Published online before print July 6, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Roswell Park Cancer Institute, Department of Immunology, Buffalo, New York; and
University of Connecticut School of Medicine, Farmington
2Correspondence: Department of Immunology, CCC 401, Elm and Carlton Streets, Buffalo, NY 14263. E-mail: elizabeth.repasky{at}roswellpark.org
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and lipopolysaccharide (LPS) were needed to induce NO at 37°C, we observed that addition of only LPS was sufficient for production of NO if there were a pretreatment at 39.5°C. Further, if IFN- and LPS were given after thermal exposure, a substantial increase in NO and iNOS was observed over that seen using cells kept at normothermic conditions. Macrophages isolated from mice lacking heat shock factor-1 did not attenuate the ability of mild thermal stress to modulate NO production. Reverse transcriptase-polymerase chain reaction data revealed that thermal regulation of iNOS expression is not entirely at the transcriptional level, suggesting possible points of post-transcriptional thermal sensitivity. These data support the concept that altering the thermal microenvironment is an important means by which the host can manipulate macrophage responses. Increases in temperature (e.g., during fever) may function to lower the activation threshold needed for production of effector molecules in times of infection.
Key Words: heat shock proteins • stress response • hyperthermia
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tissue macrophages are major producers of NO by the action of the inducible form of NOS (iNOS) [17
], which was first defined in macrophages [18
]. iNOS is induced by cytokines, such as interferon- (IFN)- and tumor necrosis factor 
(TNF-), and by bacterial products such as lipopolysaccharide (LPS), a component of gram-negative bacterial cell walls. As activated macrophages produce the proinflammatory cytokines that induce local inflammation and systemic increases in temperature [19
, 20
], NO production often occurs in environments of elevated temperature. Is macrophage production of NO regulated or enhanced by mild (physiologically relevant) changes in environmental temperature?
To obtain some information about this unexplored question, we decided to examine whether a mild, physiologically relevant temperature increase can affect NO production in primary macrophages isolated from BALB/c and C57BL/6 mice. In addition, we looked at iNOS protein and mRNA levels, as well as for the involvement of heat shock factor-1 (HSF-1), in regulation of NO production by macrophages exposed to mild thermal stress. Overall, our results suggest that small increases in temperature can significantly enhance the production of this critical effector molecule; however, this enhancement was found to be highly regulated and is not simply a result of a nonspecific increase in general cellular metabolism. This highly selective thermal response could provide a mechanism by which immune effector cells could distinguish between a response appropriate for resolution of disease states, in which there is thermal stress associated with fever or inflammation, and that associated with exercise or sauna, in which there may be a nondisease-related hyperthermia.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Preparation of murine peritoneal macrophages
Three days prior to harvest of peritoneal cells, mice were treated with 2.0 mL sterile, 3% solution of Brewer’s thioglycollate, intraperitoneally. Cells were harvested from killed mice using sterile harvest medium [0.1% bovine serum albumin, 10 U/mL heparin, 0.54 mM EDTA, in phosphate-buffered saline (PBS)]. Once all peritoneal cells were collected, the cell suspension was spun at 1200 rpm (350 g) for 10 min, resuspended in RPMI-10 (RPMI-1640 medium with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, and 55 mM ß-mercaptoethanol), and then plated at a concentration of 3 x 106 cells/mL. For some experiments, 100 µL of this cell suspension was plated into sterile tissue culture-treated, 96-well plates, and for other experiments, 3.0 mL of this cell suspension was plated into 60 mm sterile tissue culture-treated dishes. Macrophages were enriched from the total peritoneal exudates by 1 h adherence to plastic. After washing away nonadherent cells, warmed RPMI-10 was added to each well or plate to a total volume of 200 µL (96-well plates) or 5.0 mL (60 mm dishes).
Exposure of macrophages to mild-temperature increase, stimulation with IFN- and LPS, and determination of nitrite (NO2–) concentration
Macrophages were incubated at 37°C or 39.5°C for 7 h in separate, humidified incubators with 5% CO2. After 7 h of incubation, warmed RPMI-10 containing IFN- (1 U/mL or 5 U/mL) and/or LPS (2.5 ng/mL or 50 ng/mL) was added to appropriate wells or plates. Each experimental condition was performed in at least triplicate. To a separate set of wells, the NOS inhibitor NG-monomethyl-L-arginine (L-NMA) was added in the presence and absence of IFN- and/or LPS at a concentration of 250 µM. Cells were then further incubated at 37°C for 10 h or 16 h. NO2– concentration in the cell culture supernatants was used as a surrogate marker of NO biosynthesis and was measured using the Griess assay. Absorbances of the standards and samples were read at 550 nM using a MRX plate reader (Dynex Technologies, Chantilly, VA), and results were calculated from the standard curve using Biolinx software (Version 2.2, Dynex Technologies).
Immunoblot analysis of the iNOS
The medium from 60 mm tissue-culture dishes originally inoculated with 9 x 106 total peritoneal cells was removed, and the cell monolayer was washed with 5 mL sterile PBS, twice. The cells were then lysed in 300 µL boiling sodium dodecyl sulfate (SDS) cell lysis buffer (1.0% SDS, 1.0 mM sodium orthovanadate, and 10 mM Tris, pH 7.4), scraped into microfuge tubes, and boiled in a microwave for an additional 5 min, spun at 12,000 rpm (13,200 g) for 10 min, after which the supernatant was removed and stored at –70°C for future analysis. The Micro BCATM protein assay reagent kit (Pierce, Rockford, IL) was used to determine protein concentration using dilutions that prevented the interference of SDS present in the samples, which for SDS-polyacrylamide gel electrophoresis (PAGE)/immunoblotting was made with concentrations between 500 µg/mL and 1000 µg/mL, depending on the experiment. Briefly, the samples were diluted in 4x SDS sample buffer (8% SDS, 400 mM dithiothreitol, 200 mM Tris, pH 6.8, 40% glycerol, and a pinch of bromophenol blue), boiled for 5 min, and then stored at –20°C until use. Samples were resolved in a 10% SDS-PAGE (12.5–25.0 µg/lane) and transferred to methanol-treated, Immobilon P polyvinylidene difluoride membranes (Millipore, Bedford, MA) for 1 h at 380 mA. The membranes were then cut into two pieces, horizontally, so that iNOS and ß-actin antibodies (Ab) were kept separate while probing. Nonspecific binding of Ab to the membrane was minimized by blocking with PBS containing 5% nonfat dry milk and 0.05% Tween (NFDM/PBST) for 30 min at room temperature with agitation and then probed for iNOS or ß-actin (as a loading control). The mouse anti-mouse iNOS monoclonal Ab [mAb; immunoglobulin G2a (IgG2a), clone 6, BD Transduction Laboratories, Lexington, KY] was used at a dilution of 1:2000 in 5% NFDM/PBST, and the mouse anti-mouse ß-actin mAb (IgG1, clone AC15, Sigma Chemical Co., St. Louis, MO) was used at a dilution of 1:20,000 in 5% NFDM/PBST. Membranes were incubated with the primary Ab for 1 h at room temperature with agitation. Membranes were then washed four times, 10 min each, in large volumes of PBST with agitation. The horseradish peroxidase-conjugated goat anti-mouse secondary Ab (BD Transduction Laboratories) was used at a dilution of 1:2500 for the iNOS-probed membranes and at a dilution of 1:100,000 for the ß-actin-probed membranes in 5% NFDM/PBST for 1 h at room temperature with agitation. The membranes were washed six times, 10 min each, in large volumes of PBST and then incubated in Supersignal West Pico chemiluminescent substrate (Pierce) for 5 min prior to detection of iNOS and ß-actin.
Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of iNOS message steady-state levels
Four hours after addition of IFN- and LPS, total RNA was harvested from peritoneal macrophages using the RNeasy Mini Kit (Qiagen, Valencia, CA) per the manufacturer’s instructions. Superscript II RNaseH– RT and oligo-dT12–18 primers (Invitrogen, Life Technologies, Carlsbad, CA) were used to synthesize cDNA from 1 µg total macrophage RNA. Mus musculus-specific iNOS cDNAs were amplified by PCR using Platinum Taq polymerase (Invitrogen, Life Technologies) with the following primer sequences: iNOS forward primer, 5'CCA ACC GGA GAA GGG GAC GAA CT3'; iNOS reverse primer, 5'GGA GGG TGG TGC GGC TGG AC3', producing a PCR product of 295 base pairs (bp) in length. Thermal cycle conditions for the PCR were as follows: 1 cycle at 94°C for 2 min, 94°C for 45 s, 55°C for 45 s, and 72°C for 75 s for 30 cycles, followed by a final extension at 72°C for 7 min. Analysis of Mus musculus glyceraldehyde 3-phosphate dehydrogenase (GAPDH; forward primer, 5'CAT GTA GGC CAT GAG GTC CAC CAC3'; reverse primer, 5'TGA AGG TCG GTG TGA ACG GAT TTG GC3', 983 bp product amplified) was used as a control in these experiments. After visualization of PCR products in a 0.6-µg/mL ethidium bromide-containing 1% agarose gel, integrated density volumes (IDV) were calculated for each PCR product using the Alpha Imager spot density analysis software. Ratios of iNOS to GAPDH were then calculated from these values and plotted as bar graphs.
Statistics
The unpaired Student’s t-test was used to compare differences between control and experimental data points. P values of <0.05 were considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, or both and then determined the concentration of NO2– in cell culture supernatants 10 h later. NO2– measurement is accomplished using the Griess reaction and is recognized as an acceptable surrogate marker of NO biosynthesis [21
]. Normally, IFN- and LPS are needed to induce NO production from macrophages, indicating the need for relevant danger signals to be present before this highly potent effector molecule is released. We used two different doses of these molecules to help us evaluate the extent of effects of hyperthermia on NO production; moreover, we decided to include the evaluation of two mouse strains (C57BL/6 and BALB/c), which are known to differ in terms of their responsiveness to the same doses of LPS and/or IFN- [22
].
In untreated cultures of macrophages from both strains of mice, NO production was seen neither at normothermic conditions (37°C) nor at mild hyperthermic conditions (39.5°C; Fig. 1
). This result confirms the fact that NO production is tightly controlled and does not normally occur in untreated cells nor can it be induced by simply raising the temperature of untreated cells, even over a period of 7 h. However, in heated macrophage cultures derived from BALB/c mice, in which a low dose of IFN- + LPS was added following thermal exposure, we saw an eightfold induction of NO production over that seen in normothermic cultures to which the same doses of IFN- and LPS were added (Fig. 1)
. When a higher dose of IFN- + LPS is used, NO production is still enhanced greater than twofold over that seen in normothermic controls (Fig. 1)
. More importantly, we saw a statistically significant increase in NO production from heated cells when only LPS was added (especially for the cultures treated with the higher dose of LPS). For BALB/c macrophages under normothermic conditions, we found that LPS and IFN- are needed for generation of NO and that neither treatment used alone, even at the high dose, resulted in NO production. This suggests that a mild, physiologically achievable increase in BALB/c macrophage temperature can lower the criteria needed for NO production so that only LPS is needed, rather than LPS and IFN-.
 View larger version (19K): [in a new window] |
Figure 1. NO production is thermally regulated. Thioglycollate-elicited, peritoneal macrophages were incubated at 37°C or 39.5°C for 7 h [control (cont)] and then stimulated with IFN-, LPS, or IFN- + LPS and returned to 37°C. NO production was determined by measuring the concentration of NO2– in cell culture supernatants, using the Griess assay, 10 h after the addition of IFN- and LPS. Low dose = 1 U/mL IFN-, 2.5 ng/mL LPS; high dose = 5 U/mL IFN-, 50 ng/mL LPS. Data are representative of at least 10 separate experiments. Triplicate-to-quadruplicate measurements were made for each experimental condition. The peritoneal cells from three mice were pooled for each experiment. *, P < 0.01.
|
or LPS, and their macrophages were seen to produce more NO under normothermic conditions with IFN- or LPS (at the higher dose level) than do cells from BALB/c mice. However, the C57BL/6 macrophages responded similarly to BALB/c cells when exposed to hyperthermic conditions. C57BL/6-derived macrophages, incubated at 39.5°C prior to treatment with low-dose LPS, produced tenfold more NO than controls by 10 h after the addition of LPS (Fig. 1)
. When heating was used in conjunction with high-dose IFN-, a high-dose LPS, or with a low-dose IFN- + LPS, we observed a four-, two-, or threefold enhancement of NO production compared with the NO production of cultures incubated at 37°C. Moreover, C57BL/6-derived macrophages, when given thermal stimuli, not only responded by making NO when LPS was used alone but also produced significant levels of NO when IFN- was used alone (at the high-dose level). However, in contrast to the data obtained using BALB/c cells, the resultant, large amount of NO production achieved when high dose IFN- + LPS was used in combination on C57BL/6 macrophages at normothermic conditions was inhibited by the presence of hyperthermia (Fig. 1)
. L-NMA, an inhibitor of NO production by all NOS, was able to block the synthesis of NO by all cells used in these studies when used at a dose of 250 µM (data not shown).
Using both strains of mice, kinetic experiments were performed to assess the concentration of NO2– in cell culture supernatants at 15, 30, and 45 min and 1, 2, 4, 8, 12, 16, 20, and 24 h after IFN- and LPS addition. These studies revealed that changes in the concentration of NO2– observed in the above studies were maintained throughout the time course. In addition, in BALB/c and C57BL/6 mice, total protein determination was used to calculate ratios of µM NO2– measured to µg total protein per well (from 3x105 total peritoneal cells originally plated) and confirmed that insignificant amounts of cell death occurred, which could have been responsible for these differences in NO production (data not shown).
Mild-temperature increase regulates the synthesis of iNOS by murine macrophages
To begin to elucidate the molecular events by which a mild thermal stimulus regulates NO biosynthesis in this system, we examined the level of iNOS protein from macrophage lysates at the same time and from the same cultures from which the NO2– determination was made. Using BALB/c-derived macrophages, we found that the level of iNOS protein lysates mirrored the level of NO2– measured in the cell culture supernatants at that same time-point (i.e., if the NO2– were induced or enhanced, this result was recapitulated in an induced or enhanced level of iNOS found in our immunoblots; compare Figs. 1
and 2
).
 View larger version (44K): [in a new window] |
Figure 2. iNOS steady-state protein levels are altered by mild, fever-range hyperthermia. Thioglycollate-elicited peritoneal macrophages were incubated at 37°C or 39.5°C for 7 h and then stimulated with IFN-, LPS, or IFN- + LPS and returned to 37°C. Low dose = 1 U/mL IFN-, 2.5 ng/mL LPS; high dose = 5 U/mL IFN-, 50 ng/mL LPS. Macrophage lysates were prepared 10 h after addition of IFN- and LPS and used in an immunoblot to detect iNOS protein. Blots are representative of three separate experiments. The peritoneal cells from three mice were pooled for each experiment. ß-actin was used as a loading control.
|
and/or LPS recapitulated the NO2– results under the same experimental conditions (compare Figs. 1
and 2 ). Where NO2– was induced/enhanced, iNOS was induced/enhanced in macrophage lysates from those same cultures. It is interesting that we were able to detect the induction of iNOS protein in macrophages treated with low-dose IFN- and mild hyperthermia in a faint band in the immunoblot but were unable to detect a similar band from normothermic macrophages treated with low-dose IFN- (Fig. 2)
. The presence of this band precedes the measurable accumulation of NO2– in supernatants of those same cultures. When we treated macrophages with a high dose of LPS after incubation at 39.5°C, we observed a strong iNOS band in our immunoblot. Although NO2– was detectable in the normothermic macrophage cultures from C57BL/6 mice treated with high-dose LPS, iNOS protein was below the level of detection by immunoblot, even after a 60-min exposure time was used. This apparent dichotomy may reflect an alteration in the kinetics of iNOS protein synthesis and degradation in this mouse strain between control (normothermic) and heated macrophage cultures. However, this hypothesis remains to be tested. Finally, when macrophages were incubated with high-dose IFN- and LPS after exposure to thermal stimuli, iNOS was undetectable after a 5-min exposure (Fig. 2)
but was detectable after a lengthy (60 min) exposure (data not shown), once again correlating with NO2– data from the same cultures.
Lack of HSF-1 does not diminish the ability of a mild-temperature increase to enhance NO production by murine macrophages but is not tightly correlated with iNOS protein levels
As gene transcription can be regulated by the transcription factor HSF-1 during times of cellular stress {.e., during nonphysiological, 42–45°C heat shocks (for review, see ref. [23
])}, and as a consensus sequence or heat-shock element (HSE) is present in the promoter/enhancer of the mouse iNOS gene to which HSF-1 has been shown to bind during heat shock [24
], we examined whether HSF-1 played a role in the NO production induced by mild thermal conditions used here. To accomplish this, we incubated elicited peritoneal macrophages from hsf–/– (knockouts) and hsf+/– (heterozygote controls) at 39.5°C for 7 h, after which we stimulated those cells with IFN-, LPS, or a combination of IFN- and LPS and placed all cultures at 37°C. Sixteen hours later, we measured the concentration of NO2– in supernatants using the Griess assay and detected the amount of iNOS protein in macrophage cell lysates, prepared concomitantly, by immunoblotting. It is interesting that the lack of HSF-1 in these cells did not preclude their ability to produce a greater amount of NO after incubation at febrile temperatures when compared with controls maintained at 37°C (Fig. 3
). However, as opposed to wild-type cells from BALB/c mice, these knockout and heterozygote cells did not respond to LPS but instead responded more robustly to IFN-. The difference in IFN- and LPS responsiveness between wild-type hsf-1+/– and hsf-1–/– macrophages is not known but is currently under investigation. Although the difference in the iNOS protein steady-state levels between heated and normothermic control macrophages from heterozygote animals reflected the changes we observed in NO2–, measured in the cell culture supernatant, the difference in iNOS protein level between heated and control cells from hsf-1–/– macrophages was diminished (Fig. 4
). These data correlated with the presence of heat-shock protein (HSP)70, or relative lack thereof, in the heterozygote or homozygote knockout mouse strain, respectively (Fig. 5
). Specifically, the lack of HSP70 correlated with a diminished difference in iNOS level between heated and control macrophages from the hsf-1 knockout mice, and the presence of HSP70 in the heterozygote animals correlated with a notable difference in iNOS level in those same macrophages.
 View larger version (18K): [in a new window] |
Figure 3. Lack of HSF-1 does not diminish the ability of mild hyperthermia to enhance NO production. Thioglycollate-elicited, peritoneal macrophages from two hsf-1–/– and two hsf-1+/– mice were incubated at 37°C or 39.5°C for 7 h [control (cont)] and then stimulated with high-dose IFN-, LPS, or IFN- and LPS and returned to 37°C. Sixteen hours later, NO production was determined by measuring the concentration of NO2– in cell culture supernatants using the Griess assay. This experiment was only performed once with two mice whose cells were not pooled. Triplicate measurements were made for each experimental condition. *, P < 0.01.
|
 View larger version (29K): [in a new window] |
Figure 4. HSF-1 does not appear to be required for synthesis of the iNOS enzyme. Thioglycollate-elicited peritoneal macrophages from two hsf-1 null (–/–) and two hsf-1 heterozygote (+/–) mice were incubated at 37°C or 39.5°C for 7 h [control (Cont)] before stimulation with high-dose IFN-, LPS, or IFN- and LPS. Sixteen hours later, cell lysates were made and prepared for immunoblot detection of iNOS. This blot (one mouse) is representative of two separate blots.
|
 View larger version (30K): [in a new window] |
Figure 5. HSF-1 is required for induction of HSP70 by mild hyperthermia. Thioglycollate-elicited peritoneal macrophages from two hsf-1 null (–/–) and two hsf-1 heterozygote (+/–) mice were incubated at 37°C or 39.5°C for 7 h [control (Cont)] before stimulation with IFN- and/or LPS. Sixteen hours later, cell lysates were made and prepared for immunoblot detection of HSP70. This blot (one mouse) is representative of two separate blots.
|
and/or LPS. After 4 h of incubation at 37°C, total RNA was isolated for analysis of iNOS mRNA steady-state levels using RT-PCR. When compared with the level of GAPDH in each sample, little difference in iNOS steady-state message level could be demonstrated between heated and control macrophages. These findings were in contrast to the rather pronounced differences in iNOS protein level that we had previously observed, suggesting that mild thermal stress may predominantly regulate this gene at post-transcriptional levels.
 View larger version (40K): [in a new window] |
Figure 6. Mild hyperthermia minimally affects iNOS message steady-state levels. Thioglycollate-elicited, peritoneal macrophages from BALB/c and C57BL/6 mice were incubated at 37°C or 39.5°C for 7 h [control (cont)] and then incubated with low- or high-dose IFN-, LPS, or IFN- and LPS. Four hours later, total RNA was harvested from all cells. RT-PCR was performed to determine the relative level of iNOS transcript to that of GAPDH in each mouse strain. ImageQuant was used to calculate IDV for the intensity of each PCR product in the agarose gels. Bar graphs illustrate the ratio of iNOS to GAPDH for each experimental condition. The background level of iNOS expression is depicted by the horizontal line drawn though each graph. Control reactions (lacking RT) confirmed purity of the RNA samples (data not shown).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Additional levels of complexity in the thermal regulation of NO production were revealed in macrophages from C57BL/6 mice. When elicited peritoneal macrophages from these mice were treated with high-dose IFN- in combination with LPS, conditions that result in high levels of NO production, febrile temperatures inhibited NO and iNOS production. As a high level of NO is known to negatively regulate the continued production of NO [17
], it is likely that incubation of macrophages at 39.5°C over-stimulated the macrophages and thus, inhibited NO biosynthesis. It is interesting that we did not observe negative-feedback regulation of NO or iNOS in our experiments using macrophages from BALB/c mice. This observation supports other studies, which demonstrate that BALB/c mice are relatively hyporesponsive to various stimuli, whereas the C57BL/6 mice are more hyper-responsive to the same stimuli [22
].
Collectively, these data support the notion that thermally sensitive, regulatory pathways exist in macrophages, which are highly sensitive to existent regulatory or feedback mechanisms. Thermal stimulation in the absence of an appropriate additional signal, i.e., LPS or IFN-, does not result in a global stimulation of effector molecule production. This suggests that a second, relevant signal (e.g., bacterial-like molecule or specific cytokines) is needed for the cell to respond to thermal elevation and, in turn, additional mechanisms are in place to prevent the potentially harmful over-production of NO. These data are supported by earlier findings from our group that the acute-phase response proteins [including interleukin (IL)-1, IL-6, and TNF-] are not increased by thermal stress alone but are significantly enhanced in serum following addition of LPS (presumably mimicking bacterial presence) in the presence of thermal stress [35
]. Further, these data also support our earlier observations that fever-like thermal stress can also enhance the maturation and function of another antigen-presenting cell, epidermal dendritic cells [36
]. Although at the initial moment of infection in a physiological setting, local macrophages have not been previously exposed to a sustained thermal stress (such as that experienced in the in vitro protocol examined here), we speculate that this protocol is highly relevant to the sequence of events involving more distantly recruited macrophages in the following manner: Immune cells, such as macrophages involved in the initial encounter with antigen, help to generate a systemic or regional hyperthermia in the form of fever or inflammation by their production of proinflammatory cytokines. Thus, distant macrophages may now be exposed to increased temperature in their microenvironment, prior to their being recruited to the site and brought into contact with bacteria (in our case, simulated by LPS). These "heat-sensitized" macrophages may now be in a position to respond to antigen by having a lowered threshold of activation for the production of important effector molecules such as NO. The fact that small increases in temperature can significantly enhance the production of critical effector molecules, only if IFN- or LPS are provided, suggests that this selective regulation could be important to reduce the possibility of inappropriate responses to temporary hyperthermia caused by other nonpathogenic events (e.g., exercise or sauna). As the current report exclusively examines the effects of mild thermal stress on primary macrophages, in vitro, concomitant with multiple sources of stimulation, this hypothesis must next be evaluated in vivo with conditions that result in the production of NO or other effector molecules from macrophages.
It is important to note that the majority of the data presented here is in contrast to previously published reports that also examined the effect of elevated temperatures on NO production on cells in which a decrease in NO production was observed [37 38 39 40 41 ]. However, there are potentially important differences between the current study and those done previously. One important difference may be that the models used in the previous studies did not use cells of the immune system. These earlier studies demonstrated that exposure of rat and human hepatocytes [37 , 38 ], rat pulmonary artery smooth muscle cells [39 ], and rat astrocytes [40 , 41 ] to heat-shock conditions decreases cytokine-induced NO and iNOS production. It is intriguing to speculate that macrophages may have evolved to respond to temperature shifts as part of their selective response to the host, as they generate fever in the presence of bacterial or viral infections. However, another important difference between the earlier studies and those presented here is that temperatures used in the previous studies are significantly higher (i.e., between 41°C and 43°C) than those used here or those usually attained during febrile episodes [28 ], in which the maximum is usually 40–41°C. It is reasonable to assume that there could be important differences in this response depending on the temperature to which the cells were exposed and that there may be a narrow window of physiological responses to temperature gradients. A more thorough examination of how temperature affects NO/iNOS production in various cell types, including the macrophage, would illuminate just how much heat is required to obtain the enhanced NO/iNOS that we have already observed and at what temperature, if any, enhancement of NO/iNOS ceases.
It is interesting that Goldring et al. [24] reported that the promoter of the iNOS gene contains a region that is hypomethylated at guanine –898/9 when RAW 264.7 cells were heated at 41°C for 20 min. The hypomethylation, observed as a result of heat treatment, directly mimicked hypomethylation at that same site when cells were treated with LPS. These authors characterized this site as a potential HSE, representing one of several putative HSEs in the iNOS promoter. Although the function of these elements in iNOS gene regulation has not been characterized completely, this particular study showed that mutations of the HSEs, which abrogated HSF-1 binding to this element, led to altered response of the iNOS gene to high-temperature heat regulation. Because of the presence of HSEs in the mouse iNOS promoter and the notion that HSF-1 may play a role in iNOS gene regulation in high-temperature heat protocols, we examined the ability of macrophages from hsf-1 knockout mice to produce NO after incubation at 39.5°C and treatment with IFN- and/or LPS. Our initial work suggests that HSF-1 is not playing a major role in the stress-induced/enhanced production of NO by murine macrophages. However, HSF-1 may impact macrophage sensitivity to different stimuli as well as affect iNOS steady-state levels after synthesis. The latter idea correlates with the inability of hsf-1 knockouts to produce HSP70. These data suggest that heat can enhance NO production in a way distinct from a direct effect of HSF-1 activity, as was previously demonstrated by Goldring et al. [24]. It is possible that the difference in iNOS protein steady-state levels that we noted between control and heated macrophages after treatment with IFN- and/or LPS is a result of, at least in part, HSP production and/or to post-transcriptional regulatory mechanisms activated by mild thermal stress within the cell involving, but not limited to, effects on mRNA stability and translational efficiency.
It is well recognized that small changes in RNA half-life can dramatically alter the amount of protein ultimately expressed [42 ] and that many different cellular stresses can increase mRNA stability [43 ]. As the febrile temperatures we used here are able to induce a stress response in murine macrophages, as judged by the production of HSP70 demonstrated above, we hypothesize that iNOS mRNA stability may be enhanced in cells exposed to mild thermal stress. iNOS mRNA half-life is controlled, in part, by adenine and uracil-rich elements (AREs) found in its 3'-untranslated region [44 45 46 ], to which RNA binding proteins such as ARE-binding factor 1 and human-antigen R bind and control mRNA stability [47 48 49 ]. HSP70 and HSP110 also bind to AREs [50 , 51 ]. These data suggest that HSPs may play a role in mRNA metabolism as "RNA chaperones" separate from their well-known function as protein chaperones. We have previously demonstrated that mild thermal stress induces HSP70 and HSP110 in certain mouse tissues [52 ] and HSP70 in lymphocytes [53 ]. This report demonstrates that HSP70 is induced in macrophages as well after mild thermal stress. The expression of HSPs under these circumstances correlates with their ability to function as RNA and protein chaperones during mild thermal stress. It should also be noted here that earlier reports indicate that exogenously added HSP70, HSP60, and gp96 result in the induction of iNOS and in the production of NO [54 55 ]. How these data relate to the role of thermal induction of intracellular HSPs is not clear but does indicate the potential for other levels of HSP-dependent regulation of this system. Finally, although other kinases play important roles in controlling iNOS synthesis, one recent publication [56 ] has identified a pivotal role for the c-Jun NH2-terminal kinase in stabilizing the iNOS mRNA and further, may do so during conditions of cellular stress. Analysis of changes in iNOS mRNA stability in macrophages exposed to mild thermal stress involving each of the mechanisms described above is currently underway.
In summary, these data support the hypothesis that macrophages may be highly sensitive to physiologically relevant thermal gradients and that conditions such as fever or inflammation can lower the threshold of activation required for NO production. Mild hyperthermia may thus serve as a danger signal similar to a role recently described for HSP [57 ] (see also refs. [58 59 60 61 ]). It is possible that febrile temperatures may represent those at which macrophages have evolved to function optimally during infection. These data lay the foundation for continued exploration of the exact mechanism(s) by which macrophages are able to sense and respond to differences in their microenvironmental temperature and further suggest the presence of cellular signaling pathways or gene regulatory mechanisms that are activated by mild thermal stimuli.
Received April 2, 2004; revised May 16, 2005; accepted May 18, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and lipopolysaccharide J. Biol. Chem. 268,1908-1913
B activation J. Biol. Chem. 271,17724-17732B activation, nitric oxide synthase type 2 expression, and macrophage/microglial activation in brain J. Cereb. Blood Flow Metab. 20,800-811[Medline] mRNA Mol. Cell. Biol. 19,4311-4323 B pathway Int. Immunol. 12,1539-1546This article has been cited by other articles:
 |
|
 |
|
  Y. Y. Tyurina, L. V. Basova, N. V. Konduru, V. A. Tyurin, A. I. Potapovich, P. Cai, H. Bayir, D. Stoyanovsky, B. R. Pitt, A. A. Shvedova, et al. Nitrosative Stress Inhibits the Aminophospholipid Translocase Resulting in Phosphatidylserine Externalization and Macrophage Engulfment: IMPLICATIONS FOR THE RESOLUTION OF INFLAMMATION J. Biol. Chem., March 16, 2007; 282(11): 8498 - 8509. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Gobeil Jr., T. Zhu, S. Brault, A. Geha, A. Vazquez-Tello, A. Fortier, D. Barbaz, D. Checchin, X. Hou, M. Nader, et al. Nitric Oxide Signaling via Nuclearized Endothelial Nitric-oxide Synthase Modulates Expression of the Immediate Early Genes iNOS and mPGES-1 J. Biol. Chem., June 9, 2006; 281(23): 16058 - 16067. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
