Abstract

OBJECTIVE

Toll-like receptor 4 (Tlr-4) mediates many biological effects of lipopolysaccharide (LPS), which has antitumoral effects on glioblastoma both in vivo and in vitro. However, the precise role of Tlr-4 in these antitumoral effects remains unknown.

METHODS

The role of Tlr-4 in the antitumoral effect of LPS on glioblastomas was assessed in wild-type BALB/c mice and in Tlr-4 knockout (KO) BALB/c mice. Mice were implanted with DBT glioblastoma cells intracranially or subcutaneously, were treated with intratumoral LPS, and were assessed by histopathological examination for degrees of tumor progression and inflammation. Flow cytometry and Western blotting with antibodies to the Tlr-4 receptor and flow cytometry to the related CD14 moiety were performed to quantitate the expression levels of these two receptors by glioblastoma cells.

RESULTS

For subcutaneous tumors, LPS caused near complete tumor elimination in wild-type mice, but only a 50% reduction in Tlr-4 KO mice. For mice implanted with intracranial glioblastomas, LPS increased survival times modestly in wild-type mice, but showed no benefit in the Tlr-4 KO mice. There were no histological differences among wild-type and Tlr-4 KO mice, except for tumor size. In both models, an early neutrophilic and later macrophage-rich inflammatory infiltrate were seen after LPS administration. Quantitative flow cytometry and Western blotting showed no Tlr-4 receptor or CD14 expression in murine and human glioblastoma cells in vitro, and Western blotting suggested that Tlr-4 effects are mediated by nontumoral elements such as microglia and inflammatory cells.

CONCLUSION

LPS-induced antitumoral effects on glioblastoma multiforme are mediated, in part, by the Tlr-4 receptor. Further understanding of this process may lead to novel treatment strategies for this uniformly fatal disease.

Glioblastoma multiforme (GBM) remains among the most deadly of all cancers. Mean patient survival, even with aggressive treatments, including surgery, radiation therapy, and chemotherapy, is typically 12 months or less (15,17). Previous in vivo investigations have demonstrated powerful antitumoral effects of lipopolysaccharide (LPS) on GBM, mediated in part by the tumor-bearing host's immune system (6,38). However, the mechanisms of this antitumoral response have not been fully elucidated. Since its first descriptions in vertebrates in the late 1990s (20,29,33), there has been an explosion of research showing the role of the toll-like receptors (TLRs), the associated signal transduction pathways, and the downstream nuclear factor κ-B transcription factors in various immune-mediated processes (1,7,12,20,21,23,37). The TLRs are a class of evolutionarily conserved transmembrane receptors belonging to the interleukin-1 receptor/TLRsuperfamily that has been described recently (1,4,7,12,39). The toll-like nomenclature relates to the homology of these molecules to the toll protein that mediates various physiological and developmental processes in the fruit fly, Drosophila melanogaster (7,37,39). Tlr-4, in conjunction with the CD14 molecule, has high specificity for LPS binding and initiates a cascade of downstream events mediated by nuclear factor κ-B, which, in turn, mediates a host of transcriptional functions (1,7,13,24,29,37,39). Limited studies have demonstrated a role for Tlr-4 in anticancer immunity in head and neck cancers. OK-432 is a streptococcal agent that has been used for immunotherapy of head and neck cancer, among other malignancies, and OK-PSA is a lipoteichoic acid-related molecule isolated from OK-432. In vivo studies in Tlr-4–deficient mice have shown that the antitumoral effects of OK-432 and OK-PSA are mediated in part by Tlr-4 (25,26). The role of the TLRs, particularly Tlr-4, in the antitumoral effects of LPS on GBM has not been evaluated previously. We present studies showing that Tlr-4 plays an important in vivo role in the antitumoral effects of LPS on GBM in mouse models. We originally hypothesized that LPS may be acting directly on implanted glioblastoma cells through Tlr-4 or CD14 receptor pathways, or both. However, the quantitative flow cytometry data of human and murine glioblastoma cells in vitro and Western blot analysis of murine glioblastoma cells in vivo with and without LPS treatments showed no evidence of human or mouse Tlr-4 expressions, and flow cytometry data confirmed no CD14 expression, suggesting our original hypothesis to be incorrect. Therefore, other cell types in the tumor-bearing host must bear Tlr-4, which would account for the diminished antitumoral response to LPS in Tlr-4 knockout (KO) mice.

MATERIALS AND METHODS

Maintenance of DBT Glioblastoma Cells in Culture

The delayed brain tumor (DBT) cell line was established from a tumor induced in an adult mouse by intracerebral injection of the Rous sarcoma virus (16). Tumors generated after intracranial or subcutaneous implantation of the DBT cells are analogous to human glioblastomas in terms of their aggressive growth pattern, histopathological characteristics, and immunoreactivity for glial fibrillary acidic protein (6). DBT glioma cells (previously donated to our laboratory by Dr. Michael M.C. Lai, Department of Molecular Microbiology and Immunology, University of Southern California) were used for subcutaneous and intracranial implantations in Tlr-4 KO mice (The Jackson Laboratory, Bar Harbor, ME) and in appropriately matched BALB/c wild-type mice (The Jackson Laboratory). As previously described (6,35,38), DBT cells were grown in Dulbecco's modified Eagle's medium (DMEM; Gibco BRL Life Technologies, Rockville, MD) with 15% fetal bovine serum (FBS; Gemini Bioproducts, Calabasas, CA), 0.2 mM glutamine (Gibco BRL Life Technologies), 50 𝛍g/ml neomycin (Sigma-Aldrich Corp., St. Louis, MO), 100 𝛍g/ml penicillin, and 100 𝛍g/ml streptomycin and plated in culture flasks for incubation in a 5% CO2 humidified atmosphere at 377C. On reaching confluency, cell passage was accomplished by removing feeding medium, washing attached cells with calcium and magnesium-free Hunk's balanced salt solution, and then detaching the cells from the flask with exposure to 0.25% trypsin-ethylenediamine tetra-acetic acid (EDTA; Gibco BRL Life Technologies). Trypsin-EDTA was neutralized with feeding medium containing FBS, and the suspension was centrifuged at 1200 rpm for 10 minutes. Supernatant was removed, the pellet resuspended in feeding medium, and the cells replated at 1 × 105 cells per flask.

Maintenance of the J774a.1 and RAW 264.7 Cell Line

The J774A.1 cell line (American Type Culture Collection [ATCC], Manassas, VA) is a murine macrophage cell line derived from a female BALB/c mouse and is active in antibody-dependent phagocytosis (31). The RAW 264.7 cell line (ATCC) is a murine macrophage cell line established from a tumor induced by the Abelson murine leukemia virus in a male BALB/c mouse (32). For our quantitative flow cytometry studies of Tlr-4 and CD14, both cell lines served as positive controls; the J774A.1 line was also used as a positive control for the Tlr-4 Western blotting. Both cell lines were grown in high-glucose DMEM with 10% FBS, 0.2 mM glutamine, 50 𝛍g/ml neomycin, 100 𝛍g/ml penicillin, and 100 𝛍g/ml streptomycin and plated in 100 mm tissue culture dishes for incubation in a 5% CO2 humidified atmosphere at 377C. On reaching confluency, passage of the cells was accomplished by removing old feeding media, washing the cells with calcium and magnesium-free Hunk's balanced salt solution, adding fresh feeding media, then scraping the cells off the dish using a sterile cell scraper. The dislodged cells then were dispensed into new tissue culture dishes at a subcultivation ratio of 1:3 to 1:6. Media was either replaced or added two to three times weekly until cells reached confluency.

Maintenance of the U-87 MG Cell Line

The U-87 MG cell line (ATCC) is a human glioblastoma cell line derived from a 44-year-old Caucasian woman. This is a hypodiploid human cell line with the modal chromosome 44 occurring in 48% of cells. This is one of a number of cell lines derived from malignant gliomas by Ponten and Macintyre (30) from 1966 through 1969. The original cultures were established as explants on grid-supported lens paper or gelatin foam with Eagle's minimum essential medium and 10% bovine calf serum as the culture fluid. Trypsinization of the outgrowth of cells attached to the vessel floor with subsequent transfer to standard vessels in growth medium permitted cell line development. A culture at passage 108 was deposited at ATCC by Ponten in 1973. For our quantitative flow cytometry studies of Tlr-4 and CD14, U-87 MG cells were grown in Eagle's minimal essential medium with Earle's balanced salt solution and 2 mmol/L l-glutamine that is modified to contain 1.0 mmol/L sodium pyruvate and 0.1 mmol/L nonessential amino acids (Mediatech, Inc., Herndon, VA). In addition, 10% FBS, 50 𝛍g/ml neomycin, 100 𝛍g/ml penicillin, and 100 𝛍g/ml streptomycin were added to the medium. Cells were plated in culture flasks for incubation in a 5% CO2 humidified atmosphere at 377C. On reaching confluency, passage of the cells was accomplished by removing feeding medium, washing attached cells with calcium and magnesium-free Hunk's balanced salt solution, and then detaching the cells from the flask with exposure to 0.25% trypsin-EDTA. Trypsin-EDTA was neutralized with feeding medium containing FBS, and the suspension was centrifuged at 1200 rpm for 10 minutes. The supernatant was removed and the pellet resuspended in feeding medium at a subcultivation ratio of 1:2 to 1:5.

Preparation and Care of Animals

Mice used in this study were cared for according to the guidelines of the American Association for Accreditation of Laboratory Animal Care as approved by the Division of Comparative Medicine at Washington University. For animal procedures, including implantation of DBT cells and injection of LPS, general anesthesia was induced using intraperitoneal administration of a rodent cocktail (87 mg/Kg ketamine and 13 mg/Kg xylazine).

Subcutaneous Implantation of DBT Cells in Tlr-4 KO and Wild-Type BALB/C Mice

General anesthesia was induced for female Tlr-4 KO BALB/c mice (The Jackson Laboratory) or corresponding wild-type female BALB/c mice as described previously. The flank region of each mouse was prepared with 95% ethanol, and 2 million DBT mouse glioblastoma cells in 200 ml serum-free DMEM were injected subcutaneously. On Day 21, mice were sacrificed by intraperitoneal administrations of 0.01 ml Euthasol (Delmarva Laboratory Inc., Des Moines, IA). Tumors were harvested and mean tumor mass was compared between treatment and control groups of mice implanted with subcutaneous DBT tumors.

Intratumoral LPS Treatments for Mice Implanted Subcutaneously with DBT Glioblastomas

Mice implanted with subcutaneous DBT glioblastomas were treated with intratumoral LPS (Sigma; 400 𝛍g Escherichia coli serotype O55:B5) on Days 7 and 14. Control Tlr-4 KO and control wild-type mice did not receive LPS, but rather were injected intratumorally with phosphate-buffered saline (PBS). Tumors were harvested on Day 21 and mean tumor masses in grams were compared using unpaired t tests between LPS-treated mice and controls for both Tlr-4 KO mice and BALB/c wild-type mice.

Intracranial Implantation of DBT Cells in Tlr-4 KO and Wild-Type BALB/c Mice

Female Tlr-4 KO mice and corresponding wild-type female BALB/c mice were anesthetized as described previously. The heads were prepped with 95% ethanol, a 1-cm midline longitudinal incision was made, and the mouse was fixed in a stereotactic frame (Stoelting Co., Wood Dale, IL). A right paramedian craniotomy was made (2 mm lateral and 2 mm posterior to bregma) using a dental drill with a 2-mm bit (Foredom Electronic Co., Bethel, CT). Five hundred thousand DBT cells suspended in 9 𝛍l DMEM were injected over 30 seconds to a depth of 4 mm into the brain using a Hamilton syringe (Hamilton Company, Reno, NV). The animal was removed from the stereotactic frame and the scalp was closed with Autoclips (Becton Dickinson & Company, Sparks, MD). After recovery from anesthesia, mice were returned to cages to resume normal activity and diet ad libitum. Mice were observed daily to the end point of death or a moribund state, at which time they were sacrificed using Euthasol. Survival in days from the time of tumor cell implantation was recorded for each animal implanted with an intracranial DBT tumor. Differences in survival were compared between controls and the various LPS treatment groups using the log-rank 2 test.

Intratumoral LPS Treatments for Mice Implanted Intracranially with DBT Glioblastomas

On Day 0, while under anesthesia and in a stereotactic frame as mentioned previously, mice were injected intracranially with 100 𝛍g LPS via the same craniotomy site in the skull. The scalp was closed and the mice recovered as described previously.

Preparation of Tumor Specimens with and without LPS Treatment for Histopathological Evaluation

At predetermined intervals after subcutaneous implantation of DBT cells, subsets of LPS-treated and control mice were sacrificed using intraperitoneal Euthasol as mentioned previously. Subcutaneous tumors were collected and weighed. Each tumor then was postfixed with 4% paraformaldehyde in PBS at 47C for at least 24 hours, followed by placement in 30% sucrose in PBS at 47C for at least 24 hours. Fixed specimens were embedded in Tissue Tek OCT compound (Miles, Elkhart, IN) and frozen on dry ice before being cut into 10- to 20-𝛍m thick sections with a cryostat. Sections were fixed for 24 hours on frosted microscope slides. Specimens were analyzed for general histopathological features, with particular emphasis on inflammatory infiltrates.

Hematoxylin and Eosin Staining

Sections of specimens on frosted microscope slides were fixed serially in Pro-Par Clearant (Anatech, Ltd., Battle Creek, MI) for 9 minutes, absolute alcohol for 2 minutes, 95% ethanol for 1 minute, and 70% ethanol for 1 minute, followed by a 1-minute rinse in deionized water. Specimens were then placed in Harris' acidified hematoxylin (Anatech, Ltd.) for 2 minutes, followed by a brief tap water rinse, and were placed in acidic alcohol for 1 minute, followed by another brief rinse with tap water. The slides then were placed in 70% ethanol for 1 minute and in eosin (Anatech, Ltd.) for 1.5 minutes, after which serial rinses with 95% ethanol (1 min), 100% ethanol (3 min), and Pro-Par Clearant (3 min) were completed. The slides were mounted with Permount (Fisher, St. Louis, MO), and coverslips were applied.

Statistical Analyses

Statistical analyses were conducted using SAS software (SAS Institute, Cary, NC). Data derived from mice with subcutaneous DBT tumors treated with LPS were assessed using a one-way analysis of variance and post hoc tests using Tukey's studentized range test. Survival in days from the time of intracranial tumor implantation was analyzed using Kaplan-Meier product-limit survival estimates. Differences in survival were compared between controls and the various LPS treatment groups using the log-rank 2 test.

In Vitro Glioblastoma Cells Labeled with Antibodies to Human Tlr-4, Mouse Tlr-4, and CD14

Flow cytometry of glioblastoma cells incubated with phycoerythrin (PE)-labeled antibody to Tlr-4 or fluorescein isothiocyanate (FITC)-labeled antibody to CD14 was used to quantitatively assess the presence of human Tlr-4, mouse Tlr-4, and CD14 on glioblastoma cells and macrophages before and after LPS treatments. Assays for the mouse cell lines, DBT, RAW, and J774, were performed with rat antimouse antibodies to Tlr-4 and CD14, respectively, and assays for the human glioblastomas U87 were performed with mouse antihuman antibodies to Tlr-4 and CD14, respectively. One million cells of each cell line tested were plated on 100-mm tissue culture dishes 1 day before flow evaluation. After 24 hours, each cell line was exposed to either 100 or 200 𝛍g/ml LPS in serum-free media for approximately 30 minutes. Controls were exposed to an equivalent amount of sterile PBS in serum-free media for the same amount of time. After 30 minutes, the cells were washed, trypsinized, and spun down for 5 minutes at 1200 rpm. Then, each cell pellet for each cell line tested was resuspended in cold PBS, incubated with PE-labeled monoclonal antibody to human or mouse Tlr-4 (eBioscience, San Diego, CA) or FITC-labeled monoclonal antibody to CD14 (Pharmingen, San Diego, CA) for 1 hour at 47C, washed once in PBS, spun down for 5 minutes at 1200 rpm, and finally resuspended in 1 ml cold PBS.

Flow Cytometry Technique

Flow cytometric analyses of FITC-CD14–labeled and PE-human Tlr-4–labeled cells (or PE-mouse Tlr-4 labeled) were conducted using a Becton Dickinson FACS 440 flow cytometer (BD Biosciences Immunocytometry Systems, San Jose, CA). The system was interfaced to a CICERO/CYCLOPS data acquisition system (Cytomation, Inc., Ft. Collins, CO) for data storage and analysis. Both fluorochromes were excited with 300 mW at 488 nm of an argon ion laser. FITC fluorescence was detected through a 530-nm band-pass filter, whereas PE fluorescence was detected through a 580-nm band-pass filter. A minimum of 20,000 events were acquired in the list mode for each sample, and fluorescence distributions of live cells were obtained by gating.

Western Blot Technique

Western blot analysis of disaggregated specimens from implanted tumors in Tlr-4 KO BALB/c mice and wild-type BALB/c mice with or without LPS treatments were conducted using a standard sodium dodecyl sulfate-polyacrylamide gel electrophoresis technique. The blot was incubated with an anti-Tlr-4 antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA). A horseradish peroxidase conjugated antigoat polyclonal (Santa Cruz Biotechnology Inc.) was used as the secondary antibody. The prepared blot was visualized with the SuperSignal West Femto Maximum Sensitivity Substrate (Pierce Biotechnology, Inc., Rockford, IL).

RESULTS

Subcutaneous DBT Glioblastomas are More Resistant to the Antitumoral Effects of LPS When Implanted in Tlr-4 KO Mice than When Implanted in Wild-type BALB/c Mice

In the subcutaneous model, LPS caused near complete tumor elimination in wild-type BALB/c mice; however, in Tlr-4 KO BALB/c mice, only a 50% tumor reduction was achieved (Figs. 1 and 2, Table 1). These results indicate that the Tlr-4 KO mice bearing subcutaneously implanted DBT cells are more resistant to the antitumoral effects of LPS than wild-type BALB/c mice bearing subcutaneous implanted DBT cells.

FIGURE 1.

Diagram summarizing the methodology and results for subcutaneous DBT glioblastoma implantations in Tlr-4 KO mice and wild-type BALB/c mice and LPS treatments.

FIGURE 1.

Diagram summarizing the methodology and results for subcutaneous DBT glioblastoma implantations in Tlr-4 KO mice and wild-type BALB/c mice and LPS treatments.

FIGURE 2.

Bar graph showing mean tumor masses for mice subcutaneously implanted with DBT glioblastomas, including wild-type BALB/c mice with or without LPS treatments and Tlr-4 KO mice with or without LPS treatments (see text and Table 1for details). The P value for BALB/c control mice versus BALB/c + LPS mice was 0.0135 and for Tlr-4 KO control mice versus Tlr-4 KO + LPS mice was 0.1174. For BALB/c control mice versus Tlr-4 KO control mice, the P value was 0.9130.

FIGURE 2.

Bar graph showing mean tumor masses for mice subcutaneously implanted with DBT glioblastomas, including wild-type BALB/c mice with or without LPS treatments and Tlr-4 KO mice with or without LPS treatments (see text and Table 1for details). The P value for BALB/c control mice versus BALB/c + LPS mice was 0.0135 and for Tlr-4 KO control mice versus Tlr-4 KO + LPS mice was 0.1174. For BALB/c control mice versus Tlr-4 KO control mice, the P value was 0.9130.

TABLE 1.

Summary of results for subcutaneous implantationa

Intracranial DBT Glioblastomas are More Resistant to the Antitumoral Effects of LPS When Implanted in Tlr-4 KO Mice than When Implanted in Wild-type BALB/c Mice

Intracranial LPS caused a modest increase in survival for mice implanted with intracranial DBT cells in wild-type BALB/c mice (Fig. 3A), but did not increase the survival of Tlr-4 KO BALB/c mice bearing intracranial DBT cells (Fig. 3B). These results indicate that the Tlr-4 KO BALB/c mice bearing intracranially implanted DBT cells are more resistant to the antitumoral effects of LPS than wild-type BALB/c mice bearing intracranially implanted DBT cells.

FIGURE 3.

A, Kaplan-Meier survival plot for female BALB/c mice intracranially implanted with 5 × 105 DBT glioblastoma cells mixed with 100 μg LPS or PBS on Day 0. Median survival was 19 days (mean, 20.8 d; range, 12–34 d) for the five mice treated with 100 μg LPS versus 10 days (mean, 11.2 d; range, 9–23 d) for the 13 control mice. Log-rank analysis of Kaplan-Meier plots showed statistically significant prolonged survival for the LPS-treated mice compared with control mice (P = 0.0102). B, Kaplan-Meier survival plot for female Tlr-4 KO BALB/c mice intracranially implanted with 5 × 105 DBT glioblastoma cells mixed with 100 μg LPS or PBS on Day 0. Median survival was 14 days (mean, 16.4 d; range, 11–28 d) for the 10 Tlr-4 KO mice treated with 100 μg LPS versus 12 days (mean, 13.8 d; range, 10–21 d) for the nine control Tlr-4 KO mice treated with PBS. Log-rank analysis of Kaplan-Meier plots showed no increase in survival for the LPS-treated Tlr-4 KO mice compared with Tlr-4 KO control mice (P = 0.3277).

FIGURE 3.

A, Kaplan-Meier survival plot for female BALB/c mice intracranially implanted with 5 × 105 DBT glioblastoma cells mixed with 100 μg LPS or PBS on Day 0. Median survival was 19 days (mean, 20.8 d; range, 12–34 d) for the five mice treated with 100 μg LPS versus 10 days (mean, 11.2 d; range, 9–23 d) for the 13 control mice. Log-rank analysis of Kaplan-Meier plots showed statistically significant prolonged survival for the LPS-treated mice compared with control mice (P = 0.0102). B, Kaplan-Meier survival plot for female Tlr-4 KO BALB/c mice intracranially implanted with 5 × 105 DBT glioblastoma cells mixed with 100 μg LPS or PBS on Day 0. Median survival was 14 days (mean, 16.4 d; range, 11–28 d) for the 10 Tlr-4 KO mice treated with 100 μg LPS versus 12 days (mean, 13.8 d; range, 10–21 d) for the nine control Tlr-4 KO mice treated with PBS. Log-rank analysis of Kaplan-Meier plots showed no increase in survival for the LPS-treated Tlr-4 KO mice compared with Tlr-4 KO control mice (P = 0.3277).

Histopathological Examination

Microscopic examination of subcutaneously implanted DBT glioblastomas with or without LPS treatments showed no clear differences between wild-type BALB/c and Tlr-4 KO BALB/c mice, except that the tumors tended to be larger in the Tlr-4 KO mice (Fig. 4, A–D). Both subcutaneously and intracranially implanted DBT tumors showed histopathological features of glioblastoma, including hypercellularity and pseudopalisading necrosis. There was a peritumoral neutrophilic infiltrate within a few days after LPS treatment, followed by a macrophage-rich infiltrate 1 week after LPS administration (Fig. 4, E–H).

FIGURE 4.

Photomicrographs of subcutaneous DBT tumors stained with hematoxylin and eosin at 3200 magnification. A–C, photomicrograph of subcutaneous tumor stained with hematoxylin and eosin 7 days after implantation of 2 × 106 DBT cells in BALB/c mice (A and B) and Tlr-4 KO BALB/c mice (C) without treatment (A) or after treatment (B and C) with PBS. The hypercellular histological results of these control tumors are identical when comparing specimens from the wild-type BALB/c mice with specimens from the Tlr-4 KO BALB/c mice. C contains several entrapped skeletal muscle fibers from the abdominal wall at the site of tumor growth. D, photomicrograph of similar subcutaneous tumor in Tlr-4 KO mouse 14 days after implantation of DBT cells. The tumor demonstrated pseudopalisading necrosis characteristic of glioblastoma multiforme. E and F, photomicrographs demonstrating the inflammatory response 2 days after intratumoral injection of 400 μg LPS in wild-type BALB/c mouse (E) and Tlr-4 KO mouse (F) implanted with subcutaneous tumors. In both the wild-type and Tlr-4 KO mice, most of the specimen consisted of inflamed granulation tissue rich in neutrophils. Small nests of neoplastic cells comprise a minority of the lesion (F, lower left). Histologically, the two tumors were similar, although the wild-type mice had much smaller tumors. G and H, photomicrographs of histologically similar subcutaneous tumors encountered in wild-type BALB/c (G) and Tlr-4 KO (H) mice 7 days after intratumoral injection of 400 μg LPS, although the former typically were smaller. The inflammatory component consisted predominantly of macrophages and lymphocytes (G), and the tumor had characteristic foci of pseudopalisading necrosis (H).

FIGURE 4.

Photomicrographs of subcutaneous DBT tumors stained with hematoxylin and eosin at 3200 magnification. A–C, photomicrograph of subcutaneous tumor stained with hematoxylin and eosin 7 days after implantation of 2 × 106 DBT cells in BALB/c mice (A and B) and Tlr-4 KO BALB/c mice (C) without treatment (A) or after treatment (B and C) with PBS. The hypercellular histological results of these control tumors are identical when comparing specimens from the wild-type BALB/c mice with specimens from the Tlr-4 KO BALB/c mice. C contains several entrapped skeletal muscle fibers from the abdominal wall at the site of tumor growth. D, photomicrograph of similar subcutaneous tumor in Tlr-4 KO mouse 14 days after implantation of DBT cells. The tumor demonstrated pseudopalisading necrosis characteristic of glioblastoma multiforme. E and F, photomicrographs demonstrating the inflammatory response 2 days after intratumoral injection of 400 μg LPS in wild-type BALB/c mouse (E) and Tlr-4 KO mouse (F) implanted with subcutaneous tumors. In both the wild-type and Tlr-4 KO mice, most of the specimen consisted of inflamed granulation tissue rich in neutrophils. Small nests of neoplastic cells comprise a minority of the lesion (F, lower left). Histologically, the two tumors were similar, although the wild-type mice had much smaller tumors. G and H, photomicrographs of histologically similar subcutaneous tumors encountered in wild-type BALB/c (G) and Tlr-4 KO (H) mice 7 days after intratumoral injection of 400 μg LPS, although the former typically were smaller. The inflammatory component consisted predominantly of macrophages and lymphocytes (G), and the tumor had characteristic foci of pseudopalisading necrosis (H).

Flow Cytometry Analysis Shows No Expression of Human Tlr-4, Mouse Tlr-4, or CD14 in Murine and Human Glioblastoma Cells

Quantitative flow cytometry did not show detectable Tlr-4 receptor or the CD14 moiety for either murine (DBT) or human (U87) glioblastoma cells in vitro (Fig. 5). The murine macrophage cell lines J774A.1 and RAW 264.7 served as Tlr-4– and CD14-positive controls.

FIGURE 5.

A–D, graphs of flow cytometry data from four cell lines labeled with FITC-tagged CD14 antibodies. RAW macrophages (A), DBT mouse glioblastoma cells (B), U87 human glioblastoma cells (C), and J774 macrophages (D) were incubated with the FITC-labeled CD14 antibody. After labeling in vitro, flow cytometry was performed as described in the Materials and Methods section. For each graph, the y axis represents the cell counts and the x axis represents the labeling intensity. The red peaks (vertical lines) represent the untreated controls (no LPS), the green peaks (angled lines from the lower left to the upper right) represent the 100 μg/ml LPS-treated cells, and the brown peaks (angled lines from the upper left to the lower right) represent the 200 μg/ml LPS-treated cells. The latter dose, 200 mg/ml, was tested only in the DBT and U87 cells. The rightward shift of the peaks for the two macrophage cell types indicates the presence of the CD14. The leftward position of the peaks for the two glioblastoma cell lines indicates the absence of expression of CD14. E–G, graphs of flow cytometry data from similar experiments performed using a PE-labeled Tlr-4 antibody using the J774 macrophage cell line (E), a human glioblastoma cell line (F), and one murine glioblastoma cell line (G). The red peaks (vertical lines) represent the untreated controls (no LPS and no Tlr-4 antibody), the green peaks (angled lines from the lower left to the upper right) represent untreated cells (no LPS) labeled with Tlr-4 antibody, and the brown peaks (angled lines from the upper left to the lower right) represent the 100 μg/ml LPS-treated cells. The rightward shift with the macrophage cell line indicates labeling of the Tlr-4 in these cell lines and the absence of a rightward shift of the peaks. In the two panels of glioblastoma cell cultures, the lack of a shift indicates no expression in these cell types. Note that neither the Tlr-4 nor the CD14 expression was affected by exposure in vitro to LPS, except for the CD14 expression in the J774 macrophage cells, which was increased by LPS exposure.

FIGURE 5.

A–D, graphs of flow cytometry data from four cell lines labeled with FITC-tagged CD14 antibodies. RAW macrophages (A), DBT mouse glioblastoma cells (B), U87 human glioblastoma cells (C), and J774 macrophages (D) were incubated with the FITC-labeled CD14 antibody. After labeling in vitro, flow cytometry was performed as described in the Materials and Methods section. For each graph, the y axis represents the cell counts and the x axis represents the labeling intensity. The red peaks (vertical lines) represent the untreated controls (no LPS), the green peaks (angled lines from the lower left to the upper right) represent the 100 μg/ml LPS-treated cells, and the brown peaks (angled lines from the upper left to the lower right) represent the 200 μg/ml LPS-treated cells. The latter dose, 200 mg/ml, was tested only in the DBT and U87 cells. The rightward shift of the peaks for the two macrophage cell types indicates the presence of the CD14. The leftward position of the peaks for the two glioblastoma cell lines indicates the absence of expression of CD14. E–G, graphs of flow cytometry data from similar experiments performed using a PE-labeled Tlr-4 antibody using the J774 macrophage cell line (E), a human glioblastoma cell line (F), and one murine glioblastoma cell line (G). The red peaks (vertical lines) represent the untreated controls (no LPS and no Tlr-4 antibody), the green peaks (angled lines from the lower left to the upper right) represent untreated cells (no LPS) labeled with Tlr-4 antibody, and the brown peaks (angled lines from the upper left to the lower right) represent the 100 μg/ml LPS-treated cells. The rightward shift with the macrophage cell line indicates labeling of the Tlr-4 in these cell lines and the absence of a rightward shift of the peaks. In the two panels of glioblastoma cell cultures, the lack of a shift indicates no expression in these cell types. Note that neither the Tlr-4 nor the CD14 expression was affected by exposure in vitro to LPS, except for the CD14 expression in the J774 macrophage cells, which was increased by LPS exposure.

Western Blot Analysis Shows No Expression of Tlr-4 in Murine DBT Glioblastoma Cells Implanted Subcutaneously

Western blotting did not show detectable Tlr-4 receptor in the disaggregated tumors harvested from Tlr-4 KO and wild-type mice previously implanted with DBT glioblastoma cells (Fig. 6). The murine macrophage cell line J774A.1 served as the Tlr-4–positive control.

FIGURE 6.

Western blots of Tlr-4 receptor and α-actin expression in disaggregated specimens harvested from Tlr-4 KO and wild-type BALB/c mice after implantation of DBT glioblastoma cells. Lane 1 represents tumor specimen from Tlr-4 KO mice implanted with DBT glioblastoma cells without LPS treatment. Lane 2 represents tumor specimen from Tlr-4 KO mice implanted with DBT glioblastoma cells and treated with LPS. Lane 3 represents tumor specimen from wild-type BALB/c mice implanted with DBT glioblastoma cells without LPS treatment. Lane 4 represents disaggregated spleen specimen from wild-type BALB/c mice without LPS. Lanes 1, 2, and × have no expression for Tlr-4 receptor. Lane 4 shows expression of the Tlr-4 receptor in the spleen. Lane 5 represents the J774A.1 murine macrophage line, which shows a strong Tlr-4 expression. Lane 6 represents disaggregated tumor specimen from wild-type BALB/c mice implanted with DBT glioblastoma cells and treated with LPS. Lane 7 represents a disaggregated spleen specimen from wild-type BALB/c mice treated with LPS. Lane 8 represents disaggregated spleen specimen from Tlr-4 KO mice without LPS. Lanes 6 and 8 show no expression for the Tlr-4 receptors, whereas the wild-type spleen extract in Lane 7 shows expression of the Tlr-4 receptor. The Western blot shows how in vivo DBT glioblastoma cells with and without LPS treatment in both Tlr-4 KO and wild-type mice did not express Tlr-4 receptors; however, the macrophage cell line as well as the wild-type BALB/c spleen extracts showed expression of Tlr-4 with and without LPS treatment. Each lane additionally was probed for α-actin expression as a control. Lanes 1 through 8 showed expression of α-actin.

FIGURE 6.

Western blots of Tlr-4 receptor and α-actin expression in disaggregated specimens harvested from Tlr-4 KO and wild-type BALB/c mice after implantation of DBT glioblastoma cells. Lane 1 represents tumor specimen from Tlr-4 KO mice implanted with DBT glioblastoma cells without LPS treatment. Lane 2 represents tumor specimen from Tlr-4 KO mice implanted with DBT glioblastoma cells and treated with LPS. Lane 3 represents tumor specimen from wild-type BALB/c mice implanted with DBT glioblastoma cells without LPS treatment. Lane 4 represents disaggregated spleen specimen from wild-type BALB/c mice without LPS. Lanes 1, 2, and × have no expression for Tlr-4 receptor. Lane 4 shows expression of the Tlr-4 receptor in the spleen. Lane 5 represents the J774A.1 murine macrophage line, which shows a strong Tlr-4 expression. Lane 6 represents disaggregated tumor specimen from wild-type BALB/c mice implanted with DBT glioblastoma cells and treated with LPS. Lane 7 represents a disaggregated spleen specimen from wild-type BALB/c mice treated with LPS. Lane 8 represents disaggregated spleen specimen from Tlr-4 KO mice without LPS. Lanes 6 and 8 show no expression for the Tlr-4 receptors, whereas the wild-type spleen extract in Lane 7 shows expression of the Tlr-4 receptor. The Western blot shows how in vivo DBT glioblastoma cells with and without LPS treatment in both Tlr-4 KO and wild-type mice did not express Tlr-4 receptors; however, the macrophage cell line as well as the wild-type BALB/c spleen extracts showed expression of Tlr-4 with and without LPS treatment. Each lane additionally was probed for α-actin expression as a control. Lanes 1 through 8 showed expression of α-actin.

DISCUSSION

Survival for patients diagnosed with GBM remains poor even after maximal treatment; therefore, novel treatment strategies are needed. We previously demonstrated that LPS leads to dramatic regression of malignant gliomas implanted subcutaneously in mice (6,38), but the mechanisms responsible for this antitumoral effect have not been well defined. The current study shows that the antitumoral effect of LPS against subcutaneously and intracranially implanted DBT glioblastoma cells is attenuated in Tlr-4 KO BALB/c mice compared with the antitumoral effect of LPS against similar tumors implanted in corresponding BALB/c wild-type mice. Therefore, these studies indicate that the antitumoral effect of LPS against glioblastoma is mediated, at least in part, by Tlr-4.

Since their description just a few years ago, the TLRs have been characterized in considerable detail and are known to play a critical role in the immunological responses to bacteria and other pathogens (1,7,12,21,23,29,33,37). In particular, Tlr-4 is known to mediate many of the effects of LPS on macrophages and other immune cell types. Before this article, the role of Tlr-4 in the antitumoral effects of LPS on glioblastoma had yet to be described.

As in our previous studies, the antitumoral effect of LPS on glioblastoma was more effective for tumors implanted subcutaneously compared with those implanted intracranially. This effect has been ascribed to the immune privilege of the central nervous system. There are other possible explanations for the differences in responses to LPS for glioblastomas implanted in the brain compared with those implanted subcutaneously. The central nervous system is also unique in its absence of a lymphatic system and in the presence of the blood-brain barrier. Mice are much less tolerant of LPS administered directly intracranially in comparison with LPS administered subcutaneously or by other routes. This further points to the unique and somewhat fragile nature of the central nervous system and may account for the less robust antitumoral effect of LPS for intracranial tumors. Despite the limitations in the central nervous system, a statistically significant increase in survival, albeit modest, was achieved in the wild-type BALB/c mice. The absence of this increase in survival in the Tlr-4 KO BALB/c mice indicates that Tlr-4 factors in the LPS-induced antitumoral response, even in the unique immunological environment of the central nervous system. We originally hypothesized that LPS may be acting directly on implanted glioblastoma cells through Tlr-4 or CD14 receptor pathways, or both. However, the quantitative flow cytometry data of glioblastoma cells in vitro and Western blot analysis of glioblastoma cells in vivo with and without LPS treatments showed no evidence of Tlr-4, and flow cytometry confirmed no CD14 expression, suggesting that this hypothesis was incorrect. Therefore, other cell types in the tumor-bearing host must bear Tlr-4, which accounts for the diminished antitumoral response to LPS in the Tlr-4 KO mice. This is in agreement with other investigations that have reported the presence of Tlr-4 activity in the central nervous system (8), particularly in microglia (18). These same investigators also reported the absence of Tlr-4 expression in astrocytes and oligodendroglial cells (18), which agrees with our flow cytometry and Western blotting studies, given that the glioblastoma cells likely originated in astrocytic or oligodendroglial progenitor cells.

The antitumoral effect of LPS on glioblastoma being mediated by Tlr-4 further supports the concept that the tumor-bearing host's immune system regulates the antitumoral effects of LPS. The role of the immune system in this response has been shown in our previous studies in which the antitumoral effect of LPS was attenuated in immunodeficient mice (38). Other investigators have also implicated T-cell immunity (3,27), production of tumor necrosis factor (2,9,10,14,19,27,28), interferons (11,27), and interleukins (2,10,11,14,19,27) in the antitumoral effects of LPS. Further analysis of the role of Tlr-4–expressing cells in the immunological response of the tumor-bearing host may provide further insights into immunotherapy strategies for patients with malignant gliomas. In addition to immune-mediated effects, LPS has been shown to have many other biological effects, including proapoptotic effects (34,36), antiapoptotic effects (5,36), and effects on cellular migration and motility (22). This cascade of complex and multifaceted effects of LPS may work in conjunction with and independent of the Tlr-4 immune-mediated effects of LPS. Further understanding of all of the different mechanisms involved in the antitumoral effects of LPS on glioblastoma multiforme is needed.

There were no histopathological differences in the tumors from the Tlr-4 KO BALB/c mice compared with the wild-type BALB/c mice, with the exception of the finding of a more robust LPS effect in the wild-type mice, manifested by smaller tumor sizes. The early neutrophilic and later macrophage-rich infiltrates in both types of mice indicate that the absence of Tlr-4 does not completely eliminate the immune-mediated antitumoral effect of LPS, but in some way does reduce its intensity. Further studies will be needed to improve understanding of the immune modulation of the antitumoral effect of LPS on glioblastoma by Tlr-4.

In summary, we have further elucidated the nature of the antitumoral mechanisms of LPS on glioblastoma and have shown that the recently described Tlr-4 receptor plays an important role in this process. The mechanisms by which Tlr-4 mediates the antitumoral effects of LPS on glioblastoma are in need of further investigation. Nonetheless, these results do suggest that the rapidly expanding topic of the TLRs may have applicability in the field of neuro-oncology and may provide novel treatment strategies for patients with malignant gliomas and other cancers.

CONCLUSION

This study provides the first evidence that Tlr-4 plays a role in the immune-mediated antitumoral effects of LPS on GBM in a mouse model. Further analysis of these mechanisms will provide clues as to how these innate antitumoral mechanisms may be harnessed most efficiently to treat patients with GBM and other cancers of the central nervous system.

References

1.
Barton
GM
,
Medzhitov
R
Toll-like signaling pathways
.
Science
 
300
:
1524
1525
,
2003
.
2.
Ben-Nathan
D
,
Padgett
DA
,
Loria
RM
Androstenediol and dehydroepiandrosterone protest mice against lethal bacterial infections and lipopolysaccharide toxicity
.
J Med Microbiol
 
48
:
425
431
,
1999
.
3.
Bowles AP Jr, Perkins
E
Long-term remission of malignant brain tumors after intracranial infection: A report of four cases
.
Neurosurgery
 
44
:
636
643
,
1999
.
4.
Byrd-Leifer
CA
,
Block
EF
,
Takeda
K
,
Akira
S
,
Ding
A
The role of MyD88 and TLR4 in the LPS-mimetic activity of Taxol
.
Eur J Immunol
 
31
:
2448
2457
,
2001
.
5.
Chao
W
,
Shen
Y
,
Zhu
X
,
Zhao
H
,
Novikov
M
,
Schmidt
U
,
Rosenzweig
A
Lipopolysaccharide improves cardiomyocyte survival and function after serum deprivation
.
J Biol Chem
 
280
:
21997
2005
,
2005
.
6.
Chicoine
MR
,
Won
EK
,
Zahner
M
Intratumoral injection of lipopolysaccharide causes regression of subcutaneously implanted mouse DBT glioblastoma multiforme
.
Neurosurgery
 
48
:
607
614
,
2001
.
7.
Dunne
A
,
O'Neill LA
The interleukin-1 receptor/Toll-like receptor superfamily: Signal transduction during inflammation and host defense
. Sci STKE 2003:re3,
2003
.
8.
Eklind
S
,
Mallard
C
,
Leverin
A
,
Gilland
E
,
Blomgren
K
,
Mattsby-Baltzer
I
,
Hagberg
H
Bacterial endotoxin sensitizes the immature brain to hypoxic-ischaemic injury
.
Eur J Neurosci
 
13
:
1101
1106
,
2001
.
9.
Engelhardt
B
,
Conley
FK
,
Kilshaw
PJ
,
Butcher
EC
Lymphocytes infiltrating the CNS during inflammation display a distinctive phenotype and bind toVCAM-1 but not to MAdCAM-1
.
Int Immunol
 
7
:
481
491
,
1995
.
10.
Engelhardt
R
,
Mackensen
A
,
Galanos
C
,
Andreesen
R
Biological response to intravenously administered endotoxin in patients with advanced cancer
.
J Biol Response Modif
 
9
:
480
491
,
1990
.
11.
Engelhardt
R
,
Otto
F
,
Mackensen
A
,
Mertelsmann
R
,
Galanos
C
Endotoxin (Salmonella abortus equi) in cancer patients
.
Clinical and immunological findings. Prog Clin Biol Res
 
392
:
253
261
,
1995
.
12.
Ewbank
JJ
,
Gorvel
J
,
Vivier
E
A widely conserved smell of danger
. Sci STKE 2003:pe9,
2003
.
13.
Flo
TH
,
Ryan
L
,
Latz
E
,
Takeuchi
O
,
Monks
BG
,
Lien
E
,
Halaas
O
,
Akira
S
,
Skjak-Braek
G
,
Golenbock
DT
,
Espevik
T
Involvement of Toll-like Receptor (TLR) 2 and TLR4 in cell activation by mannuronic acid polymers
.
J Biol Chem
 
277
:
35489
35495
,
2002
.
14.
Goto
S
,
Sakai
A
,
Kera
J
,
Suma
Y
,
Soma
G
,
Takeuchi
S
Intradermal administration of lipopolysaccharide in treatment of human cancer
.
Cancer Immunol Immunother
 
42
:
255
261
,
1996
.
15.
Kleihues
P
,
Cavenee
WK
(eds):
World Health Organization Classification of Tumours-Pathology and Genetics—Tumors of the Nervous System
.
Lyon, France
,
IARC Press
,
2000
, pp
9
52
.
16.
Kumanishi
T
Brain tumors induced with Rous sarcoma virus, Scmidt-Ruppin strain
.
I. Induction of brain tumors in adult mice with Rous chicken sarcoma cells. Jpn J Exp Med
 
37
:
461
474
,
1967
.
17.
Landis
SH
,
Murray
T
,
Bolden
S
,
Wingo
PA
Cancer statistics, 1999
.
CA Cancer J Clin
 
49
:
8
31
,
1999
.
18.
Lehnardt
S
,
Lachance
C
,
Patrizi
S
,
Lefebvre
S
,
Follett
PL
,
Jensen
FE
,
Rosenberg
PA
,
Volpe
JJ
,
Vartanian
T
The toll-like receptor is necessary for lipopolysaccharide-induced oligodendrocyte injury in the CNS
.
J Neurosci
 
22
:
2478
2486
,
2002
.
19.
Mackensen
A
,
Galanos
C
,
Engelhardt
R
Treatment of cancer patients with endotoxin induces release of endogenous cytokines
.
Pathobiology
 
59
:
264
267
,
1991
.
20.
Medzhitov
R
,
Preston-Hurlburt
P
,
Janeway
CA
Jr
A human homologue of the Drosophila Toll protein signals activation of adaptive immunity
.
Nature
 
388
:
394
397
,
1997
.
21.
Medzhitov
R
,
Janeway
CA
Jr
Self-defense: The fruit fly style
.
Proc Natl Acad Sci U S A
 
95
:
429
430
,
1998
.
22.
Mohan
K
,
Cordeiro
E
,
Vaci
M
,
McMaster
C
,
Issekutz
TB
CXCR3 is required for migration to dermal inflammation by normal and in vivo activated T cells: Differential requirements by CD4 and CD8 memory subsets
.
Eur J Immunol
 
35
:
1702
1711
,
2005
.
23.
Monick
MM
,
Powers
L
,
Butler
N
,
Yarovinsky
T
,
Hunninghake
GW
Interaction of matrix with integrin receptors is required for optimal LPS-induced MAP kinase activation
.
Am J Physiol Lung Cell Mol Physiol
 
283
:
L390
L402
,
2002
.
24.
Muroi
M
,
Ohnishi
T
,
Tanamoto
K
Regions of the mouse CD14 molecule required for toll-like receptor2- and 4-mediated activation of NF-Kappa B
.
J Biol Chem
 
277
:
42372
42379
,
2002
.
25.
Okamoto
M
,
Oshikawa
T
,
Ohe
G
,
Nishikawa
H
,
Furuichi
S
,
Tano
T
,
Moriya
Y
,
Saito
M
,
Sato
M
Severe impairment of anti-cancer effect of lipoteichoic acid-related molecule isolated from a penicillin-killed Streptococcus pyogenes in toll-like receptor 4-deficient mice
.
Int Immunopharmacol
 
1
:
1789
1795
,
2001
.
26.
Okamoto
M
,
Oshikawa
T
,
Tano
T
,
Ohe
G
,
Furuichi
S
,
Nishikawa
H
,
Ahmed
SU
,
Akashi
S
,
Miyake
K
,
Takeuchi
O
,
Akira
S
,
Moriya
Y
,
Matsubara
S
,
Ryoma
Y
,
Saito
M
,
Sato
M
Involvement of Toll-like receptor 4 signaling in interferon-gamma production and antitumor effect by streptococcal agent OK-432
.
J Natl Cancer Inst
 
95
:
316
326
,
2003
.
27.
Old
LJ
Tumor necrosis factor
.
Sci Am
 
258
:
59
75
,
1988
.
28.
Otto
F
,
Schmid
P
,
Mackesen
A
,
Wehr
U
,
Seiz
A
,
Braun
M
,
Galanos
C
,
Mertlesmann
R
,
Englehardt
R
Phase II trial of intravenous endotoxin in patients with colorectal and non-small cell lung cancer
.
Eur J Cancer
 
32A
:
1712
1718
, 1996.
29.
Poltorak
A
,
He
X
,
Smirnova
I
,
Liu
MY
,
Van Huffel
C
,
Du
X
,
Birdwell
D
,
Alejos
E
,
Silva
M
,
Galanos
C
,
Freudenberg
M
,
Ricciardi-Castagnoli
P
,
Layton
B
,
Beutler
B
Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: Mutations in Tlr4 gene
.
Science
 
282
:
2085
2088
,
1998
.
30.
Ponten
J
,
Macintyre
EH
Long term culture of normal and neoplastic human glia
.
Acta Pathol Microbiol Scand
 
74
:
465
486
,
1968
.
31.
Ralph
P
,
Nakoinz
I
Phagocytosis and cytolysis by a macrophage tumour and its cloned cell line
.
Nature
 
257
:
393
394
,
1975
.
32.
Raschke
WC
,
Baird
S
,
Ralph
P
,
Nakoinz
I
Functional macrophage cell lines transformed by Abelson leukemia virus
.
Cell
 
15
:
261
267
,
1978
.
33.
Rock
FL
,
Hardiman
G
,
Timans
JC
,
Kastelein
RA
,
Bazan
JF
A family of human receptors structurally related to Drosophila Toll
.
Proc Natl Acad Sci U S A
 
95
:
588
593
,
1998
.
34.
Schildberg
FA
,
Schulz
S
,
Dombrowski
F
,
Minor
T
Cyclic AMP alleviates endoplasmic stress and programmed cell death induced by lipopolysaccharides in human endothelial cells
.
Cell Tissue Res
 
320
:
91
98
,
2005
.
35.
Sherburn
EW
,
Kim
P
,
Wanebo
JA
,
Song
S
,
Chicoine
MR
,
Woolsey
TA
Gliomas in the rodent barrel cortex: A novel tumor model
.
J Neurosurg
 
91
:
814
821
,
1999
.
36.
Sherwin
C
,
Fern
R
Acute lipopolysaccharide-mediated injury in neonatal white matter glia: Role of TNF-alpha, IL-1beta, and calcium
.
J Immunol
 
175
:
155
161
,
2005
.
37.
Suzuki
N
,
Suzuki
S
,
Duncan
GS
,
Millar
DG
,
Wada
T
,
Mirtsos
C
,
Takada
H
,
Wakeham
A
,
Itie
A
,
Li
S
,
Penninger
JM
,
Wesche
H
,
Ohashi
PS
,
Mak
TW
,
Yeh
WC
Severe impairment of interleukin-1 and Toll-like receptor signaling in mice lacking IRAK-4
.
Nature
 
416
:
750
756
,
2002
.
38.
Won
EK
,
Zahner
MC
,
Grant
EA
,
Gore
P
,
Chicoine
MR
Analysis of the antitumoral mechanisms of lipopolysaccharide against glioblastoma multiforme
.
Anticancer Drugs
 
14
:
457
466
,
2003
.
39.
Xu
Y
,
Tao
X
,
Shen
B
,
Horng
T
,
Medzhitov
R
,
Manley
JL
,
Tong
L
Structural basis for signal transduction by the Toll/interleukin-1 receptor domains
.
Nature
 
408
:
111
115
,
2000
.

Acknowledgments

We thank Ralph G. Dacey, Jr., M.D., Department of Neurosurgery, Washington University, School of Medicine, for providing the resources for creating and maintaining our neuro-oncology research laboratory; Betsy Grant, Ph.D., Division of Biostatistics, Washington University School of Medicine, for performing statistical analyses; Michael M. C. Lai, Ph.D., Department of Molecular Microbiology and Immunology, University of Southern California, for his donation of the DBT glioblastoma cell line; Sherry Boeckelmann for assistance with manuscript preparation; and Claire Kramer and Karen Dodson, B.S., for editorial assistance.

COMMENTS

This experimental study supports the role of toll-like receptor 4 (TLR-4) in mediating lipopolysaccharide (LPS) antitumoral effects in wild-type and TLR-4 knockout mice models implanted with intracranial and subcutaneous glioblastoma. The TLR-4 knockout models showed less benefit from LPS than the wild-type models, suggesting a significant role for TLR-4 in vivo. For the subcutaneous tumors in knockout models, LPS had some effect on tumor regression but not as significant as that seen in the wild types. In vitro studies did not support such a role for TLR-4 when studied with glioblastoma cell lines, suggesting that the receptor requires environmental factors for its full effect, as seen in the in vivo models.

This study is an important contribution to the scientific literature and adds to our understanding of the complex cascades involved in tumor growth and apoptosis. The experiments are well done with clear results that support the hypothesis of the important, but not absolute, role of TLR-4 in mediating the antitumoral effect of LPS. However, as there was still some effect of LPS in the knockout models, there are obviously other pathways in addition to TLR-4 that mediate the effect of LPS on glioblastoma regression. Further research elucidating these other pathways must be conducted; the findings reported in the current study provide a strong base for pursuing such work.

Deborah T. Blumenthal

Neurologist, Salt Lake City, Utah

Philip H. Gutin

New York, New York

The authors present some very interesting work that, unfortunately, only suggests that LPS mediation by TLR-4 is an unlikely strategy for immune-mediated antitumor activity. The research is well done and the results are candidly presented. The modest improvement in the wild-type model with intracranial tumor has many causes. Further investigation into the reasons seems to be a long run for a short slide. Regardless, immunotherapy offers an exciting opportunity for glioma therapy. These data are important to direct future investigation.

Joseph M. Piepmeier

New Haven, Connecticut