Skip to main content
NCBI home page
Cell Stress & Chaperones logoLink to Cell Stress & Chaperones
. 2001 Oct;6(4):306–315. doi: 10.1379/1466-1268(2001)006<0306:ctsvit>2.0.co;2

Cell type–specific variations in the induction of hsp70 in human leukocytes by feverlike whole body hyperthermia

Rudolf Oehler 1,1, Erich Pusch 1, Maria Zellner 1, Peter Dungel 1, Nicole Hergovics 2, Monika Homoncik 2, Maja Munk Eliasen 1, Marianne Brabec 1, Erich Roth 1
PMCID: PMC434413  PMID: 11795467

Abstract

Fever has been associated with shortened duration and improved survival in infectious disease. The mechanism of this beneficial response is still poorly understood. The heat-inducible 70-kDa heat shock protein (Hsp70) has been associated with protection of leukocytes against the cytotoxicity of inflammatory mediators and with improved survival of severe infections. This study characterizes the induction of Hsp70 by feverlike temperatures in human leukocytes in vitro and in vivo. Using flow cytometry, Hsp70 expression was determined in whole blood samples. This approach eliminated cell isolation procedures that would greatly affect the results. Heat treatment of whole blood in vitro for 2 hours at different temperatures revealed that Hsp70 expression depends on temperature and cell type; up to 41°C, Hsp70 increased only slightly in lymphocytes and polymorphonuclear leukocytes. However, in monocytes a strong induction was already seen at 39°C, and Hsp70 levels at 41°C were 10-fold higher than in the 37°C control. To be as close as possible to the physiological situation during fever, we immersed healthy volunteers in a hot water bath, inducing whole body hyperthermia (39°C), and measured leukocyte Hsp70 expression. Hsp70 was induced in all leukocytes with comparable but less pronounced cell type–specific variations as observed in vitro. Thus, a systemic increase of body temperature as triggered by fever stimulates Hsp70 expression in peripheral leukocytes, especially in monocytes. This fever-induced Hsp70 expression may protect monocytes when confronted with cytotoxic inflammatory mediators, thereby improving the course of the disease.

INTRODUCTION

For centuries fever has been considered a protective body response, and physicians induced fevers to combat infections (Coxe 1846). When antipyretic drugs were introduced, treatments focused on the reduction of fever, and induction of fever was virtually abandoned (Kramer and Campbell 1993). However, recent animal studies describe protective effects of hyperthermia during infection and thus the concept of fever as a host defense response has again attracted the attention of physicians and scientists (Kluger et al 1998). Fever is a systemic response to regional infections. Invading microorganisms stimulate tissue macrophages to produce high amounts of interleukin 1 (Kluger 1991). Interleukin 1 reaches the brain and activates the thermoregulatory center, which then stimulates fat catabolism in the brown adipose tissue for heat production. This heat production finally leads to an increase of body temperature. However, little is known about the effects of this systemic hyperthermia on leukocytes distant from the primary site of infection.

It is well known that in response to strongly elevated temperatures cells express a distinct group of proteins, the heat shock proteins (Hsps). This heat shock response provides cells with increased resistance against a variety of environmental insults such as heat (Lindquist and Craig 1988), UV irradiation (Tyrell 1996), and, as we could recently show, also osmotic challenge (Oehler et al 1998). In addition, Hsps exert beneficial effects in inflammation, including protection against the cytotoxicity of inflammatory mediators (Polla and Cossarizza 1996). High expression of Hsps was also connected with increased survival in animal models for inflammation such as sepsis and ischemia-reperfusion syndrome. The 70-kDa Hsp (Hsp70) plays a central role in these effects. Hsp70 is strongly heat inducible, and its expression was found to be increased after mild feverlike hyperthermia in rat lung, liver, kidney, retina (Tytell et al 1994; Fawcett et al 1997), and rabbit cerebellum (D'Souza et al 1998). However, there is no report on whether the thermal effect of fever stimulates Hsp70 expression in human leukocytes. The present study characterizes the effects of feverlike hyperthermia on Hsp70 expression in the major human leukocyte populations: lymphocytes, monocytes, and polymorphonuclear leukocytes (PMNs). To determine the effect of rising temperatures, Hsp70 expression was measured in heat-treated whole blood samples by flow cytometry. In addition, to be as close as possible to the in vivo situation during fever, we externally applied heat to induce whole body hyperthermia in healthy volunteers and measured the effect on Hsp70 expression in leukocytes. The experiments revealed that Hsp70 expression can be stimulated by feverlike hyperthermia in all leukocytes but to a different extent. These results indicate that the systemic increase of body temperature during fever is able to stimulate Hsp70 expression in peripheral blood leukocytes. The fever-induced Hsp70 expression could be a beneficial response, protecting leukocytes against the increased concentrations of cytotoxic mediators during inflammation.

MATERIALS AND METHODS

Flow cytometry

Intracellular staining of Hsp70 was performed according to Bachelet et al (1998), with modifications. Briefly, 100 μL of heparinized whole blood was mixed with 5 mL of lysis solution (155 mM ammonium chloride (NH4Cl), 10 mM potassium bicarbonate (KHCO3) [pH 7.4], and 0.1 mM ethylenediamine-tetraacetic acid [EDTA]) and incubated for 10 minutes at room temperature for elimination of erythrocytes. The remaining leukocytes were washed in Hank's Balanced Salt Solution (HBSS) (Bio-Whittaker, Verviers, Belgium) containing 3% fetal calf serum (FCS) at 4°C, resuspended in 250 μL of fixation solution (Cytofix/Cytoperm Plus Kit from BD-Pharmingen, San Jose, CA, USA, containing paraformaldehyde), and incubated for 15 minutes at 4°C for fixation. Then cells were centrifuged for 5 minutes at 600 × g, resuspended in 100 μL of a 1:300 dilution of antibody specific for the inducible Hsp70 (SPA810 by Stressgen, Victoria, Canada) in permeabilization solution (Cytofix/Cytoperm Plus Kit containing saponin), and incubated for 30 minutes at 4°C. After washing in permeabilization solution, cells were incubated in a 1:250 dilution of a fluorescein isothiocyanate–labeled anti-mouse immunoglobulin G1 (IgG1) antibody (BD-Pharmingen, San Jose, CA, USA) for 30 minutes at 4°C. Then cells were washed again in permeabilization solution and incubated in a 1:10 dilution of a PerCP-labeled anti-CD14 antibody (clone RMO52 from Immunotech, Marseille, France) in HBSS containing 3% FCS for 30 minutes at 4°C. Negative control staining was performed with appropriate fluorescence-labeled isotype IgG antibodies (Immunotech, Marseille, France). Stained cells were analyzed using a flow cytometer (Epics XL, Coulter, Miami, FL, USA). A total of 10 000 events were acquired, and gating based on CD14 staining was performed to differentiate lymphocytes (CD14−), PMNs (low CD14 expression), and monocytes (high CD14 expression). The mean fluorescence intensity (MFI) detected at 515 nm of the specifically labeled sample subtracted by the MFI of the isotype control sample was used for quantitation of Hsp70 expression. For identification of lymphocyte subpopulations, leukocyte preparations were stained with additional antibodies: CD3-PC5 (clone UCHT1), CD4-PE (clone 13B8.2), CD8-PE (clone B9.11), CD16-PE (clone 3G8), and CD56-PE (clone NKH-1). All antibodies were obtained from Immunotech, Marseille, France.

Isolation and culture of monocytes

Monocytes were prepared from human leukocyte concentrates by immune affinity chromatography using the MACS monocyte isolation kit (Milteny Biotech, Bergisch-Gladbach, Germany). Briefly, 30 mL of leukocyte concentrate was layered on 60 mL of Ficoll-Paque (Amersham Pharmacia Biotech, Freiburg, Germany) and centrifuged for 30 minutes at 1600 rpm. The plasma supernatant was removed and used later as autologous plasma. The peripheral blood mononuclear cell layer was transferred to 30 mL of phosphate-buffered saline containing 2 mM EDTA and washed 2 times in this solution. Afterward, cells were resuspended in 30 μL/107 cells of MACS buffer (containing 10% autologous plasma). Then T cells, granulocytes, natural killer cells, B cells, dendritic cells, and basophils were labeled by a cocktail of hapten-modified CD3, CD7, CD19, CD45RA, CD56, and anti-IgE antibodies. After washing cells in MACS buffer, the nonmonocytes were magnetically labeled using MACS MicroBeads coupled to an antihapten antibody. Highly pure monocytes (more than 90%) were achieved by retaining the nonmonocytes on a separation column. Monocytes were washed in phosphate-buffered saline, suspended in RPMI-1640 (containing 2 mM glutamine and 10% human AB serum) to a final concentration of 2 × 106 cells/mL, and placed in Teflon flasks to avoid unspecific activation by contact with plastic. After an overnight culture, monocytes were used for the experiments described.

In vitro heat treatment

To trigger Hsp70 synthesis, 1 mL of either whole blood or isolated monocyte preparation was transferred to a Teflon flask and placed for 2 hours in an incubator at the indicated temperature (39°C, 41°C, or 42°C). Then cell suspensions were transferred to a 37°C warm incubator and cultured for 3 hours. This treatment had no effect on the viability of cells or on the composition of the leukocyte subpopulations as assayed by propidium iodide exclusion assay and a quantitative hematology analyzer.

In vivo heat treatment: study design

The study was designed as a prospective, randomized, crossover, controlled clinical trial in 6 healthy men and 6 healthy women aged 22–29 years. They were placed in a hot water bath (water temperature, 39.5°C) for 2 hours. The volunteers were immersed up to the neck with one arm above the water. Tympanic temperature was recorded using an electronic thermometer (Braun AG, Kronberg, Germany). One week later volunteers served as their own controls, being immersed in a thermoneutral water bath (36°C). Blood samples were taken before the experiment and 3, 6, and 22 hours following treatment. Blood was drawn into heparin Vacutainer tubes, in a total amount of 110 mL, and analyzed within 30 minutes. The protocol was approved by the local ethics committee of the University of Vienna.

RESULTS

Constitutive Hsp70 expression in human leukocytes

Hsp70 is a stress-inducible member of the Hsp70 protein family (Lindquist and Craig 1988). However, Hsp70 was found to be expressed also in several cell types and tissues under nonstress conditions (Fincato et al 1991; Weingartmann et al 1999). To investigate whether Hsp70 is also expressed constitutively in human leukocytes, whole blood samples were taken from healthy volunteers and Hsp70 was quantified by flow cytometry. The results are shown in Figure 1. Whereas lymphocytes express a very low level of Hsp70, monocytes and PMNs show considerably higher Hsp70 levels (10-fold and 20-fold compared with lymphocytes, respectively). The difference in basal levels of Hsp70 seems to be specific for the cell type and not related to the size of leukocytes, because the signal does not correlate with the cell volume (lymphocytes, 0.23 picoliter [pl;] monocytes, 0.47 pl; and PMNs, 0.45 pl). As shown in Figure 2, the absolute levels vary among the individuals but the rations remain the same. Higher constitutive Hsp70 expression of one leukocyte population correlates with higher expression in all other leukocyte subtypes. Thus, Hsp70 is constitutively expressed in human leukocytes but varies considerably between the different cell types.

Fig. 1.

Fig. 1.

Open in a new tab

 Constitutive Hsp70 expression in leukocytes. Whole blood was taken from healthy volunteers (n = 12), and Hsp70 expression was investigated by flow cytometry. Hsp70 is expressed as MFI, and the error bars represent SDs. The statistical evaluation by Student's t-test is indicated.

Fig. 2.

Fig. 2.

Open in a new tab

 Comparison of the constitutive Hsp70 expression in different leukocyte subtypes. Whole blood was taken from healthy volunteers, and Hsp70 expression was investigated by flow cytometry. Hsp70 levels in leukocyte subtypes of the same donor are compared. (A) Comparison of PMNs with monocytes, (B) comparison of PMNs with lymphocytes, and (C) comparison of monocytes with lymphocytes. Each data point represents one human donor (n = 36). Hsp70 levels are expressed as MFI, and the error bars represent SDs. The black line represents the linear regression, and the correlation coefficient is indicated

Induction of Hsp70 expression in leukocytes by in vitro hyperthermia

During a febrile response, leukocytes are confronted with a rise in body core temperature of 38°C to 41°C. Occasionally, the body temperature can increase up to 42°C (Tomlinson 1975). To investigate the influence of different temperatures on Hsp70 expression, we took whole blood from healthy volunteers, exposed it to feverlike temperatures of 39°C, 41°C, and 42°C for 2 hours, and allowed cells to recover for an additional 3 hours at 37°C. Then Hsp70 expression was assayed by flow cytometry. Cells cultivated continuously at 37°C showed no increase in Hsp70 compared with the expression immediately after blood drawing (Fig 3). Hyperthermia induced Hsp70 in all leukocytes. However, the degree of induction varied remarkably among the 3 subtypes. In PMNs no increase in Hsp70 expression could be found after heat treatment at 39°C. At higher temperatures (41°C or 42°C), PMNs showed only weak induction (30%). Monocytes behaved differently; at 39°C they already showed a 6-fold higher Hsp70 content than the 37°C control. After heat treatment of monocytes at 41°C or above, Hsp70 expression increased further to 10-fold. In contrast, lymphocytes showed no induction at 39°C, a small increase (30%) at 41°C, and a very strong induction of Hsp70 (500%) at 42°C. However, because of the low basal Hsp70 expression, lymphocytes remain the population with the lowest absolute Hsp70 levels even after heat shock.

Fig. 3.

Fig. 3.

Open in a new tab

 Induction of Hsp70 in human leukocytes by heat shock of whole blood. Whole blood was taken from healthy volunteers (n = 6) and incubated for 2 hours at 37°C, 39°C, 41°C, or 42°C. Then cells were allowed to recover for 3 hours at 37°C. Hsp70 expression in PMNs, monocytes, and lymphocytes was investigated by flow cytometry. MFI was used to quantitate the flow cytometry signal, and the results are expressed as percentage of 37°C control cells (black bar). The error bars represent SDs. The hatched bars indicate Hsp70 expression measured immediately after taking blood without subsequent culturing. The statistical evaluation by Student's t-test is indicated (n.s. = not significant).

Monocytes are the most remarkable population. Although they exhibit an intermediate constitutive Hsp70 expression (Fig 1), monocytes have by far the strongest Hsp70 induction following hyperthermia. Heat treatment increases Hsp70 expression of monocytes to an MFI of 117.5, whereas in PMNs the MFI reaches only 24.6. In the experiments described herein, all blood cells were heat treated together in one vial. Thus, the strong induction of Hsp70 in monocytes could be either the result of a direct thermal effect on monocytes or mediated by additional activation by other leukocytes. For example, heat treatment might stimulate one leukocyte population to produce cytokines, which could lead to an activation of other populations, thereby affecting their Hsp70 expression. To prove that the strong Hsp70 induction in monocytes of whole blood samples is due to a direct thermal effect, we repeated the in vitro heat shock experiments described herein with isolated monocytes. Monocytes were isolated from whole blood by magnetic cell sorting and cultured overnight. Then monocytes were exposed for 2 hours to 42°C and allowed to recover at 37° for 3 hours. Hsp70 expression was determined by flow cytometry, and the results were compared with whole blood samples heat treated in the same way. Figure 4 shows 1 of 2 independent experiments. The constitutive Hsp70 expression of isolated monocytes corresponds to that observed in monocytes from whole blood samples. In the heat-treated samples, Hsp70 expression increased in both isolated monocytes and monocytes of whole blood to the same level. This indicates that the strong Hsp70 induction in monocytes observed in the heat-treated whole blood reflects a direct effect of heat on these cells. However, a small part (<10%) of isolated monocytes remained Hsp70 negative after heat treatment, whereas Hsp70 was induced in all monocytes of the whole blood. This suggests that the isolation procedure might affect the ability of monocytes to express Hsp70.

Fig. 4.

Fig. 4.

Open in a new tab

 Induction of monocyte Hsp70 by heat shock of whole blood (panels C and D) or isolated monocytes (panels A and B). Monocytes were isolated from whole blood of a healthy volunteer and cultured overnight. Then cells were exposed for 2 hours to 37°C and 42°C, and Hsp70 was determined by flow cytometry. Whole blood of the same donor was heat treated in the same way and subjected to flow cytometric analysis. (A and C) CD14 expression (shown as scatter graph against sideward scatter) used to identify monocytes in the whole blood sample (A) and to investigate the purity of the monocyte culture (C). (B and D) The distribution of Hsp70 expression within the monocyte population from whole blood (B) and cell culture (D). The dotted line represents the signal of the isotype control antibody indicating unspecific binding. The additional lines represent the Hsp70 expression of cells heat treated at the indicated temperatures

On closer examination of lymphocytes treated at 42°C, we became aware of a broad distribution of Hsp70 expression possibly related to the different lymphocyte subsets. To clarify this question, we repeated the experiment, exposing whole blood for 2 hours to 37°C and 42°C and allowing cells to recover for 3 hours at 37°C. For identification of the major lymphocyte subsets, cells were stained with specific antibodies and Hsp70 expression was determined by flow cytometry. As shown in Figure 5, the constitutive Hsp70 expression in T-helper cells, cytotoxic T cells, B cells, and natural killer cells differs only slightly from each other (maximum, 25% deviation from the mean). When heat treated, all lymphocytes show a pronounced induction of Hsp70 but the response varied among subtypes. Although T-helper cells and B cells show an increase in Hsp70 of about 200%, cytotoxic T cells exhibit an induction by 400% and natural killer cells by 500%. Taken together, these results show that Hsp70 induction by heat shock depends strongly on cell type and temperature.

Fig. 5.

Fig. 5.

Open in a new tab

 Induction of Hsp70 in different lymphocyte subtypes by heat shock of whole blood. Whole blood was taken from healthy volunteers (n = 6) and incubated for 2 hours at 37° (HS−) and 42°C (HS+). Then cells were allowed to recover for 3 hours at 37°C. Hsp70 expression was investigated by flow cytometry. Leukocyte subtypes were identified by staining with specific antibodies: Th (T-helper cells) CD3+, CD4+, CD8−; Tc (cytotoxic T cells) CD3+, CD8+, CD4−; B (B cells) CD19+, CD3−; NK (natural killer cells) CD3+, CD56+, CD16+. Hsp70 is expressed as MFI, and the error bars represent SDs. Statistical evaluation of the results revealed that the heat-mediated increase of Hsp70 expression was highly significant in all lymphocyte subtypes (P < 0.001)

Induction of Hsp70 expression in leukocytes by in vivo heat treatment of healthy volunteers

The results of the in vitro heat shock experiments reflect the in vivo situation during fever only partially. The stress response of leukocytes to elevated temperatures may be influenced by additional components of the body. To analyze the thermal effect of fever on Hsp70 expression in human leukocytes under physiological conditions, we induced whole body hyperthermia by immersing healthy volunteers in a hot water bath. As shown in Figure 6, a 36°C warm water bath for 2 hours had no effect on the body temperature. In contrast, a 39.5°C hot water bath led to an increase of body temperature to 39°C within 40 minutes. After the end of the hyperthermic exposure, the body temperature decreased immediately to normal values. The heart rate behaved similarly: at 39.5°C the heart rate increased to 100 beats per minute, whereas at 36°C the heart rate remained unaffected. Furthermore, blood pressure was affected by hyperthermia. Although the systolic blood pressure remained stable, the diastolic blood pressure decreased by about 20% within the first 15 minutes and remained at this level until the end of treatment (P < 0.05 at every time during heat treatment). Measurements of the white blood cell counts revealed that all 3 leukocyte populations remained unaffected by a 36°C warm water bath (Table 1). However, an immersion in a 39.5°C hot water bath induced a substantial increase of PMNs at 3 hours and monocytes at 6 hours after the hot water bath. In contrast, lymphocytes showed a slight decrease during hyperthermia, which persisted up to 3 hours after the end of exposure.

Fig. 6.

Fig. 6.

Open in a new tab

 Influence of a warm water bath on body temperature. Healthy volunteers (n = 12) were immersed for 2 hours in a 39.5°C and a 36°C warm water bath. (A) Body temperature as measured by a tympanic thermometer during and after the water bath. (B) Heart rate. (C) Blood pressure

Table 1.

 Influence of hyperthermia on white blood cell counts

graphic file with name i1466-1268-6-4-306-t01.jpg

Open in a new tab

Thus, a 39.5°C hot water bath seems to be a valid model for whole body hyperthermia, whereas the 36°C water bath reflects an isothermal situation and serves as a regular control. We took whole blood samples before and 3, 6, and 22 hours after the water bath and examined Hsp70 expression by flow cytometry. Hsp70 levels observed before heat exposure were in the same range as the constitutive expression shown in Figure 1. Immersing healthy volunteers in a 36°C warm water bath had no effect on Hsp70 expression of their leukocytes (Fig 7). In contrast, a 39.5°C hot water bath led to an increase in Hsp70 levels. As soon as 3 hours after the end of hyperthermia, PMNs showed a slightly elevated Hsp70 expression (15%), which further increased to 40% at 22 hours after treatment. Similar to the in vitro experiments, monocytes showed the highest Hsp70 induction of all leukocyte subtypes: 30% at 3 hours after treatment, 50% at 6 hours, and 60% at 22 hours. Six hours after heat treatment, monocyte Hsp70 expression was more pronounced in volunteers, which reached a higher maximum temperature during the hot water bath (r = 0.770). Lymphocytes also behaved similarly to what was observed in the in vitro experiments. Hyperthermia had only a very weak effect on Hsp70 expression in these cells: the expression increased slowly and only 22 hours after treatment reached a significant difference to the isothermal control. Taken together, these results show that an elevation of body temperature to 39°C induces Hsp70 expression in all leukocyte subtypes. Similarly to the in vitro results, the induction varies, depending on the cell type. Again, monocytes show the strongest Hsp70 induction. However, in the in vivo experiment, there was much less induction of monocyte Hsp70 than with a similar in vitro warming.

Fig. 7.

Fig. 7.

Open in a new tab

 Induction of Hsp70 expression in leukocytes by in vivo heat treatment of healthy volunteers. Healthy volunteers (n = 12) were immersed for 2 hours in a 36°C or 39.5°C warm water bath to increase body temperature. Whole blood samples were taken at different time points, and Hsp70 expression in leukocytes was investigated by flow cytometry. MFI was used to quantitate the flow cytometry signal, and the results are expressed as percentage of the signal before treatment. The statistical analysis compares the heat-treated group with the isothermal control (*P = 0.05–0.01; **P = 0.01–0.005; ***P = 0.005–0.001)

DISCUSSION

This study characterizes the influence of feverlike hyperthermia on Hsp70 expression in different human leukocyte subsets: lymphocytes, monocytes, and PMNs. Using flow cytometry, we determined Hsp70 expression directly in heat-treated whole blood samples. This approach circumvents cell isolation and cultivation that could affect cell physiology and consequently our measurements. In addition, to be as close as possible to the physiological situation during fever, we induced whole body hyperthermia in healthy volunteers and determined the effect on Hsp70 expression in leukocytes. Lymphocytes, monocytes, and PMNs showed a varying degree of Hsp70 induction in both experimental settings, in vitro and in vivo. Thus, our results give new insights on the immunological effect of fever. Furthermore, they indicate that the response to stress depends not only on the degree of stress but also on the cell type.

Hsp70 was found to be constitutively expressed in untreated leukocytes. Although Hsp70 was barely detectable in lymphocytes, it could be measured in high amounts in phagocytic leukocytes: monocytes and PMNs. This corresponds to earlier observations at the messenger RNA (mRNA) level. Using Northern blot analysis, Fincato et al (1991) showed that appreciable levels of Hsp70 mRNA are present in monocytes and PMNs, but the transcript was nearly absent in lymphocytes. Phagocytes are the main producers of reactive oxygen species, which have been proposed to be important regulators of Hsp expression (Polla and Cossarizza 1996). Since phagocytes become activated very early in the inflammatory process, we suppose that a high constitutive Hsp70 level is helpful to protect these cells from the first reactive oxygen species challenge.

Heat treatment of whole blood samples led to a remarkable induction of Hsp70 in monocytes: they increased their Hsp70 expression 10-fold after heat treatment at 41°C. A similar effect was observed after heat treatment of isolated monocytes in Western blot experiments (Pizurki and Polla 1994; Jacquier-Sarlin et al 1995) and using flow cytometry (Vayssier et al 1998). A recent study reported that heat-induced Hsp70 expression is not uniform within the monocyte population (Bachelet et al 1998). The authors exposed isolated monocytes to different heat treatments and determined Hsp70 expression by flow cytometry. They found that the increase in Hsp70 expression is due to both an increase in Hsp70 expression levels and an increase in the number of cells expressing Hsp70. Higher Hsp70 inducibility coincided with higher CD14 expression observed in a subset of the monocyte population.

In the present study, however, monocytes from heat-treated whole blood samples showed a homogeneous expression of Hsp70 and CD14 at all temperatures (Fig 5). CD14 expression is known to be induced by a number of different environmental stimuli, including cell isolation procedures (Bernardo et al 1997), attachment to plastic surface (Hopkins et al 1995), and culturing in the presence of FCS (Bernardo et al 1997). Concomitant with this increase in CD14, monocytes become activated, which is known to affect synthesis of Hsps (Polla et al 1995). In the present study, we tried to reduce these effects to a minimum. Monocytes were investigated directly in whole blood within 4 hours after blood sampling. This may explain the homogeneous expression pattern of CD14 and Hsp70 in these samples. Furthermore, in the cultures of isolated monocytes, we tried to keep cell activation low. Monocytes were isolated by a gently magnetic sorting protocol, suspended in medium containing human AB serum instead of FCS, and cultured in Teflon flasks, which prevent monocyte adhesion. Despite all precautions, isolated monocytes showed a slightly irregular Hsp70 expression pattern: approximately 10% remained Hsp70 negative.

In contrast to monocytes, lymphocytes showed a heterogeneous Hsp70 expression pattern after heat treatment of whole blood. This is in accordance with recently published observations that demonstrate by in vitro heat shock of isolated human lymphocytes that Hsp70 induction does not occur in all cells and occurs to a varyingly extent (Dressel and Gunther 1999). We further characterized that among all major lymphocyte subsets natural killer cells show by far the strongest (more than 6-fold) Hsp70 induction after heat treatment at 42°C. In these cells Hsp70 reaches levels similar to those observed in untreated monocytes. However, at slightly lower temperatures (41°C) the Hsp70 induction in lymphocytes was negligible, which indicates that lymphocytes respond only to very strong heat stress. Since the body temperature rarely exceeds 41°C during febrile illness, it cannot necessarily be concluded that the Hsp70 induction observed at 42°C also occurs under physiological conditions. Furthermore, the data may not reflect changes that occur after prolonged heat exposure for several hours. Nevertheless, it can be said that the heterogeneity of Hsp70 induction observed in lymphocytes certainly reflects a cell type–specific variation of the stress response.

In contrast to lymphocytes, PMNs showed a weak Hsp70 induction at all temperatures applied (only 30% ± 13% at 42°C). Thus, the response of PMNs to hyperthermia differs strongly from that of the other phagocytic leukocytes, the monocytes. It has already been reported that isolated PMNs and isolated monocytes induce Hsps to a different extent in response to exogenous stimuli (Polla et al 1995). These experiments investigated the ability of cells to synthesize Hsps in response to phagocytosis and showed that only monocytes were responsive, although both cell types showed the same degree of phagocytosis. In comparison, our in vitro experiments revealed strong variations in the inducibility of Hsp70 in human leukocytes by feverlike temperatures. These variations depend on temperature and cell type. Especially at mild hyperthermia (39°C), which is commonly reached during febrile illness, monocytes show a much stronger Hsp70 expression than PMNs or lymphocytes.

To mimic the in vivo situation during fever more closely, we determined the effect of whole body hyperthermia on leukocyte Hsp70 expression. Therefore, we immersed healthy volunteers in a hot water bath, thereby increasing the body temperature. A similar approach was used by others to characterize the thermal effect of fever on natural killer cell activity (Kappel et al 1991) and cytokine production (Robins et al 1995). In our experiments, whole body hyperthermia resulted in an increase in body temperature to 39°C, an elevated heart rate, and a decrease in diastolic blood pressure. Similar changes of these parameters can also be observed during fever or other hyperthermia-inducing treatments such as sauna conditions (Rowell 1983). In addition, we found a slight decrease of lymphocytes during the hyperthermic exposure and a PMN and monocyte leukocytosis after the hot water bath. This is in accordance with data by Fabricius et al (1978) that reported that hyperthermia induced by radiofrequency leads to leukocytosis and a decrease in lymphocyte counts in healthy subjects.

Hsp70 expression was induced in all leukocytes. The degree of induction varied among the different cell types as observed in the in vitro experiments: monocytes showed a stronger response than PMNs, whereas the Hsp70 induction in lymphocytes was barely detectable. However, it has to be mentioned that the induction in monocytes was not as pronounced as that in the in vitro setting. Heating the body for 2 hours at 39.5°C resulted in an increase of Hsp70 to 160% of normal values, whereas exposing whole blood for 2 hours to 39°C lead to an increase to 600%. However, comparison of in vitro and in vivo heat treatment is of course limited. On one hand, the hyperthermic challenge of monocytes during whole body hyperthermia is substantially milder then during whole blood heat treatments. The core body temperature of volunteers reached 39°C only after 45 minutes, thus reducing the net incubation at 39°C to about 75 minutes (Fig 7A), whereas the blood samples were kept at 39°C for the entire 120 minutes. On the other hand, monocytes under physiological body conditions are better supplied with plasma glutathione (GSH) than monocytes in isolated blood samples. The plasma GSH level decreases rapidly by auto-oxidation, when blood is being taken (Vina et al 1995). GSH is involved in many cellular functions, especially in the antioxidant defense, and Hsp70 induction is known to be enhanced in GSH-depleted cells (Freeman et al 1993). Therefore, it is to be expected that leukocytes are more susceptible to heat stress in vitro than in vivo. In addition, the monocyte leukocytosis observed after heat exposure may account for the moderate Hsp70 induction in these cells. Newly formed monocytes might exhibit a lower sensitivity to heat treatment, thereby reducing the overall monocyte Hsp70 content. In that case, we would expect 2 monocyte populations: one exhibiting high and one showing low Hsp70 expression. Closer evaluation of the cytometry data, however, gave no indication of heterogeneous monocyte Hsp70 expression. Thus, the role of leukocytosis in Hsp70 induction after heat treatment of healthy subjects remains unclear.

Despite the differences observed in the extent of monocyte responses, the cell type–specific variations in heat-mediated Hsp70 expression of leukocytes are consistent in both experimental settings and therefore seem to be due to differences in the nature of cell types. For that reason, we propose that the strong Hsp70 induction observed in monocytes in this study is of particular importance for monocyte function. When confronted with invading microorganisms, monocytes phagocytose and kill the pathogen by superoxide radical formation. Hsp70 is known to protect cells against a superoxide radical challenge (Polla and Cossarizza 1996). Thus, the high Hsp70 induction by febrile temperatures will help monocytes survive the oxidative stress during inflammation. However, in PMNs, which also act as phagocytes, Hsp70 is induced by hyperthermia to a lesser degree. We propose that the higher Hsp70 induction in monocytes helps these cells survive long enough to fulfill their major function, antigen processing and presentation. In contrast, PMNs act primarily to eliminate pathogens. Accordingly, PMNs have only a short half-life of 6 to 7 hours, whereas monocytes can survive 1 to 2 days in the blood flow (Handin et al 1995). Hsp70 induction is mainly regulated by the heat-inducible transcription factor heat shock factor 1 (HSF-1) (Morimoto et al 1996). A recent study shows that HSF-1 plays an important role in regulating monocyte and macrophage cytokine expression (Singh et al 2000). HSF-1 down-regulates lipopolysaccharide-induced transcription of tumor necrosis factor α in the raw 264.7 cell line. The induction of Hsp70 in monocytes by whole body hyperthermia as demonstrated by us implies that HSF-1 is also activated under these in vivo conditions and suggests that tumor necrosis factor α transcription is reduced by elevated body temperatures. The results of this study not only underline the importance of understanding how shifts in body temperature affect biological processes but also suggest that Hsp70 induction in human leukocytes by mild hyperthermia helps leukocytes fulfill their function in the immune response during febrile illness.

Acknowledgments

We are grateful to Christine Brostjan and Susanne Oehler for helpful discussions and for proofreading the manuscript. This study was supported by the Jubilaeumsfonds der Oesterreichischen Nationalbank 8311.

REFERENCES

  1. Bachelet M, Mariéthoz E, and Banzet N. et al. . 1998 Flow cytometry is a rapid and reliable method for evaluating heat shock protein 70 expression in human monocytes. Cell Stress Chaperones. 3(3):168–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bernardo J, Billingslea AM, Ortiz MF, Seetoo KF, Macauley J, Simons ER. Adherence-dependent calcium signaling in monocytes: induction of a CD14-high phenotype, stimulus-responsive subpopulation. J Immunol Methods. 1997;209(2):165–175. doi: 10.1016/s0022-1759(97)00157-9. [DOI] [PubMed] [Google Scholar]
  3. Coxe JR 1846 The Writings of Hippocrates and Galen. Lindsay and Blakiston, Philadelphia. [Google Scholar]
  4. Dressel R, Gunther E. Heat-induced expression of MHC-linked Hsp70 genes in lymphocytes varies at the single-cell level. J Cell Biochem. 1999;72(4):558–569. [PubMed] [Google Scholar]
  5. D'Souza CA, Rush SJ, Brown IR. Effect of hyperthermia on the transcription rate of heat-shock genes in the rabbit cerebellum and retina assayed by nuclear run-ons. J Neurosci Res. 1998;52(5):538–548. doi: 10.1002/(SICI)1097-4547(19980601)52:5<538::AID-JNR6>3.0.CO;2-D. [DOI] [PubMed] [Google Scholar]
  6. Fabricius HA, Neumann H, Stahn R, Engelhardt R, Lohr GW. Changes in clinical and immunological laboratory parameters of healthy adults after exposure to a one-hour 40 degree C-whole-body hyperthermia (author's transl) Klin Wochenschr. 1978;56(21):1049–1056. doi: 10.1007/BF01476552. [DOI] [PubMed] [Google Scholar]
  7. Fawcett TW, Xu Q, Holbrook NJ. Potentiation of heat stress-induced hsp70 expression in vivo by aspirin. Cell Stress Chaperones. 1997;2(2):104–109. doi: 10.1379/1466-1268(1997)002<0104:pohsih>2.3.co;2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fincato G, Polentarutti N, Sica A, Mantovani A, Colotta F. Expression of a heat-inducible gene of the Hsp70 family in human myelomonocytic cells: regulation by bacterial products and cytokines. Blood. 1991;77(3):579–586. [PubMed] [Google Scholar]
  9. Freeman ML, Sierra Rivera E, Voorhees GJ, Eisert DR, Meredith MJ. Synthesis of hsp-70 is enhanced in glutathione-depleted Hep G2 cells. Radiat Res. 1993;135(3):387–393. doi: 10.2307/3578879. [DOI] [PubMed] [Google Scholar]
  10. Handin I, Lux S, and Stossel T 1995 Blood, Principles, and Practice of Hematology. Lippincott, Philadelphia. [Google Scholar]
  11. Hopkins HA, Monick MM, and Hunninghake GW 1995 Lipopolysaccharide upregulates surface expression of CD14 on human alveolar macrophages. Am J Physiol. 269(6 pt 1). L849–L854. [DOI] [PubMed] [Google Scholar]
  12. Jacquier-Sarlin MR, Jornot L, Polla BS. Differential expression and regulation of hsp70 and hsp90 by phorbol esters and heat shock. J Biol Chem. 1995;270(23):14094–14099. doi: 10.1074/jbc.270.23.14094. [DOI] [PubMed] [Google Scholar]
  13. Kappel M, Stadeager C, Tvede N, Galbo H, Pedersen BK. Effects of in vivo hyperthermia on natural killer cell activity, in vitro proliferative responses and blood mononuclear cell subpopulations. Clin Exp Immunol. 1991;84(1):175–180. doi: 10.1111/j.1365-2249.1991.tb08144.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kluger MJ. Fever: role of pyrogens and cryogens. Physiol Rev. 1991;71(1):93–127. doi: 10.1152/physrev.1991.71.1.93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kluger MJ, Kozak W, Conn CA, Leon LR, Soszynski D. Role of fever in disease. Ann N Y Acad Sci. 1998;856:224–233. doi: 10.1111/j.1749-6632.1998.tb08329.x. [DOI] [PubMed] [Google Scholar]
  16. Kramer MS, Campbell H 1993 The Management of Fever in Young Children With Acute Respiratory Infections in Developing Countries. World Health Organization, Geneva. WHO/ARI 93.30. [Google Scholar]
  17. Lindquist S, Craig EA 1988 The heat-shock proteins. Annu Rev Genet 22: 631–677. [DOI] [PubMed] [Google Scholar]
  18. Morimoto R, Kroeger P, and Cotto J 1996 The transcriptional regulation of heat shock genes: a plethora of heat shock factors and regulatory conditions. EXS 77: 139–163. [DOI] [PubMed] [Google Scholar]
  19. Oehler R, Zellner M, Hefel B, Weingartmann G, Spittler A, Struse HM, and Roth E 1998 Influence of heat shock on cell volume regulation: protection from hypertonic challenge in a human monocyte cell line. FASEB J 12: 553–560. [DOI] [PubMed] [Google Scholar]
  20. Pizurki L, Polla BS 1994 cAMP modulates stress protein synthesis in human monocytes-macrophages. J Cell Physiol 161: 169–177. [DOI] [PubMed] [Google Scholar]
  21. Polla BS, Cossarizza A 1996 Stress proteins in inflammation. In: Stress-Inducible Cellular Responses, ed Feige U, Morimoto R, Yahara I, Polla B. Birkhäuser Verlag, Basel, 375–391. [DOI] [PubMed] [Google Scholar]
  22. Polla BS, Stubbe H, Kantengwa S, Maridonneau Parini I, Jacquier Sarlin MR. Differential induction of stress proteins and functional effects of heat shock in human phagocytes. Inflammation. 1995;19(3):363–378. doi: 10.1007/BF01534393. [DOI] [PubMed] [Google Scholar]
  23. Robins HI, Kutz M, and Wiedemann GJ. et al. . 1995 Cytokine induction by 41.8 degrees C whole body hyperthermia. Cancer Lett. 97(2):195–201. [DOI] [PubMed] [Google Scholar]
  24. Rowell LB 1983 Cardiovascular adjustments of thermal stress. In: Handbook of Physiology, ed Shepherd JT, Abbound EM, Geiger SR. American Physiological Society, Bethesda, MD, 967–1023. [DOI] [PubMed] [Google Scholar]
  25. Singh IS, Viscardi RM, Kalvakolanu I, Calderwood S, Hasday JD. Inhibition of tumor necrosis factor-alpha transcription in macrophages exposed to febrile range temperature: a possible role for heat shock factor-1 as a negative transcriptional regulator. J Biol Chem. 2000;275(13):9841–9848. doi: 10.1074/jbc.275.13.9841. [DOI] [PubMed] [Google Scholar]
  26. Tomlinson WA. High fever: experience in private practice. Am J Dis Child. 1975;129(6):693–696. doi: 10.1001/archpedi.1975.02120430033009. [DOI] [PubMed] [Google Scholar]
  27. Tyrell R 1996 UV activation of mammalian stress proteins. EXS 77: 255–271. [PubMed] [Google Scholar]
  28. Tytell M, Barbe MF, Brown IR. Induction of heat shock (stress) protein 70 and its mRNA in the normal and light-damaged rat retina after whole body hyperthermia. J Neurosci Res. 1994;38(1):19–31. doi: 10.1002/jnr.490380105. [DOI] [PubMed] [Google Scholar]
  29. Vayssier M, Favatier F, Pinot F, Bachelet M, Polla BS. Tobacco smoke induces coordinate activation of HSF and inhibition of NFkappaB in human monocytes: effects on TNFalpha release. Biochem Biophys Res Comm. 1998;252(1):249–256. doi: 10.1006/bbrc.1998.9586. [DOI] [PubMed] [Google Scholar]
  30. Vina J, Sastre J, Asensi M, and Packer L 1995 Assay of blood glutathione oxidation during physical exercise. In: Methods in Enzymology, ed Packer L. Academic Press, San Diego, 237–43. [DOI] [PubMed] [Google Scholar]
  31. Weingartmann G, Oehler R, Derkits S, Oismuller C, Fugger R, Roth E. Hsp70 expression in granulocytes and lymphocytes of patients with polytrauma: comparison with plasma glutamine. Clin Nutr. 1999;18(2):121–124. doi: 10.1016/s0261-5614(99)80064-8. [DOI] [PubMed] [Google Scholar]

Articles from Cell Stress & Chaperones are provided here courtesy of Elsevier

RESOURCES