Journal of the National Cancer Institute Advance Access originally published online on September 23, 2008
JNCI Journal of the National Cancer Institute 2008 100(19):1389-1400; doi:10.1093/jnci/djn308
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2008. Published by Oxford University Press.
ARTICLES |
A Genetically Enhanced Anaerobic Bacterium for Oncopathic Therapy of Pancreatic Cancer
Affiliations of authors: Departments of Gene and Cell Medicine (ZL, SLCW), Pathology (JF), Community and Preventive Medicine (JM), and Microbiology (JW), Mount Sinai School of Medicine, New York, NY
Correspondence to: Savio L. C. Woo, PhD, Department of Gene and Cell Medicine, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1496, New York, NY 10029-6574 (e-mail: savio.woo{at}mssm.edu).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
Background: A major obstacle in treatment of solid tumors is the inefficient delivery of therapeutic agents to the hypoxic cores. Hypoxia offers the potential for anaerobic bacteria colonization and tumor destruction by the bacteria, and dormant spores of wild-type Clostridium perfringens (Cp) germinate and proliferate within the hypoxic cores of pancreatic tumors in mice. However, the oncopathic effects of Cp were limited by host inflammatory responses and by Cp’s residual tolerance to oxygen, which caused toxic effects in animals.
Methods: Recombinant Cp strains in which superoxide dismutase, a major oxygen tolerance gene, was deleted (Cp/sod–) were constructed to enhance its selective growth in tumors. In addition, Panton-Valentine Leukocidin (PVL), an inflammation-suppressing gene from Staphylococcus aureus, was inserted into the Cp/sod– genome to enhance its oncopathic potency. The ability of the recombinant Cp strains to kill tumors was investigated in C57/BL6 mice bearing murine PANC02 tumors. Systemic and organ toxic effects were assessed by monitoring serum chemistries and histopathological examination. Statistical tests were two-sided.
Results: Cp/sod– showed reduced toxic effects compared with wild-type Cp when spores were administered intravenously into PANC02 tumor–bearing mice. Mice treated with Cp/sod–/PVL spores demonstrated a reduction in neutrophils and macrophages in tumors, logarithmically elevated growth of intratumoral bacteria, enhanced tumor necrosis, and substantially prolonged survival without apparent systemic and organ toxic effects, compared with mice treated with both wild-type Cp and Cp/sod– spores. Accordingly, 47% of Cp/sod–/PVL–treated mice (n = 15) achieved tumor-free survival for over 120 days, whereas all mice treated with Cp/sod– or phosphate-buffered saline (n = 10 per group) died within 50 days. The median survival for Cp/sod–/PVL–treated mice was 77 days (95% confidence interval [CI] = 45 to 120 days) and for Cp/sod––treated mice was 30 days (95% CI = 23 to 36 days; P < .001).
Conclusions: Cp/sod–/PVL provides a prototype for a novel class of oncopathic microbes that may have potential for the safe and effective treatment of pancreatic cancer and other poorly vascularized tumors.
Prior knowledge Live anaerobic bacteria can selectively grow in the hypoxic cores of tumors, but use of Clostridium perfringens (Cp) as a cancer therapy has been limited by toxic effects due to host inflammatory responses and to this bacterium’s residual tolerance to oxygen. Study design Mutant strains of Cp were constructed in which the superoxide dismutase (sod) gene was deleted to further enhance growth in tumors and in which the Panton-Valentine Leukocidin (PVL) gene from Staphylococcus aureus was inserted to limit severe inflammation by killing tumor-infiltrating macrophages and neutrophils. Spores of the double mutant were then compared with sod– and wild-type Cp in terms of their growth properties and abilities to kill pancreatic tumors in mice. Contribution Cp/sod–/PVL bacteria grew more rapidly and killed macrophages and neutrophils more effectively than previous strains in pancreatic tumors in treated mice. Tumor necrosis and overall survival of Cp/sod–/PVL–treated mice were substantially increased compared with mice treated with earlier strains of these bacteria. Implications Deletion of sod and insertion of PVL are improvements in the development of oncopathic Cp. Limitations Cp/sod–/PVL bacteria did not eliminate all tumor cells. They were tested on only one tumor type in mice and must be made less pathogenic before tests on humans can be considered. From the Editors
|
Hypoxic cores in poorly vascularized tumors are a major hindrance in cancer therapy because they prevent the effective delivery of therapeutic medications. However, the presence of severe hypoxia in solid tumors offers the potential for specific anaerobic bacterial colonization and tumor destruction (1). Most tumors, including pancreatic cancer, contain large, poorly vascularized hypoxic areas that limit the efficacy of radiation and chemotherapeutic drugs (2). Hypoxia is a tumor characteristic that is potentially exploitable using bioreductive drugs, such as tirapazamine (3), or gene therapy, including the use of engineered nonpathogenic anaerobic bacteria (4,5). Anaerobic bacteria are microbes that thrive in oxygen-deprived areas; not unexpectedly, their growth in well-oxygenated areas is severely limited. Compared with viruses or liposomes, bacteria have unique advantages as vectors for gene therapy: They can be engineered to carry more than one gene, they are easy to produce, they do not alter the genome of the recipient, and they can be eliminated by antibiotics once treatment is complete (6). Because solid tumor cores are oxygen deficient and the normal tissues in the body are more oxygen rich, anaerobic bacteria potentially could kill tumors while sparing the normal tissues, and their preferential replication in tumors potentially could amplify the therapeutic effect.
In the 1960s, live anaerobic bacteria were shown to target tumors and replicate within the hypoxic and necrotic regions of these tumors. The genus Clostridium was shown to cause tumor regression in rodent models (7), but a subsequent clinical trial failed to demonstrate any benefit in humans that would outweigh the toxic effects (8). In recent years, this concept has been reevaluated using attenuated Gram-negative Salmonella and Gram-positive obligate anaerobes such as Clostridium and Bifidobacterium, which were shown to selectively germinate and grow in the hypoxic regions of solid tumors after intravenous injection (9–13). Although all Clostridium species are Gram-positive bacteria that require anaerobic conditions for replication and spore germination, different strains vary in their oxygen tolerance and their biochemical profiles. In the past few years, several Clostridium species have been studied for their antitumor potential (14). Intracarotid injection with spores of a nonpathogenic Clostridium strain caused complete oncolysis in vascular glioblastomas, with liquefaction. Clostridia have also been genetically engineered to selectively deliver prodrug-activating enzymes such as Escherichia coli cytosine deaminase (11) and nitroreductase (15) to enhance treatment efficacy. More recently, Vogelstein et al. (16) investigated the tumor colonization properties of a variety of anaerobic bacteria, including eight Clostridium strains. They found Clostridium novyi with a deletion of its lethal toxin gene to be the best at colonizing tumors. The mutant strain of C. novyi devoid of its lethal toxin gene (C. novyi-NT) was shown to germinate and grow within the avascular regions of tumors and to destroy surrounding tumor cells in mice, although the treatment of tumor-bearing mice was still associated with high toxic effects, including acute lethality (16). More recently, the same group showed that treatment of mice with C. novyi-NT plus a single dose of liposome-encapsidated doxorubicin led to eradication of large tumors (17).
We have begun to investigate another anaerobic bacterial strain, Clostridium perfringens (Cp), for its ability to selectively colonize and induce necrosis in advanced pancreatic cancer, for which conventional treatment options such as chemotherapy and radiation therapy are limited (18–20). A primary and orthotopic pancreatic tumor mouse model was generated by direct implantation of the murine pancreatic cancer cell line PANC02 into immune-competent and syngeneic female C57/BL6 mice. Because the wild-type Cp strain retains some oxygen tolerance and can cause toxic effects in animals (21–23), we deleted the major gene associated with oxygen tolerance, the superoxide dismutase (sod) gene, from the Cp genome. Furthermore, because microbial replication in immune-competent hosts is inhibited by inflammatory cellular responses (24), we tested the hypothesis that the replication potency of Cp/sod– in tumors and treatment efficacy could be substantially elevated by constructing recombinant Cp strains that express an inflammation suppressive gene from a heterologous microbe, specifically the Panton-Valentine Leukocidin (PVL) gene of Staphylococcus aureus (25–27).
|
Materials and Methods |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
Bacterial Growth, Spore Purification and Oxidative Stress Sensitivity of Cp
C. perfringens Type A, Subgroup 1 (strain 13124, purchased from the American Type Culture Collection, Manassas, VA) was grown anaerobically in Reinforced Clostridial Medium (RCM) (Difco, Detroit, MI). The maximum yield of Cp spores occurred at 5 days following the transfer of vegetative bacteria to the sporulation medium, which contained yeast extract, 0.4%; proteose peptone, 1.5%; soluble starch, 0.4%; sodium thioglycolate, 0.1%; and Na2HPO4•7H2O, 1.0% (Difco) (28). The sporulating bacterial culture was incubated at 90°C for 10 minutes, suspended in 70% ethanol for 20 minutes, and centrifuged at 5000 rpm. The bacterial spores were extracted with 58% Renografin solution (Renocal-76 diluted in water) (Renocal-76, Princeton, NJ) by centrifugation at 18 000 rpm, then incubated with saline-EDTA containing lysozyme (100 µg/mL) at 37°C for 45 minutes, rinsed five times with phosphate-buffered saline (PBS), and stored at –20°C at 1 x 108 spores/mL as quantitatively determined by morphological identification.
Bacterial sensitivity to various oxidative stresses in vitro was evaluated using methods previously described (21). Aliquots of 1 x 104 per mL bacterial cells were diluted in PBS and exposed to the addition of H2O2, t-butylhydroperoxide, or plumbagin at various concentrations (0, 1, 3.3, 10, 33, 100, 333, and 1000 µM) for 1 hour. Alternatively, in vitro hypoxia was achieved by admixture of N2 and normal air inside a hypoxic incubator, and monitored with a Fisher Oakton DO-110 Oxygen Meter (29). Bacterial survival and growth were determined by incubating 1 x 104 log-phase vegetative bacterial cells per milliliter of RCM at oxygen concentrations including 0%, 0.1%, 0.33%, 1.0%, 3.3%, 10%, and 21%. Bacterial survival and growth under these conditions was determined by agar plating at 6 hours after exposure, in the mid-log phase of bacterial proliferation. The doses of the oxidative chemicals that caused 50% of bacteria death (LD50) were determined from the survival curves of bacterial cells using the StatPlus Mac 4.7.2 program (AnalystSoft software, Washington, DC). All determinations were repeated three times.
Orthotopic Pancreatic Cancer Model in Immune-Competent and Syngeneic Mice, Intravenous Bacterial Spore Injection, and Inflammatory Cell Depletion In Vivo
The PANC02 murine pancreatic cancer cell line was established from a murine pancreatic ductal adenocarcinoma in our previous study (30) and grown in DMEM (Mediatech, Herndon, VA) supplemented with 10% heat-inactivated FBS and 100 U/mL penicillin-streptomycin (Sigma-Aldrich, St Louis, MO). To generate a primary and orthotopic pancreatic tumor model for the proposed studies, 1 x 105 PANC02 cells that were harvested by brief trypsinization and resuspended in 10 µL of Earle's balanced salt solution buffer were directly injected into the pancreas of 6- to 8-week-old immune-competent and syngeneic female C57BL/6 mice under a protocol of the Institutional Animal Care and Use Committee (IACUC) of the Mount Sinai School of Medicine. Tumor nodules grew to greater than 5 x 5 mm in the pancreata of more than 90% of the animals in 18 days. Various doses (1 x 104 up to 1 x 108) of the purified bacterial spores in 0.2 mL of PBS were then injected via the tail vein. For depletion of neutrophils and natural killer (NK) cells, monoclonal rat anti-mouse polymorphonuclear neutrophils (RB6-8C5, BD Pharmingen, San Diego, CA) (31) and anti-mouse NK (PK-136, eBioscience, San Diego, CA) antibodies were used. Liposome-encapsidated dichloromethylene-bisphosphonate (Clodronate, kindly provided by Dr Nico van Rooijen, clodronateliposomes.org, The Netherlands) was used for macrophage depletion (24,32). To deplete neutrophils, NK cells, and macrophages in mice, 150 µg RB6-8C5, 100 µg PK-136, or 20 mg Clodronate, respectively, were administered intraperitoneally into PANC02 tumor–bearing mice 1 day before bacterial spore injection. Control mice were inoculated with an equal volume of normal rat IgG (Sigma, St Louis, MO) or PBS. Mice were then injected with Cp/sod– spores at the maximum tolerated dose (MTD) (1 x 107) and killed by CO2 at 2 days after injection. Tumor tissues were obtained by dissection, and bacterial titers in tumor extracts were determined by agar plating. Additionally, tumor sections were used for immunohistochemical staining for neutrophils, NK cells, and macrophages using rat anti-mouse Gr-1 monoclonal antibodies (BD Pharmingen), PK-136 monoclonal antibodies (eBioscience) and rat anti-mouse F4/80 monoclonal antibodies (AbD Serotec, Raleigh, NC), as well as hematoxylin and eosin (H&E) and Gram staining, followed by morphometric analysis (see below). All tumor-bearing mice in all animal experiments were determined to have palpable tumors before use in treatment experiments, with the criterion of palpation at 18 days after tumor cell implantation when the tumor nodules grow to 5 x 5 mm or greater.
Construction and Characterization of Cp/sod– and Cp/sod–/PVL Strains
The sod-deletion and PVL-containing strains were constructed through homologous recombination. All polymerase chain reaction (PCR) primers and probes used in this study were synthesized by Gene Link, Inc (Hawthorne, NY). The amplification, purification, ligation, and cloning of PCR products were performed with commercial kits from Roche (Roche Diagnostics, Belleville, NJ). Electroporation is the most efficient method for delivery of DNA fragments into Cp, which is a Gram-positive bacterium and difficult to transform by conventional chemical-based procedures. The method was optimized (data not shown) with bacterial cells at 1 x 108 per mL, transforming DNA at 1 µg/mL, pulse capacity of 25 µF, 600
, 6 KV/cm in 0.27 M sucrose, 5 mM Na2HPO4, and 1 mM MgCl2.
The sod-deletion Cp strain was constructed by replacing sod with firefly luciferase (LUC) by homologous recombination. The 1.8-kb LUC gene was amplified from a commercially available plasmid (Promega, Madison, WI) by PCR. To construct a homologous recombination fragment (an exogenous DNA fragment that is flanked by regions of DNA that are identical to the sequences flanking the gene to be replaced), the LUC gene and the upstream and downstream regions (each 0.6 kb) flanking the sod gene were amplified by PCR. The three fragments were assembled by overlapping PCR to generate one homologous recombination fragment. The entire fragment (LUC surrounded on each side by sod flanking sequences) was cloned in the pUC19 plasmid vector (Roche Diagnostics) and transformed into E. coli with the Expand Cloning Kit (Roche Diagnostics). The plasmid was then extracted and digested with the appropriate restriction enzymes (EcoRI and XbaI) to release the homologous recombination fragment. The fragment was introduced into Cp by optimized electroporation. The recombinant cells were screened by bacterial colony in situ hybridization with a LUC gene–specific probe and confirmed by DNA sequencing (33).
The PVL expressing strain was also constructed by homologous recombination. The PVL gene of S. aureus was inserted into the pyruvate ferredoxin oxidoreductase (pfoR) gene locus of Cp/sod– to construct Cp/sod–/PVL. The pfoR locus was chosen as an insertion site because this gene is highly expressed in Cp and is likely to have a strong promoter (34). The pfoR gene coding sequence (3516 bps) and up- and downstream sequences (each 0.8 kb) were amplified from Cp by PCR. The sequences of the two subunits of PVL gene, LukS and LukF, were amplified from S. aureus (ATCC strain 49775). The promoterless LukS and LukF sequences were linked together with a TAA stop codon between them. A cDNA fragment containing the pfoR leader (promoter and SD ribosome binding sequence) and coding region was cloned in Cp, as was its downstream sequence. The three fragments (pfoR, LukS-LukF, pfoR downstream) were linked by overlapping PCR (see Figure 3, A). This homologous recombination fragment was designed to insert the promoterless LukS–LukF coding sequences following the translation termination codon of pfoR gene, which allowed the transgene PVL to be driven by the native promoter of pfoR and expressed as a polycistronic mRNA, and to have no negative impact on the expression level of the resident gene. The composite fragment was cloned in E. coli, released from the amplified plasmids with the appropriate restriction enzymes (HindIII and EcoRI), and transformed into Cp/sod– by electroporation as described above. The colonies were screened by in situ colony hybridization using PVL-specific probes and the insertions in the recombinant strain, Cp/sod–/PVL, were sequenced.
The functional activity of PVL was assayed using serially diluted bacterial culture supernatants in a killing assay of mouse peripheral monocytes and neutrophils. C57/BL6 mouse peripheral monocytes and neutrophils were isolated from whole blood by immunomagnetic separation, suspended in RPMI-1640 medium, and distributed into 96-well plates (1 x 106/mL, 100 µL per well). The bacterial culture supernatants of Cp/sod–, Cp/sod–/PVL, or S. aureus, the parent strain of the PVL gene, were obtained at mid-log phase (1 x 106 bacterial cells/mL). The leukocyte cytotoxicity was determined by the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay, which is based on the ability of a mitochondrial dehydrogenase enzyme from viable cells to cleave the tetrazolium rings of the pale yellow MTT and form dark blue formazan crystals which are largely impermeable to cell membranes, thus resulting in accumulation within healthy cells. Solubilization of the cells by the addition of a detergent results in the liberation and solubilization of the crystals. The number of surviving cells is directly proportional to the level of the formazan product created. The color can then be quantified using a colorimetric assay. The crude bacterial culture supernatants were serially diluted 1:1, 1:10, 1:100, and 1:1000 with RPMI-1640 medium. The diluted bacterial culture supernatants and control blank RPMI-1640 medium were added into leukocyte plates (50 µL per well). The plates were incubated for 60 minutes at 37°C, and the results were read on a multiwell scanning spectrophotometer (ELISA reader) (35).
Determination of Antitumor Efficacy
The MTD for each bacterial strain in PANC02 tumor–bearing C57/BL6 mice was determined by intravenous injection of escalating doses of spores that ranged from 104 to 108 at half-log increments (n = 10). The endpoint was survival, defined by the time of death or by euthanasia, which was performed when the mice appeared distressed as defined by substantial weight loss, lethargy, or ruffled fur. The highest dose without death or euthanasia of animal was defined as the MTD. The antitumor efficacy of parental and recombinant Cp strains was determined by intravenous administration of bacterial spores at their respective MTD in mice bearing orthotopic pancreatic tumors, by measuring the survival time of the mice. Survival data were analyzed by the Kaplan–Meier method, and comparisons of survival curves between different groups were made by the log-rank test.
Evaluation of Intratumoral Content of Inflammatory Cells, HIF-1
Expression, Tumor Response, Bacterial Proliferation, and Toxic Effects
PANC02 tumor–bearing mice (n = 4 mice per group) were injected intravenously with 1 x 107 Cp/sod– or Cp/sod–/PVL bacterial spores diluted in PBS or with PBS as a negative control. The treated animals were killed by CO2 compressed gas at various time points including day 0, 1, 3, 7, and 14 after spore injection. The tumors, major organs (pancreas, liver, spleen, heart, lung, kidney, and bone marrow), and blood were then collected and divided. Half of each sample was homogenized and serially diluted in PBS. Bacterial titers in tumors, major organs and blood samples were quantified by agar plating.
Alternatively, to determine the distribution of bacteria among and the histopathology of tumor and normal tissues and to evaluate tumor response, the other half of each sample was fixed overnight in 4% paraformaldehyde, embedded in paraffin, and sectioned. Paraffin-embedded tissue sections were then subjected to Gram and H & E staining. To determine the number of inflammatory cells in tumors, tumor sections were subjected to immunohistochemical staining for neutrophils using rat anti-mouse Gr-1 monoclonal antibodies (BD Pharmingen) or for macrophages using rat anti-mouse F4/80 monoclonal antibodies (AbD Serotec) in addition to H & E and Gram staining, followed by morphometric and statistical analysis (see below). Additionally, a mouse anti-mouse HIF-1 mAb (H1a67, Novus Biologicals, Littleton, CO) was used for immunohistochemical staining of HIF-1 on paraffin-embedded tumor sections as described (36). All primary antibodies were incubated with samples at room temperature overnight and then detected using an avidin–biotin complex with 3,3–diaminobenzidine as the chromogen, using the VECTASTAIN Universal Quick kit (Vector Laboratories, Burlingame, CA). Sections were counterstained with Mayer's hematoxylin, mounted and examined by microscopy. For Gram, H & E, or each of the immunohistochemical stainings, one central section was analyzed for each tumor or tissue sample from each mouse. For the immunohistochemical stainings, positive staining in tissue sections was quantified using ImagePro Software (Media Cybernetics, Inc, Silver Spring, MD), and a positive staining cell index was calculated as the ratio of tumor-infiltrating cells per unit tumor area (with 10 000 pixels from a micrograph taken at x10 magnification and 300 dpi defined as one unit tumor area).
To evaluate systemic toxicity, complete blood counts (CBC), serum chemistries, and proinflammatory cytokine levels were monitored in blood samples from the mice killed at various time points. Red blood cell (RBC), white blood cell (WBC), hemoglobin, hematocrit, alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), and direct and indirect-bilirubin levels were analyzed in the animal facility research laboratory of the Mount Sinai School of Medicine. The serum levels of interleukin-12 (IL-12), interferon-gamma (IFN-
), and tumor necrosis factor-alpha (TNF-) were determined using the ELISA kits specific for each assay, respectively (R&D Systems, Minneapolis, MN). The results were analyzed by the unpaired t test or Kruskal–Wallis one-way analysis of variance (ANOVA). The major organs, including heart, lung, spleen, liver, kidney, pancreas, and bone marrow, were harvested for H & E staining to determine organ, histopathological changes by the pathologist.
Statistical Analysis
Long-term survival of mice was analyzed by the Kaplan–Meier method, and the survival curves were compared between groups by the use of the log-rank test. Statistical analysis for the MTD study was done by the Fisher exact test. Data from the kinetic study of intratumoral inflammatory cell contents, bacterial titers, tumor necrosis, and evaluation of systemic toxicity were analyzed by Kruskal–Wallis one-way ANOVA by ranks. All other statistical analyses were done using the unpaired Student t test. Data were shown as means with 95% confidence intervals (CIs). Analyses were performed using the GraphPad Prism 3.0 program (GraphPad Software, San Diego, CA). Differences were considered statistically significant at P < .05. All statistical tests were two-sided.
|
Results |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
Enhanced Safety and Efficacy of a Recombinant Cp Strain with Reduced Oxygen Tolerance
We first constructed a sod-deleted Cp strain, Cp/sod–, by homologous recombination in which the LUC gene replaced sod and confirmed the presence of the deletion by DNA sequencing. The in vitro bacterial growth rates, spore yields, and germination efficiencies did not differ substantially between Cp/sod– and Cp (data not shown). To compare the oxygen tolerance of the two strains, bacterial survival and growth were determined under defined oxygen concentrations from 0% to 21%, and sensitivities to various conditions of oxidative stress were evaluated in vitro. Cp/sod– exhibited a statistically significant reduction in bacterial survival for each of the chemical-induced oxidative stress conditions, as well as failure to grow at 10% oxygen, where Cp grew well (Figure 1, A), indicating that Cp/sod– is indeed less oxygen tolerant. For example, mean titers of 1 x 104 Cp cells exposed to 21% O2 for 6 hours were 2.1 x 103/mL, whereas titers of Cp/sod– cells under the same conditions were 3.0 x 102/mL (difference = 1.8 x 103/mL, 95% CI = 1.4 x 103 per mL to 2.2 x 103/mL, P = .006). An even greater difference was seen when the two bacterial strains were exposed to 10% O2 (Cp titer = 1.2 x 106/mL, Cp/sod– titer = 2.4 x 102/mL; difference = 119.9 x 104/mL, 95% CI = 88 x 104/mL to 152 x 104/mL, P < .001). When the two bacterial strains were exposed to the chemical oxidants plumbagin, H2O2, and t-butylhydroperoxide, the LD50 concentrations for Cp/sod– were substantially reduced from those for Cp (Figure 1, A), indicating that Cp/sod– was killed more readily than Cp.
|
To further evaluate the safety of the Cp/sod– strain, escalating doses of spores from the Cp/sod– and Cp strains were injected into the tail veins of immune-competent and syngeneic C57/BL6 mice bearing orthotopic murine PANC02 pancreatic tumors and mice were followed until they died or appeared distressed (data not shown). The MTDs of Cp/sod– and Cp were determined to be 1 x 107 and 1 x 106 spores, respectively, indicating an improvement in safety of Cp/sod– in vivo.
A long-term survival study was performed to assess treatment efficacy using 1 x 107 or 1 x 106 Cp/sod– spores compared with 1 x 106 Cp spores or a PBS control (Figure 1, B). A statistically significant prolongation of survival over the PBS control group was observed in mice treated with 1 x 107 spores of Cp/sod– (30-day survival with 1 x 107 Cp/sod– spores = 39%, 30-day survival with PBS = 0%, difference = 39%, 95% CI = 16% to 61%, P = .013 in log-rank test) but not in those treated with 1 x 106 spores of either strain (30-day survival with 1 x 106 Cp/sod– spores = 11%; 30-day survival with 1 x 106 Cp spores = 11%; difference with 1 x 106 Cp/sod– spores vs with PBS = 11%; 95% CI = 0% to 26%, P = .4 in log-rank test; difference with 1 x 106 Cp spores vs with PBS = 11%, 95% CI = 0% to 26%, P = .3 in log-rank test. Mice treated with 1 x 107 Cp/sod– spores also exhibited a statistically significant survival advantage over those treated with 1 x 106 spores of Cp or Cp/sod– (difference at 30-day with 1 x 107 Cp/sod– vs 1 x 106 Cp = 28%, 95% CI = 12% to 47%, P = .021 in log-rank test; and difference at 30-day with 1 x 107 Cp/sod– vs 1 x 106 Cp/sod– = 28%, 95% CI = 12% to 47%, P = .028 in log-rank test).
In another set of mice, the major organs, including liver and pancreas, were removed at days 0, 1, 3, and 7 after bacterial spore treatments to evaluate their histopathology by H & E staining and to determine bacterial replication by quantitative culture in vitro. There were neither pathological changes nor bacterial proliferation in the normal tissues after bacterial spore treatments at their MTDs (data not shown). These results provided proof of principle that recombinant Cp strains with reduced oxygen tolerance could be administered intravenously at higher doses than the parental strain without apparent toxic effects in vivo, which could lead to prolonged survival in mice bearing orthotopic pancreatic tumors. The extent to which survival was prolonged, however, was modest, and most treated animals relapsed over time (Figure 1, B).
Elevation of the Oncopathic Potency of Cp/sod– by Depletion of Host Inflammatory Cells In Vivo
Because bacterial infection is known to induce a robust inflammatory response in immune-competent hosts, we examined PANC02 pancreatic tumors in treated C57/BL6 mice for the accumulation of inflammatory cells. PANC02 tumor–bearing mice were injected with PBS or Cp/sod– at its MTD (1 x 107 spores), and sacrificed on day 2 after treatment. Tumor sections were analyzed by immunohistochemical staining for neutrophils, macrophages/monocytes, and NK cells using rat anti-mouse monoclonal antibodies (one section per mouse; four mice per group). The stained inflammatory cells were quantified by morphometry (Figure 2, A). After Cp/sod– spore administration, tumors accumulated statistically significantly higher levels of neutrophils and macrophages/monocytes but not of NK cells (per unit tumor area, neutrophils with Cp/sod– = 478 cells, with PBS = 126 cells, difference = 352 cells, 95% CI = 274% to 430%, P = .003; macrophages with Cp/sod– = 437 cells, with PBS = 163 cells, difference = 274 cells, 95% CI = 226% to 321%, P = .007; NK cells with Cp/sod– = 129 cells, with PBS = 112 cells, difference = 17 cells, 95% CI = –26% to 61%, P = .8). To determine the effect of macrophage, neutrophil, and NK cell depletion on bacterial replication and tumor response, PANC02 tumor–bearing mice were injected with 1 x 107 Cp/sod– spores together with clodronate-containing liposomes, which are known to deplete macrophages (24,32); with a monoclonal rat anti-mouse antibody to deplete neutrophils; or with a monoclonal rat anti-mouse antibody to deplete NK cells. After 2 days, tumor tissues were obtained and bacterial titers in tumor extracts were determined by agar plating. Additionally, tumor sections were used for immunohistochemical staining for neutrophils, macrophage/monocytes, and NK cells, as well as for Gram staining for bacteria and H & E for tumor necrosis measurements.
|
As expected, neutrophil, macrophage, and NK cell depletion effectively reduced neutrophil, macrophage/monocyte, and NK cell counts in tumors (Figure 2, B, top). The reductions in neutrophils and macrophages/monocytes were associated with a one-log elevation of intratumoral bacteria titers (Figure 2, B, middle panel) as well as a statistically significant enhancement in tumor necrosis (for neutrophil depletion, with antibody = 82%, 95% CI = 75% to 89%, with control IgG = 52%, 95% CI = 43% to 61%; P = .032; for macrophage depletion, with clodronate = 72%, 95% CI = 66% to 78%, with control liposome = 48%, 95% CI = 42% to 54%; P = .027) (Figure 2, B, bottom). However, the successful depletion of NK cells was not associated with any enhancement of intratumoral bacterial titers or tumor necrosis. Collectively, these results suggest that neutrophils and macrophages/monocytes, but not NK cells, were rapidly recruited to the tumor sites in response to intravenous Cp/sod– spore administration and were effective in attenuating intratumoral bacterial replication.
Construction and In Vitro Characterization of Cp/sod–/PVL
Cp/sod–/PVL, a recombinant Cp/sod– strain expressing the PVL gene from S. aureus that is known to cause direct damage to phagocytic cell membranes, was constructed by homologous recombination (Figure 3, A). We expected that Cp/sod–/PVL would not only better replicate in tumors because of the deletion of the sod gene but also better kill tumor-infiltrating macrophages and neutrophils because of the inclusion of the PVL gene. The proliferation, sporulation, and germination profiles of Cp/sod–/PVL were determined in vitro and showed no statistically significant differences from those of Cp and Cp/sod– (Figure 3, B). The functional activity of PVL was assayed by incubation of peritoneal macrophage and polymorphonuclear cells with the bacterial culture supernatants in vitro, followed by an inflammatory cell viability assay, as described in "Materials and Methods." Statistically significant reductions in both macrophage/monocyte and neutrophil cell survival were observed after incubation with non- and 10-fold–diluted supernatants from cultures of Cp/sod–/PVL or S. aureus (which normally expresses PVL), compared with equivalently diluted Cp/sod– culture supernatants (Figure 3, C). For example, when neutrophils were added 1:1 to undiluted Cp/sod–/PVL or S. aureus supernatants, 18% and 21%, respectively, of neutrophils survived, whereas when neutrophils were similarly added to Cp/sod–, 61% survived, (difference, Cp/sod–/PVL vs. Cp/sod– = 43%; 95% CI = 34% to 52%; P < .001). Similarly, when monocytes were added 1:1 to undiluted Cp/sod–/PVL or S. aureus supernatants, 22% and 17%, respectively, survived, whereas when Cp/sod– supernatants were used, only 58% survived (difference, Cp/sod–/PVL vs Cp/sod– = 36%; 95% CI = 24% to 48%, P = .008). These results suggest that PVL expressed from Cp/sod–/PVL was functional.
|
Oncopathic Potency and Treatment Efficacy of Cp/sod–/PVL in Tumor-Bearing Mice
In total, 1 x 107 spores of Cp/sod– or Cp/sod–/PVL, or PBS as a control, were administered by tail vein injection into PANC02 tumor–bearing mice (5 mice/group) that were sacrificed on day 2 after spore administration. Tumors were collected and sections were analyzed by histology, Gram staining, and immunohistochemical staining for neutrophils, macrophages/monocytes, and HIF-1 expression, followed by morphometric and statistical analyses. Bacterial titers in tumors were determined by quantitative bacterial culture from tumor extracts. Markedly more neutrophils (Figure 4, A) and macrophages/monocytes (Figure 4, B) were found in tumors from Cp/sod––treated mice than in PBS-treated control mice, and numbers of both types of inflammatory cells were substantially reduced in tumors from Cp/sod–/PVL–treated mice compared with those from Cp/sod––treated mice. Substantial reductions in HIF-1–positive cells were found in the tumor sections from both Cp/sod– and Cp/sod–/PVL–treated mice compared with PBS-treated mice (Figure 4, C). These findings were associated with substantially and statistically significant elevated bacterial titers (Figure 4, D) (bacterial titer in Cp/sod–/PVL = 1.4 x 105/mL, in Cp/sod– = 2.2 x 104/mL, difference = 1.2 x 105/mL, 95% CI = 0.9 x 105/mL to 1.4 x 105/mL, P = .029) and levels of necrosis in the bacterially treated tumors (Figure 4, E) (percentage in Cp/sod–/PVL = 65%, in Cp/sod– = 41%, difference = 24%, 95% CI = 16% to 32%, P = .037).
|
To determine the kinetics of intratumoral proinflammatory cell accumulation, bacterial proliferation, and tumor necrosis, PANC02 tumor–bearing mice were treated with 1 x 107 spores of Cp/sod–/PVL or 1 x 107 spores of Cp/sod–. The mice were sacrificed at days 0, 1, 3, 7, and 14 after spore injection (4 mice/group) for the collection of tumors, blood, and major organs. Neutrophils and macrophages/monocytes in Cp/sod–/PVL–treated mice showed substantial reductions at days 1, 3, and 7 after spore injection (Figure 5, A and B). Bacterial titers in Cp/sod–/PVL–treated tumors were enhanced by one or more logs relative to those in Cp/sod––treated tumors at days 3, 7, and 14 (Figure 5, C), and enhanced bacterial growth was associated with the substantially enhanced tumor necrosis on the same days (Figure 5, D).
|
The effectiveness of Cp/sod–/PVL in tumor treatment was then evaluated by survival studies (Figure 6). There was a statistically significant improvement in survival for PANC02 tumor–bearing mice treated with 1 x 107 Cp/sod– spores compared with PBS-treated controls (median survival for Cp/sod––treated mice = 30 days, 95% CI = 23 to 36 days; for PBS-treated mice = 22 days, 95% CI = 17 to 26 days; P = .014). However, despite their extended median survival, all Cp/sod––treated mice died with tumor relapse within 50 days. A substantial and statistically significant prolongation of survival was observed after treatment with 1 x 107 spores of Cp/sod–/PVL vs Cp/sod– (median survival for Cp/sod–/PVL–treated mice = 77 days, 95% CI = 45 to 120 days; for Cp/sod––treated mice = 30 days, 95% CI = 23 to 36 days; P < .001). Here, the added presence of the PVL gene was associated with a median survival time prolonged by 47 days, and 7 out of 15 (47%) of the Cp/sod–/PVL–treated mice remained alive at 120 days, at which time all remaining mice were killed and found to be tumor free. The experiment was repeated once with similar results (Supplementary Figure 1, available online). The median survival in mice treated with Cp/sod–/PVL was 94 days (95% CI = 66 to 120 days), with Cp/sod– was 35 days (95% CI = 27 to 43 days), and with PBS was 21 days (95% CI = 16 to 25 days). Eight out of 16 (50%) of the Cp/sod–/PVL–treated animals remained alive at 120 days. Using the log-rank test, statistically significant survival advantages were seen in mice treated with Cp/sod– and Cp/sod–/PVL spores, compared with those treated with PBS (P = .009 and P < .001, respectively), and were also seen in Cp/sod–/PVL–treated mice vs Cp/sod––treated mice (P < .001). Taken together, these results suggest that the PVL gene product was effective in suppressing the intratumoral accumulation of neutrophils and macrophages/monocytes, which permitted the bacteria to replicate to much higher titers in the lesions and led to considerably enhanced tumor necrosis and substantially prolonged animal survival.
|
Apparent Lack of Systemic and Organ Toxic Effects in Tumor-Bearing Mice Treated with Cp/sod–/PVL
To evaluate systemic toxic effects of the bacterial spore treatment, blood samples collected from mice treated in the kinetic study described above were analyzed for complete blood count, blood chemistries, described and serum proinflammatory cytokine levels (Supplementary Figure 2, A, available online). Based on the United States Medical Licensing Examination (USMLE) standard laboratory values, ALT, AST, BUN, direct-bilirubin and indirect-bilirubin values in all treated mice remained within their respective normal ranges at all time points, indicating normal liver and kidney functions. RBC, WBC, hemoglobin, and hematocrit values also remained within their normal ranges at all time points, indicating that there were no apparent hematologic toxic effects. Transient induction of IL-12, IFN-, and TNF-, which was anticipated as an initial inflammatory cytokine response to bacterial infection, was found in mice treated with Cp/sod– and Cp/sod–/PVL. However, maximal serum concentrations of these inflammatory cytokines were all below their respective toxic levels (37,38), and returned to pretreatment levels within 3 days. All major organs, including heart, lung, spleen, liver, kidney, pancreas, and bone marrow, were evaluated by H & E staining to determine their histopathology. There were no apparent pathologies in the nontumor-bearing regions of the pancreas or in the liver and spleen, organs that are close to the pancreas, at 0, 1, 3, and 14 days after treatment, in all mice injected with Cp/sod–/PVL spores (Supplementary Figure 2, B, available online). There were also no apparent pathological changes in any other organs at any of these time points. In vitro bacterial culture was also performed with extracts from all of these normal tissues, and no bacterial replication was found. These results indicated no systemic or organ toxic effects in mice after treatment with 1 x 107 Cp/sod–/PVL spores.
|
Discussion |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
In this article, we have reported the development of recombinant strains of Cp with reduced cytotoxicity and substantially enhanced oncopathic potency. After intravenous administration in immune-competent mice bearing orthotopic and syngeneic (murine) pancreatic cancer, the Cp spores were capable of preferential germination and proliferation within the hypoxic cores of tumors with oncopathic effects. However, the bacteria's residual tolerance to oxygen is such that they are still able to germinate and grow in normal tissues at reduced rates, which led to substantial toxic effects when administered intravenously at high doses in PANC02 tumor–bearing mice. The major gene associated with oxygen tolerance in Cp is the one encoding sod (21–23), and its deletion strain (Cp/sod–) exhibited reduced oxygen tolerance in vitro and limited growth in normal tissues in vivo, leading to a one-log elevation in its MTD in mice. Intravenous infusion of Cp/sod– in pancreatic tumor–bearing mice led to enhanced intratumoral bacteria replication, tumor regression, and prolonged survival of the mice. Similar responses were also observed in hepatic lesions with hypoxic cores in a mouse model of metastatic colorectal cancer (results not shown). However, the tumor response was modest and most treated animals succumbed to relapse over time. Therefore, more effective Cp strains need to be developed.
Macrophages and neutrophils have been shown to play key roles in early host defenses against infection by microbial pathogens (39). Neutrophils are first to accumulate in the infection sites and initiate cytolysis of invading microorganisms. Both local and circulating macrophages also migrate to infection sites to initiate phagocytosis. Neutrophils and macrophages have also been shown to kill clostridial bacterial spores in vitro, and thus can play another protective role in the host (24,40). In our studies, we demonstrated a substantial accumulation of neutrophils and macrophages/monocytes, but not of NK cells, in tumors after Cp/sod– spore administration. In addition, we found that reduced accumulation of neutrophils and macrophages was associated with elevated bacterial titers and necrosis in tumors, suggesting that the suppression of the inflammatory cell response could substantially enhance treatment efficacy in mice.
Many invading microbes have evolved clever strategies for evading and/or suppressing host inflammatory responses. Pathogenic bacteria are able to avoid being killed by phagocytic engulfment (41). PVL, produced by S. aureus, can directly damage membranes of phagocytes including monocytes, macrophages, and neutrophils (25–27). To enhance oncopathic potency and treatment efficacy, Cp/sod–/PVL, a recombinant Cp/sod– strain expressing PVL, was constructed. Treatment of mice with the Cp/sod–/PVL strain, compared with Cp/sod– treatment, led to a substantial reduction in intratumoral inflammatory cells, to logarithmically elevated intratumoral bacteria titers, to enhanced tumor necrosis, and to substantially prolonged survival. Importantly, a substantial fraction (47%) of Cp/sod–/PVL–treated mice remained alive after 120 days and there were no apparent systemic and organ toxic effects associated with the systemic administration of Cp/sod–/PVL spores at the effective dose.
There are a number of limitations in the use of Cp spores to treat tumors. Firstly, the antitumor efficacy of the recombinant anaerobic bacterial spores was evaluated only in one type of cancer here, and its general applicability to other types of hypoxic tumors will need to be validated in relevant animal models. Second, although effective, the treatment with Cp/sod–/PVL spores did not eliminate all tumor cells in all nodules and did not always lead to tumor rejection; as a result, tumors did recur and led to the death of about half of the mice. A rim of viable tumor cells was often seen around the necrotic cores in the lesions of mice treated with bacterial spores (data not shown). We suspect that the anaerobic bacteria could not effectively replicate in the peritumoral regions, which are relatively oxygen rich by virtue of their access to blood supply. In contrast, tumor cells located in those well-vascularized regions are susceptible to chemotherapeutic drugs distributed via the circulation. Therefore, we further hypothesize that the anticancer efficacy of Cp/sod–/PVL spores could be complemented by the ability of chemotherapeutic drugs to destroy tumor cells in the well-vascularized and oxygen-rich regions, since the two treatments target tumor cells in different regions. Thus, the combinatorial use of these complementary agents may be better able to target tumor cells in both the well- and poorly oxygenated regions of the solid tumors and lead to substantially enhanced treatment efficacy. Third, the effectiveness of inflammation suppression to substantially enhance the oncopathic potency of Cp has only been observed in a murine tumor model, and whether the same responses will occur in humans is unknown. Fourth, Cp is a human pathogenic microbe and expresses phospholipase C (plc), which is the major virulence determinant in Cp that is known to be causative in development of gas gangrene in rodents and in humans. A deletion of the plc gene in Cp/sod–/PVL will need to be done to eliminate its gas gangrene activity and to substantially enhance treatment safety before clinical translational trials can be contemplated in the future. Finally, because the oncopathic bacteria are replication competent, the ability to completely shut down their replication by antibiotic treatment must be documented in tumor-bearing animals before the initiation of any clinical translational studies.
Taken together, this newly constructed bacterial strain, Cp/sod–/PVL, has substantially elevated tumor selectivity and oncopathic potency. These improvements may lead to the development of safe and effective oncopathic agents for the treatment of patients with pancreatic cancer and other poorly vascularized tumors in the future.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
This research was supported by a National Institutes of Health grant CA-120017. We wish to thank Ms Marcia Meseck, Dr Tian-gui Huang and Dr Lan Wu for helpful discussions, and Ms Yafang Wang and Mr Boxun Xie for technical assistance.
The authors take full responsibility for the design of the study, the collection of the data, the analysis and interpretation of the data, the decision to submit the manuscript for publication, and the writing of the manuscript.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
1. Critchley RJ, Jezzard S, Radford KJ, et al. Potential therapeutic applications of recombinant, invasive E. coli. Gene Therapy (2004) 11(15):1224–1233.[CrossRef][Web of Science][Medline]
2. Kondoh S, Inoue M, Harada H, Hiraoka M. Tumor hypoxia: a target for selective cancer therapy. Cancer Sci (2003) 94(12):1021–1028.[CrossRef][Medline]
3. Brown JM. Tumor microenvironment and the response to anticancer therapy. Cancer Biol Ther (2002) 1(5):453–458.[Web of Science][Medline]
4. Theys J, Barbe S, Landuyt W, et al. Tumor-specific gene delivery using genetically engineered bacteria. Curr Gene Ther (2003) 3(3):207–221.[CrossRef][Medline]
5. Pawelek JM, Low KB, Bermudes D. Bacteria as tumour-targeting vectors. Lancet Oncol (2003) 4(9):548–556.[CrossRef][Web of Science][Medline]
6. Sznol M, Lin SL, Bermudes D, Zheng L, King I. Use of preferentially replicating bacteria for the treatment of cancer. J Clin Invest (2000) 105(8):1027–1030.[Web of Science][Medline]
7. Engelbart K, Gericke D. Oncolysis by clostridia. Transplanted tumours of the hamster. Cancer Res (1964) 24(2):239–243.
8. Carey RW, Holland JF, Whang HY, Neter E, Bryant B. Clostridial oncolysis in man. Eur J Cancer (1967) 3(3):37–46.
9. Ryan RM, Green J, Lewis CE. Use of bacteria in anti-cancer therapies. Bioessays (2006) 28(1):84–94.[CrossRef][Web of Science][Medline]
10. Pawelek JM, Low KB, Bermudes D. Tumor-targeted Salmonella as a novel anti-cancer vector. Cancer Res (1997) 57(20):4537–4544.
11. Theys J, Landuyt W, Nuyts S, Van Mellaert L, van Oosterom A, Lambin P. Specific targeting of cytosine deaminase to solid tumors by engineered Clostridium acetobutylicum. Cancer Gene Ther (2001) 8(4):294–297.[CrossRef][Web of Science][Medline]
12. Forbes NS. Profile of a bacterial tumor killer. Nat Biotechnol (2006) 24(12):1484–1485.[CrossRef][Web of Science][Medline]
13. Yazawa K, Fujimori M, Nakamura T, et al. Bifidobacterium longum as a delivery system for gene therapy of chemically induced rat mammary tumors. Breast Cancer Res Treat (2001) 66(2):165–170.[CrossRef][Web of Science][Medline]
14. Heppner F, Mose JR. The liquefaction (oncolysis) of malignant gliomas by a non pathogenic Clostridium. Acta Neurochir (1978) 42(1–2):123–125.[CrossRef][Medline]
15. Lemmon MJ, van Zijl P, Fox ME, Mauchline ML, Giaccia AJ, Minton NP. Anaerobic bacteria as a gene delivery system that is controlled by the tumor microenvironment. Gene Ther (1997) 4(88):791–796.[CrossRef][Web of Science][Medline]
16. Dang LH, Bettegowda C, Huso DL, Kinzler KW, Vogelstein B. Combination bacteriolytic therapy for the treatment of experimental tumors. Proc Natl Acad Sci U S A (2001) 98(26):15155–15160.
17. Cheong I, Huang X, Bettegowda C, et al. A Bacterial protein enhances the release and efficacy of liposomal cancer drugs. Science (2006) 314(5803):1308–1311.
18. Casper ES. Pancreatic cancer: how can we do better? Eur J Cancer (1993) 29(2):171–172.[CrossRef]
19. Storniolo AM, Enas NH, Brown CA. An investigational new drug treatment program for patients with gemcitabine. Cancer (1999) 85(6):1261–1268.[CrossRef][Web of Science][Medline]
20. Stephens CD. Gemcitabine: a new approach to treating pancreatic cancer. Oncol Nurs Forum (1998) 25(1):87–93.[Medline]
21. Briolat V, Reysset G. Identification of the Clostridium perfringens genes involved in the adaptive response to oxidative stress. J Bacteriol (2002) 184(99):2333–2343.
22. Geissmann TA, Teuber M, Meile L. Transcriptional analysis of the rubrerythrin and superoxide dismutase genes of Clostridium perfringens. J Bacteriol (1999) 181(22):7136–7139.
23. Lehmann Y, Meile L, Teuber M. Rubrerythrin from Clostridium perfringens: cloning of the gene, purification of the protein, and characterization of its superoxide dismutase function. J Bacteriol (1996) 178(24):7152–7158.
24. Cote CK, Rea KM, Norris SL, Van Rooijen N, Welkos SL. The use of a model of in vivo macrophage depletion to study the role of macrophages during infection with Bacillus anthracis spores. Microb Pathog (2004) 37(4):169–175.[CrossRef][Web of Science][Medline]
25. Kato I. Structure and action of staphylococcal alpha-toxin and leukocidin. Nippon Saikingaku Zasshi (1981) 36(2):445–457.[Medline]
26. Okumoto M. The immunopathogenesis of staphylococcal infections. Int Ophthalmol Clin (1985) 25(2):133–142.[Medline]
27. Genestier AL, Michallet MC, Gilles P, et al. Staphylococcus aureus Panton-Valentine leukocidin directly targets mitochondria and induces Bax-independent apoptosis of human neutrophils. J Clin Invest (2005) 115(11):3117–3127.[CrossRef][Web of Science][Medline]
28. Gill CO, Penney N, Wauters AM. Survival of clostridial spores in animal tissues. Appl Environ Microbiol (1981) 41(1):90–92.
29. Vordermark D, Brown JM. Evaluation of hypoxia-inducible factor-1 alpha as an intrinsic marker of tumor hypoxia in U87 MG human glioblastoma: in vitro and xenograft studies. Int J Radiat Oncol Biol Phys (2003) 56(4):1184–1193.[CrossRef][Web of Science][Medline]
30. Block A, Chen SH, Kosai KI, Finegold MJ, Woo S. Adenoviral-mediated herpes simplex virus thymidine kinase gene transfer: regression of hepatic metastasis of pancreatic tumors. Pancreas (1997) 15(1):25–34.[Web of Science][Medline]
31. Rousseau D, Demartino S, Ferrua B, et al. In vivo involvement of polymorphonuclear neutrophils in Leishmania infantum infection. BMC Microbiol (2001) 1(17):1471–2108.
32. Rooijen NV, Sanders A, Berg TK. Apoptosis of macrophages induced by liposome-mediated intracellular delivery of clodronate and propamidine. J Immunol Methods (1996) 193(1):93–99.[CrossRef][Web of Science][Medline]
33. Ezaki T, Xu HX, Li Z, Zhao L, Kawamura Y. Molecular Ecological Detection and Identification of Intestinal Microflora (2000) Tokyo, Japan: Japan Scientific Societies Press. 214–218.
34. Karlin S, Theriot J, Mrazek J. Comparative analysis of gene expression among low G_C gram-positive genomes. Proc Natl Acad Sci USA (2004) 101(16):6182–6187.
35. Dutta A, Bandyopadhyay S, Mandal C, Chatterjee M. Development of a modified MTT assay for screening antimonial resistant field isolates of Indian visceral leishmaniasis. Parasitol Int (2005) 54(2):119–122.[CrossRef][Web of Science][Medline]
36. Sun X, Kanwar JR, Leung E, Lehnert K, Wang D, Krissansen GW. Gene transfer of antisense hypoxia inducible factor-1 alpha enhances the therapeutic efficacy of cancer immunotherapy. Gene Ther (2001) 8(8):638–645.[CrossRef][Web of Science][Medline]
37. Lyke KE, Burges R, Cissoko Y, et al. Serum levels of the proinflammatory cytokines interleukin-1 beta (IL-1beta), IL-6, IL-8, IL-10, tumor necrosis factor alpha, and IL-12(p70) in Malian children with severe Plasmodium falciparum malaria and matched uncomplicated malaria or healthy controls. Infect Immun (2004) 72(10):5630–5637.
38. Wan JM, Sit WH, Lee CL, Fu KH, Chan DK. Protection of lethal toxicity of endotoxin by Salvia miltiorrhiza BUNGE is via reduction in tumor necrosis factor alpha release and liver injury. Int Immunopharmacol (2006) 6(5):750–758.[CrossRef][Web of Science][Medline]
39. Cote CK, Van Rooijen N, Welkos SL. Roles of macrophages and neutrophils in the early host response to Bacillus anthracis spores in a mouse model of infection. Infect Immun (2006) 74(1):469–480.
40. Welkos SL, Trotter RW, Becker DM, Nelson GO. Resistance to the Sterne strain of B. anthracis: phagocytic cell responses of resistant and susceptible mice. Microb Pathog (1989) 7(1):15–36.[CrossRef][Web of Science][Medline]
41. Galan GE. Bacterial toxins and the immune system: show me the in vivo targets. J Exp Med (2005) 201(3):321–323.
Manuscript received November 30, 2007; revised July 10, 2008; accepted July 30, 2008.
Related Article in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 1337.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||