Sodium methyldithiocarbamate (SMD; trade name, Metam Sodium) is an abundantly used soil fumigant that can cause adverse health effects in humans, including some immunological manifestations. The mechanisms by which SMD acts, and its targets within the immune system are not fully understood. Initial experiments demonstrated that SMD administered by oral gavage substantially decreased IL-12 production and increased IL-10 production induced by lipopolysaccharide in mice. The present study was conducted to further characterize these effects and to evaluate our working hypothesis that the mechanism for these effects involves alteration in signaling through toll-like receptor 4 and that this would suppress innate immunity to infection. SMD decreased the activation of MAP kinases and AP-1 but not NF-κB in peritoneal macrophages. The expression of mRNA for IL-1α, IL-1β, IL-18, IFN-γ, IL-12 p35, IL-12 p40, and macrophage migration inhibitory factor (MIF) was inhibited by SMD, whereas mRNA for IL-10 was increased. SMD increased the IL-10 concentration in the peritoneal cavity and serum and decreased the concentration of IL-12 p40 in the serum, peritoneal cavity, and intracellularly in peritoneal cells (which are >80% macrophages). Similar effects on LPS-induced cytokine production were observed following dermal administration of SMD. The major breakdown product of SMD, methylisothiocyanate (MITC), caused similar effects on cytokine production at dosages as low as 17 mg/kg, a dosage relevant to human exposure levels associated with agricultural use of SMD. Treatment of mice with SMD decreased survival following challenge with non-pathogenic Escherichia coli within 24–48 h, demonstrating suppression of innate immunity.
Sodium methyldithiocarbamate (SMD) is the third most abundantly used conventional pesticide in the U.S. (U.S. EPA, 2001). In 1991, a train derailment near Dunsmuir, CA, caused a spill of 19,000 gallons of a 32% solution of SMD (Diringer, 1991). Over 700 persons in the area sought medical attention for symptoms ranging from nausea and dizziness to irritation of the eyes and upper respiratory tract (Alexeeff et al., 1994). A number of persons reported exacerbation or induction of asthma following exposure (Cone et al., 1994). Major exposure events have also occurred after routine agricultural use of SMD, leading to the evacuation of schools and of a large portion of the town of Earlimart, CA (O'Malley et al., 2004). In addition, >90,000 people in California are exposed each year to doses of the major breakdown product of SMD, methylisothiocyanate (MITC), in excess of safe dose levels (Lee et al., 2002). Considering that only about 20% of total SMD use in the U.S. occurs in California (Thongsinthusak, 2004), the actual number of exposures is almost certainly larger. The Reference doses for SMD and MITC are based on rodent studies and application of appropriate uncertainty factors, and they seem to be quite accurate with regard to actual measured effects in human subjects (O'Malley et al., 2004; Weiss and Lowit, 2004). This is of interest because the dosages used in the present study encompass the dosages used in rodent studies to determine the human reference dose (Weiss and Lowit, 2004).
In previous studies, we demonstrated that SMD suppresses several immune system parameters in mice (Keil et al., 1996; Myers et al., 2005; Padgett et al., 1992; Pruett et al., 1992), but its effects on the production of cytokines by macrophages have not been investigated. The exacerbation of asthma caused by SMD (Cone et al., 1994) suggests the possibility that it may shift the predominance of T helper cells (Th) from Th1 to Th2. The predominance of Th1 vs. Th2 cells and their corresponding cytokines is determined in part by the predominance of cytokines that support activation of Th1 cells (e.g., IL-12 or IFN-γ) compared to cytokines that favor the predominance of Th2 cells (e.g., IL-10). These cytokines are produced primarily by antigen presenting cells such as macrophages and dendritic cells (Dalod et al., 2002; Trinchieri, 1993; Yi et al., 2002). Pro-inflammatory cytokines produced by macrophages are also critically important in innate resistance to infection (Weighardt et al., 2000). In particular, peritoneal macrophages are critical in resistance to peritonitis (Dunn et al., 1987; Matsukawa et al., 1999; Vuopio-Varkila, 1988), and the macrophage-derived cytokines IL-12 and IL-10 are also involved (Sewnath et al., 2001; Takano et al., 1998; Zisman et al., 1997).
Following our initial observation in this study that SMD alters LPS-induced cytokine production by peritoneal macrophages, we developed the working hypothesis that this might be mediated by inhibition of cellular signaling and that it might lead to suppression of innate immunity. The present study was conducted to evaluate this working hypothesis. The major signaling mechanisms activated by LPS through TLR4 are NF-κB and MAP kinases, which activate several transcription factors, of which AP-1 is representative. AP-1 is also important in the expression of the cytokines investigated here. Therefore, NF-κB, MAP kinases, and AP-1 were investigated. Early events in TLR4 signaling include binding of an adapter molecule (generally MyD88) to the receptor (Kawai et al., 1999) followed by activation of interleukin 1 receptor associated kinases (IRAK) 1 and 4 (Li et al., 2002). Thus, the effects of SMD on IRAK-1 and MyD88 were examined to determine if it acts early in the signaling pathway. To evaluate innate immunity, a model in which the cell type under investigation here (peritoneal macrophages) is known to be important was selected. Thus, resistance to peritonitis caused by injection of large numbers of non-pathogenic Escherichia coli was measured.
MATERIALS AND METHODS
Mice and administration of test agents.
Female C57Bl/6×C3H F1 (B6C3F1) mice were used in this study at 8–12 weeks of age. They were obtained through the National Cancer Institute's animal program. The mice were allowed to acclimate and recover from shipping stress for at least two weeks before being used in experiments. Mice were housed in a climate-controlled environment with a 12 h light dark cycle. They were given food (Purina Lab Chow) and water ad libitum. The animal facility and animal care and use procedures are AAALAC approved. All animal care and use was in accord with the NIH Guide and the regulations of LSUHSC. In one experiment, MyD88 knockout mice on a C57BL/6 background were used. These mice were kindly provided by Dr. H. Hemmi (Hemmi et al., 2003). In this experiment heterozygous C57Bl/6 mice were used as controls.
Mice were treated with bacterial lipopolysaccharide (LPS, from Escherichia coli 0128:B12 from Sigma Chemical Co. or ultrapure LPS from Salmonella minnesota from List Labs) by iv injection in a lateral tail vein. Standard LPS is referred to as LPS, and ultrapure LPS is referred to as pure or ultrapure LPS in subsequent sections. The dosage was 60 μg/mouse, a dosage we have previously shown to induce cytokines but not to be lethal (Pruett et al., 2004b). Sodium methyldithiocarbamate (SMD) was obtained from Chem Service, Inc. and methylisothiocyanate (MITC) was obtained from Sigma Chemical. They were dissolved in tissue culture grade water. Dissolving MITC required heating to 56°C. After it was dissolved, the solution was separated into aliquots and stored frozen until use. SMD was prepared fresh for each experiment. Both solutions were filter sterilized before use. Mice were treated with SMD or MITC by oral gavage. In one experiment mice were treated with the TLR3 agonist polyinosinic polycytidylic acid (poly I:C) at 100 μg/mouse, intravenously as in our previous studies (Pruett et al., 2004a).
For host resistance studies, mice were challenged by ip injection of log phase non-pathogenic Escherichia coli (E. coli), as described previously (Pruett et al., 2004b). The E. coli strain used was isolated from one of the mice in our colony, and its identity was confirmed using the Marieux Vitek gram negative identification system. The dosages of E. coli required to cause lethal peritonitis were found to be comparable to those reported by other investigators (Takano et al., 1998; Zhao et al., 2001). After challenge, the mice were observed at least every 6 h and animals that were moribund were euthanized and counted as non-survivors. However, in the experiments shown most animals died between observations and most non-survivors shown in the graphs represent animals that died rather than animals that were euthanized.
Peritoneal fluid and cells were obtained by peritoneal lavage using phosphate buffered saline with 10% fetal bovine serum. First, a lavage was performed using 1 ml of fluid with typical recoveries of 0.6–0.7 ml. After centrifugation, the supernatant fluid was used to quantify cytokines without the dilution effect that would result from a larger volume of lavage fluid. To assure complete recovery of peritoneal cells, a second lavage was performed using 7 ml of lavage fluid. After centrifugation, the cells from both lavages were pooled. Cell counts (by Coulter counter) and differential counts in this and previous studies (Pruett et al., 2004a,b), demonstrate that typical yields of cells are 2–3 × 106/mouse, and typical differential counts are 80–95% macrophages, 4–15% lymphocytes, 1–5% neutrophils, and occasional basophils and eosinophils.
Serum corticosterone analysis.
Serum from mice treated dermally with SMD was analyzed for corticosterone to determine if dermal application induces a neuroendocrine stress response, as does po administration (Myers et al., 2005). Mice were bled by decapitation within 3 min from the time the cage was obtained from the animal room to prevent handling-induced stress. Corticosterone was analyzed using a radioimmunoassay kit from DPC (Los Angeles, CA), as in our previous studies (Pruett et al., 1999, 2000, 2003c).
Cytokine and chemokine analysis.
Samples were analyzed for quantities of IL-10, IL-12 (p40), and CXCL9 using matched ELISA reagents from BD Pharmingen. In one experiment IL-12 (p70) and IL-10 were analyzed using a BioPlex multiplexed bead array kit from BioRad. For analysis of intracellular cytokines, peritoneal cells from each mouse were resuspended in 120 μl of PBS containing 0.5% Tween 20 and sonicated for 15 s using a needle-type probe. After centrifugation at 15,000 × g for 10 min, the supernatant fluid was removed and analyzed for cytokines.
Western blot and Bioplex analysis of signaling.
For signaling studies peritoneal cells were isolated quickly using ice cold buffer, and cells were kept on ice for no more than 30 min before centrifugation (at 4°C). Whole cell extracts or nuclear extracts were prepared from peritoneal cell pellets just as described in our previous studies (Pruett et al., 2003b, 2004a). In a previous study, we determined that signaling parameters were not significantly affected following even longer periods of time than this, if they were kept cold (Pruett et al., 2003b). Protein in each sample was quantified using the BCA method (Pierce Chemical) with bovine serum albumin as a standard. An equivalent amount of protein from each mouse (10 μg for NF-κB and 20 μg for others) was mixed with sample buffer, subjected to SDS-PAGE, blotted onto a PVDF membrane, labeled with primary and secondary antibodies (anti-NF-κB, p65, #C-20; anti-IRAK-1, #H-273; peroxidase coupled anti rabbit polyclonal IgG secondary antibody, #SC-7883; Santa Cruz Biotechnology), and bands were developed using the ECL chemiluminescence system (Amersham). The membrane was then used to expose photographic film. Proteins were quantitated using NIH Image software and the included macro for gel analysis.
RNAse Protection Assay.
Expression of cytokine mRNA was analyzed using a probe set from BD Pharmingen (RiboQuant mCK2b). RNA was isolated from all cells obtained from each mouse using Tri Reagent and quantified by the ratio of absorbance at 280 and 260 nm. RNAse protection assay was done just as described in our previous studies (Pruett et al., 2003a, 2004a; Zheng and Pruett, 2000), and X-ray films of gel autoradiographs were quantified using NIH Image software with the included macro for analysis of bands on gels. A long exposure film was used to increase sensitivity for quantitation of mRNA expressed at low levels and a short exposure film was used to prevent saturation of mRNA bands expressed at high levels as well as housekeeping mRNAs L32 and GAPDH.
Data were analyzed by analysis of variance followed by the Student Newman Keul's (SNK) post hoc test to compare each mean to every other mean. Comparisons in which p < 0.05 were regarded to be significantly different. In most figures, values significantly different from the LPS or poly I:C only group are indicated by asterisks. Survival analysis was done with the Log-rank test. All analyses were done using Prism 4.0 software (GraphPad Software, San Diego, CA).
Dose-Responsive Changes in Cytokine Production Caused by SMD and MITC
Results shown in Figure 1 indicate that po administration of SMD dose responsively decreases the IL-12 (p40) response to LPS in serum and in the peritoneal cavity. The lowest dosage of SMD that caused a significant change was 50 mg/kg (decreased IL-12 in serum). Results for cytokines in serum in mice treated with MITC indicated dose responsive decreases in IL-12 concentration, with a significant decrease at a dosage of 17 mg/kg. This dosage was selected because it encompasses the dosage that would be received by a child near a field where SMD was applied (1.0–2.5 mg/kg) (Thongsinthusak, 2000), if one calculates dosages on the basis of body surface area (calculated as described in our previous study) (Padgett et al., 1992) and does not apply any other uncertainty factors. Significant increases in IL-10 in serum and peritoneal fluid occurred at higher dosages than required to suppress IL-12, but the overall dose response patterns were similar. Treatment of mice with SMD alone at 250 mg/kg did not affect basal IL-12 (p40) concentrations in blood or in the peritoneal cavity. The basal concentration of IL-10 in the peritoneal cavity also was not affected, but the basal level of IL-10 in the blood was decreased. This contrasts to the increase in IL-10 caused by SMD given with LPS and suggests that the effects of SMD on induction of cytokine production by LPS do not reflect alterations in basal signaling functions by SMD.
SMD Affects IL-12 and IL-10 Production and Induces a Stress Response following Dermal Application
Although the po route is one mode of exposure for human beings to SMD, dermal exposure may be more common (Weiss and Lowit, 2004). Therefore, it was of interest to determine if the effects noted in Figure 1 following po administration were also observed after dermal administration. The results shown in Figure 2 indicate that LPS-induced IL-10 and IL-12 responses are affected similarly (a decrease in IL-12 and an increase in IL-10) for dermal as for po administration. The results in Figure 2 also demonstrate that IL-12 p70 is affected and that an alternative method (BioPlex Reader) yields similar results as ELISA, which was used in the other experiments. The results shown here for IL-12 p70 and in Figure 1 for IL-12 p40 indicate that the concentration of the p70 form is lower than the concentration of p40, and this is consistent with the findings of others (Heinzel et al., 1996). Although there is considerable evidence that the p40 chain of IL-12 can inhibit the development of acquired immunity, recent results suggest that the p40 form can initiate signaling through the IL-12 receptor and promote innate and acquired resistance to infection (Holscher et al., 2001; Russell et al., 2003). Thus, p40 as well as p70 is of interest with regard to immune function.
The results shown in Figure 2 also indicate that dermal application of SMD substantially increases the concentration of corticosterone in the serum. In a previous study it was determined that the stress response induced by po administration of SMD was sufficient to account for SMD-induced thymic atrophy (Myers et al., 2005). The results shown here indicate that the induction of a stress response by SMD is not unique to the po route. The role of the stress response in the effects of SMD on TLR signaling and cytokine production is not clear. However, we have also found that SMD directly inhibits cytokine production by peritoneal macrophages in vitro (see supplemental data online), suggesting that either direct or indirect (stress-mediated) or both mechanisms may have a role in the effects observed in this study.
The Effect of SMD on LPS-Induced IRAK-1 Degradation and the Role of MyD88
The results shown in Figure 3 indicate that LPS induces the degradation of IRAK-1 in peritoneal cells and that SMD does not affect this degradation. Phosphorylation of IRAK-1 is one of the early events in TLR4 signaling, and hyperphosphorylation leads to ubiquitination and degradation (Yamin and Miller, 1997). Thus, the absence of change in IRAK-1 degradation suggests that SMD does not act early in the TLR signaling pathway. The major adaptor molecule in TLR4 signaling is apparently MyD88, but other adapters can also bind to TLR4 and mediate signaling. The results shown in Figure 3 demonstrate that cytokine production is considerably reduced in MyD88 knockout mice, but the key effects of SMD still occur (decreased IL-12 and increased IL-10) in MyD88 knockout mice. Again, this suggests that the action of SMD is downstream of the early events of signaling and is not specific for MyD88. This is also suggested by the observation that SMD suppresses IL-12 and CXCL9 production and enhances IL-10 production induced by polyinosinic polycytidylic acid (poly I:C), which acts through TLR3 and does not utilize MyD88 (see supplemental data online).
The Effect of SMD on LPS-Induced MAP-kinase, AP-1, and NF-κB Activation
The results shown in Figure 4 demonstrate that SMD inhibits LPS-mediated activation (phosphorylation) of p38 and JNK. The level of phosphorylation of ERK is also decreased by SMD, but there was ERK activation in cells from untreated (naive) mice, such that LPS did not significantly increase ERK activation. It is unlikely that this reflects exposure of these cells to unknown stimuli before the experiment, because the other MAP kinases were only minimally active in cells from untreated mice. As expected in cells in which activation of p38 and JNK was inhibited, the translocation of AP-1 (as indicated by the presence of p-c-Jun) to the nucleus was significantly decreased by SMD. However, the translocation of NF-κB (p65 component) to the nucleus was not affected by SMD. The activation of TLR4 leads to the formation of a multi-component signalosome complex, which activates both MAP kinases and I-κB kinase (leading to NF-κB activation) (Li et al., 2002). The normal nuclear translocation of NF-κB in the presence of SMD suggests that the complex per se is functional but that activation of MAP kinases is relatively selectively affected.
SMD and MITC Alter Production of Cytokines in the Peritoneal Cavity but Do Not Alter the Percentages of Cell Types or Cell Number
The results shown in Figure 5 indicate that SMD and MITC at equimolar dosages (300 mg/kg for SMD and 45 mg/kg for MITC) decrease peritoneal concentrations of IL-12 similarly. As noted previously, SMD increased IL-10 concentrations, and MITC caused an even greater increase. As we and others have reported previously (Goral et al., 2004; Pruett et al., 2004a,b), 80–90% of cells in the peritoneal cavity are macrophages, and this was not altered by LPS or LPS + SMD. The results indicate that LPS decreases the total cell number in the peritoneal cavity, possibly by mobilizing these cells to other areas. Neither SMD nor MITC significantly change the number of cells in the peritoneal cavity in mice treated with LPS + SMD or LPS + MITC as compared to mice treated with LPS only.
SMD Increases IL-10 mRNA Expression and Decreases Expression of Other Cytokines in Peritoneal Cells
The results shown in Figure 6 indicate that SMD (administered 30 min before LPS) increases IL-10 mRNA expression in cells harvested 3 h after LPS administration. In contrast, the expression of mRNA for a variety of other cytokines, including the p35 and p40 subunits of IL-12, is suppressed by SMD in cells harvested 2 h after LPS administration.
SMD Causes a Similar Pattern of Cytokine Changes Using Standard or Purified LPS Preparations, and Similar Changes in IL-12 Are Noted Intracellularly and in Peritoneal Lavage Fluid
The results shown in Figure 7 indicate that SMD increases the concentration of IL-10 and decreases the concentration of IL-12 in the peritoneal fluid when using purified LPS with undetectable levels of lipoprotein. Lipoprotein can activate TLR2 (Vasselon et al., 2004). Thus, this result indicates that the effects mediated by standard LPS in this system are mostly due to stimulation of TLR4. The results shown in Figure 6 are also consistent with this conclusion. The results in Figure 7 also indicate that the intracellular concentration of IL-12 is decreased to a similar degree as IL-12 in peritoneal lavage fluid. This suggests that changes in IL-12 production occur in peritoneal cells (which are >80% macrophages), and it is consistent with the idea that most of the IL-12 in peritoneal lavage fluid is produced by macrophages and not by anchored cell types in the peritoneal cavity (e.g., fibroblasts).
SMD Decreases Resistance to Escherichia coli (E. coli) and Increases IL-10 Production
The results shown in Figure 8 demonstrate that SMD (at 200 mg/kg or 300 mg/kg) substantially decreases resistance to E. coli peritonitis within a time frame (24–48 h) that indicates suppressed innate immunity. Cytokine concentrations in the peritoneal cavity were measured in a separate set of mice treated identically to those in panel A of Figure 8. The results (also shown in Fig. 8) indicate that IL-10 is significantly increased by SMD 11 h after administration of E. coli. IL-12 is decreased, but the decrease is not quite significant.
The results shown here indicate that SMD inhibits TLR4 signaling by inhibiting the activation of MAP kinases and AP-1, that it alters cytokine mRNA expression, and that it causes corresponding changes in the concentrations of IL-10 and IL-12 in the peritoneal cavity and of IL-12 in peritoneal cells, which are mostly macrophages. Specific inhibitors of MAP kinases inhibit the production of pro-inflammatory cytokines induced by LPS (Branger et al., 2002). Thus, the results presented here suggest that the decreased activation of MAP kinases is responsible for the altered mRNA expression and cytokine protein expression. This is further suggested by the observation that AP-1 (or AP-2, which is activated by the same kinases) (Tuli et al., 2002) can be involved in the transcription of IL-1 (Jeon et al., 2000), IL-18 (Wang et al., 2002), IFN-γ (Chevillard et al., 2002), MIF (Mitchell et al., 1995), and IL-12 (Ma et al., 2004; Zhu et al., 2001). The basis for the increased expression of IL-10 is not clear, but reciprocal changes in IL-10 and IL-12 are caused by a variety of agents (Correa et al., 2005; Elenkov et al., 1996; Mason et al., 2000; Pruett et al., 2003a). This suggests that one of the transcription factors or co-factors involved in the expression of pro-inflammatory cytokines negatively regulates IL-10 expression. If SMD decreases activation of this transcription factor, it would be expected to increase IL-10 expression. In this regard, it is interesting that AP-1 has been reported to negatively regulate expression of some genes (Chen et al., 2002; Mauviel et al., 1996). Thus, it is possible that inhibition of AP-1 by SMD both decreases IL-12 expression and increases IL-10 expression. It also remains possible that SMD affects NF-κB signaling, even though it does not inhibit p65 translocation into the nucleus. In some cases, NF-κB signaling requires phosphorylation of p65 as well as its translocation to the nucleus (Schmeck et al., 2004). We did not evaluate NF-κB phosphorylation, so it remains possible that SMD inhibits NF-κB signaling by inhibiting phosphorylation. This will be investigated in ongoing studies.
The molecular mechanism by which SMD acts to inhibit signaling is not revealed by these results, but the chemical properties and physiological effects of SMD suggest some possibilities. SMD has a free S− group that allows it to function as a reducing agent and free radical scavenger (Motohashi and Mori, 1986; Zanocco et al., 1989). SMD spontaneously degrades in vitro and in vivo to form methylisothiocyanate (MITC), and MITC reacts with reduced glutathione (GSH) and causes it to be transported from cells (Thompson et al., 2002). SMD is an excellent chelator of cupric ions (Gray, 1964), and dithiocarbamates may either deplete copper from the animal or increase transport of copper into cells (Nobel et al., 1995; Shimada et al., 2005). Both cellular redox status and copper availability can have profound effects on cellular signaling (including MAP kinases and AP-1) (Chen et al., 2000; Chung et al., 2000). Additionally, we have reported that SMD induces a substantial neuroendocrine stress response leading to elevated levels of corticosterone sufficient to cause thymic hypoplasia (Myers et al., 2005), and increased corticosterone was also observed following dermal dosing in the present study. Any of these mechanisms could explain the effects of SMD, and experiments are in progress to evaluate them.
The observation that MITC causes similar effects as SMD on IL-10 and IL-12 expression suggests the possibility that MITC resulting from breakdown of SMD is mostly responsible for the effects of the SMD. Further studies should demonstrate if there are effects mediated by SMD that are not caused by MITC.
The effects of SMD on TLR signaling contrast with the effects of ethanol. Ethanol acts further upstream and significantly inhibits the degradation of IRAK-1. It also inhibits activation of NF-κB as well as MAP kinases and AP-1 (Pruett et al., 2004a). It is interesting in this regard that ethanol also increases IL-10 expression and decreases IL-12 expression (Pruett et al., 2003a). This suggests that these effects depend more on suppression of AP-1 activation than on suppression of NF-κB activation (which does not occur in SMD treated mice). In addition, the observation that ethanol and SMD have different effects on signaling indicates that the action of SMD on signaling is not primarily mediated by the stress response it induces. Ethanol produces a similar stress response (Pruett et al., 2003c) as SMD (Myers et al., in press), and if stress mediators (such as corticosterone) were responsible for the alterations in TLR4 signaling, it would be expected that similar changes in signaling would be caused by ethanol and SMD.
The results presented here do not conclusively establish that the altered cytokine responses caused by SMD are responsible for the decreased resistance to E. coli. It has been established that many of the genes regulated by E. coli treatment of mice are similarly regulated by LPS (Huang et al., 2001; Nau et al., 2003). Considering that increased IL-10 has been reported to decrease E. coli clearance (Takano et al., 1998) and that some of the cytokines suppressed by SMD are involved in resistance to E. coli or similar bacteria (Cross et al., 1995; Kinoshita et al., 2004; Zisman et al., 1997), it seems likely that these changes do contribute to decreased resistance. The mouse model used here is relevant to a variety of conditions in humans (e.g., abdominal trauma, appendicitis, diverticulitis, chronic alcoholism) in which large numbers of normally non-pathogenic bacteria enter the peritoneal cavity from the gastrointestinal tract (Pruett et al., 2004b). It remains to be determined whether death is caused by overgrowth of bacteria leading to intravascular coagulation and multiple organ failure or whether there is a rebound in the production of pro-inflammatory cytokines leading to systemic inflammatory response syndrome, shock, and death. A rebound effect leading to increased concentrations of MIF, for example, after the effects of SMD subside, would be expected to increase lethality due to systemic inflammatory response syndrome (Roger et al., 2001).
The results presented here are relevant with regard to environmental and occupational exposure of persons to SMD and MITC. The dosages of SMD used here could occur as a result of occupational exposure to commercial SMD preparations (Padgett et al., 1992). In addition the reported dosage of MITC for a child near fields treated with SMD is up to 1.0–2.5 mg/kg/day (Thongsinthusak, 2000). Expressing this dosage range on the basis of body surface area instead of body weight in humans and mice indicates that this dosage range in humans encompasses a dosage of 17 mg/kg in mice (calculated as described in our previous study) (Padgett et al., 1992). This dosage was sufficient to significantly decrease serum IL-12 expression in the present study. Exacerbation of asthma has been reported following exposure to SMD or MITC, and this would be consistent with increased IL-10 and decreased IL-12 leading to a shift toward a Th2 response.
Supplementary data are available online at www.toxsci.oxfordjournals.org.
This work was supported by grant R01 ES09158 from the National Institute for Environmental Health Sciences. This paper is dedicated to Dr. Jerry Allen on the occasion of his retirement. Dr. Allen's dedication and skill as a teacher inspired one of us (S.B.P.) to pursue a career in immunology. Conflict of interest: none declared.