-affinity IgE receptor. Cell 1998;95:951–961. |
| 17. | Suzuki KG, Fujiwara TK, Sanematsu F, Iino R, Edidin M, Kusumi A. GPI-anchored receptor clusters transiently recruit Lyn and G alpha for temporary cluster immobilization and Lyn activation: single-molecule tracking study 1. J Cell Biol 2007;177:717–730. |
| 18. | Tran D, Carpentier JL, Sawano F, Gorden P, Orci L. Ligands internalized through coated or noncoated invaginations follow a common intracellular pathway. Proc Natl Acad Sci USA 1987;84:7957–7961. |
| 19. | Sheets ED, Holowka D, Baird B. Critical role for cholesterol in Lyn-mediated tyrosine phosphorylation of FcepsilonRI and their association with detergent-resistant membranes. J Cell Biol 1999;145:877–887. |
| 20. | Zhuang L, Kim J, Adam RM, Solomon KR, Freeman MR. Cholesterol targeting alters lipid raft composition and cell survival in prostate cancer cells and xenografts. J Clin Invest 2005;115:959–968. |
| 21. | Brusselmans K, Timmermans L, Van de Sande T, Van Veldhoven PP, Guan G, Shechter I, Claessens F, Verhoeven G, Swinnen JV. Squalene synthase, a determinant of Raft-associated cholesterol and modulator of cancer cell proliferation. J Biol Chem 2007;282:18777–18785. |
| 22. | Yang YF. Overexpression of cholesterol metabolism regulatory gene farnesyl diphosphate farnesyltransferase is associated with poor prognosis of non-small cell lung cancer patient and promotes metastasis through feedback regulation of SPP1–PI3K-AKT signaling pathway. Presented at the AACR 2012 annual meeting. March 31 to April 4, 2012, Chicago, IL. |
| 23. | Yang YF. SQS promotes metastasis through altered cholesterol contents and recruiting TNFR1 to lipid rafts for NFκB activation in lung cancer. Paper presented at the ASIP 2014 annual meeting at Experimential Biology for Lung Cancer Minisymposium. April 29, 2014, San Diego, CA. |
| 24. | Győrffy B, Surowiak P, Budczies J, Lánczky A. Online survival analysis software to assess the prognostic value of biomarkers using transcriptomic data in non-small-cell lung cancer. PLoS ONE 2013;8:e82241. |
| 25. | Legler DF, Micheau O, Doucey MA, Tschopp J, Bron C. Recruitment of TNF receptor 1 to lipid rafts is essential for TNFalpha-mediated NF-kappaB activation. Immunity 2003;18:655–664. |
| 26. | Bond M, Chase AJ, Baker AH, Newby AC. Inhibition of transcription factor NF-kappaB reduces matrix metalloproteinase-1, -3 and -9 production by vascular smooth muscle cells. Cardiovasc Res 2001;50:556–565. |
| 27. | Heist RS, Marshall AL, Liu G, Zhou W, Su L, Neuberg D, Lynch TJ, Wain J, Christiani DC. Matrix metalloproteinase polymorphisms and survival in stage I non-small cell lung cancer. Clin Cancer Res 2006;12:5448–5453. |
| 28. | Wu YH, Wu TC, Liao JW, Yeh KT, Chen CY, Lee H. p53 dysfunction by xeroderma pigmentosum group C defects enhance lung adenocarcinoma metastasis via increased MMP1 expression. Cancer Res 2010;70:10422–10432. |
| 29. | Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, Jacobsen A, Byrne CJ, Heuer ML, Larsson E, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401–404. |
| 30. | Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, Sun Y, Jacobsen A, Sinha R, Larsson E, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 2013;6:pl1. |
| 31. | Santos CR, Schulze A. Lipid metabolism in cancer. FEBS J 2012;279:2610–2623. |
| 32. | Hirsch HA, Iliopoulos D, Joshi A, Zhang Y, Jaeger SA, Bulyk M, Tsichlis PN, Shirley Liu X, Struhl K. A transcriptional signature and common gene networks link cancer with lipid metabolism and diverse human diseases. Cancer Cell 2010;17:348–361. |
| 33. | Ong ES, Zou L, Li S, Cheah PY, Eu KW, Ong CN. Metabolic profiling in colorectal cancer reveals signature metabolic shifts during tumorigenesis. Mol Cell Proteomics [online ahead of print] 10 Feb 2010; DOI: 10.1074/mcp.M900551-MCP200. |
| 34. | Furuta E, Okuda H, Kobayashi A, Watabe K. Metabolic genes in cancer: their roles in tumor progression and clinical implications. Biochim Biophys Acta 2010;1805:141–152. |
| 35. | Clendening JW, Pandyra A, Boutros PC, El Ghamrasni S, Khosravi F, Trentin GA, Martirosyan A, Hakem A, Hakem R, Jurisica I, et al. Dysregulation of the mevalonate pathway promotes transformation. Proc Natl Acad Sci USA 2010;107:15051–15056. |
| 36. | Chan KK, Oza AM, Siu LL. The statins as anticancer agents. Clin Cancer Res 2003;9:10–19. |
| 37. | Poynter JN, Gruber SB, Higgins PD, Almog R, Bonner JD, Rennert HS, Low M, Greenson JK, Rennert G. Statins and the risk of colorectal cancer. N Engl J Med 2005;352:2184–2192. |
| 38. | De Pinieux G, Chariot P, Ammi-Saïd M, Louarn F, Lejonc JL, Astier A, Jacotot B, Gherardi R. Lipid-lowering drugs and mitochondrial function: effects of HMG-CoA reductase inhibitors on serum ubiquinone and blood lactate/pyruvate ratio. Br J Clin Pharmacol 1996;42:333–337. |
| 39. | Wong WW, Dimitroulakos J, Minden MD, Penn LZ. HMG-CoA reductase inhibitors and the malignant cell: the statin family of drugs as triggers of tumor-specific apoptosis. Leukemia 2002;16:508–519. |
| 40. | Thompson PD, Clarkson P, Karas RH. Statin-associated myopathy. JAMA 2003;289:1681–1690. |
| 41. | Hu J, La Vecchia C, de Groh M, Negri E, Morrison H, Mery L; Canadian Cancer Registries Epidemiology Research Group. Dietary cholesterol intake and cancer. Ann Oncol 2012;23:491–500. |
| 42. | Kok DE, van Roermund JG, Aben KK, den Heijer M, Swinkels DW, Kampman E, Kiemeney LA. Blood lipid levels and prostate cancer risk; a cohort study. Prostate Cancer Prostatic Dis 2011;14:340–345. |
| 43. | Ushio-Fukai M, Hilenski L, Santanam N, Becker PL, Ma Y, Griendling KK, Alexander RW. Cholesterol depletion inhibits epidermal growth factor receptor transactivation by angiotensin II in vascular smooth muscle cells: role of cholesterol-rich microdomains and focal adhesions in angiotensin II signaling. J Biol Chem 2001;276:48269–48275. |
| 44. | Zhuang L, Lin J, Lu ML, Solomon KR, Freeman MR. Cholesterol-rich lipid rafts mediate akt-regulated survival in prostate cancer cells. Cancer Res 2002;62:2227–2231. |
| 45. | Murata M, Peränen J, Schreiner R, Wieland F, Kurzchalia TV, Simons K. VIP21/caveolin is a cholesterol-binding protein. Proc Natl Acad Sci USA 1995;92:10339–10343. |
| 46. | Jeon JH, Kim SK, Kim HJ, Chang J, Ahn CM, Chang YS. Lipid raft modulation inhibits NSCLC cell migration through delocalization of the focal adhesion complex. Lung Cancer 2010;69:165–171. |
| 47. | Murai T, Maruyama Y, Mio K, Nishiyama H, Suga M, Sato C. Low cholesterol triggers membrane microdomain-dependent CD44 shedding and suppresses tumor cell migration. J Biol Chem 2011;286:1999–2007. |
| 48. | Pao W, Girard N. New driver mutations in non-small-cell lung cancer. Lancet Oncol 2011;12:175–180. |
| 49. | Gaglio D, Metallo CM, Gameiro PA, Hiller K, Danna LS, Balestrieri C, Alberghina L, Stephanopoulos G, Chiaradonna F. Oncogenic K-Ras decouples glucose and glutamine metabolism to support cancer cell growth. Mol Syst Biol 2011;7:523. |
| 50. | Ying H, Kimmelman AC, Lyssiotis CA, Hua S, Chu GC, Fletcher-Sananikone E, Locasale JW, Son J, Zhang H, Coloff JL, et al. Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism. Cell 2012;149:656–670. |
*These authors contributed equally to this work.
Supported by grants from the National Science Council (NSC-102-2320-B-001-027-MY3) and Academia Sinica.
Author Contributions: Designed and wrote the manuscript, Y.-F.Y., Y.-H.J., and M.H. Provided materials, C.-J.Y. and M.-S.H. Performed experiments, Y.-F.Y., Y.-H.J., H.-Y.T., Y.-C.C., J.C., and T.-C.L. Interpreted data, Y.-F.L., Y.-P.L., C.-Y.S., J.L., C.-N.S., J.-Y.S., and P.-J.L.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201404-0714OC on August 23, 2014
Author disclosures are available with the text of this article at www.atsjournals.org.
