SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

In many recent publications, supposed athermal effects of water-filtered infrared A (wIRA) irradiation are discussed. Those effects are mainly attributed to wavelengths in the range from 780 to 1440 nm, and should not result from warming of cellular water or any aqueous medium surrounding the irradiated sample caused by wIRA absorption. Athermal effects are considered to be induced directly by absorption of different wavelengths of the wIRA spectrum by cellular molecules or structures except water. To distinguish between thermal and athermal effects, irradiated samples have to be subjected to a very effective and precise temperature homeostasis. Any experimental effects can only be attributed to pure athermal effects, if the temperature of the irradiated samples is verifiably constant and does not result in hyperthermia. Here, data of temperature distribution in Petri dishes of different types filled with aqueous medium are presented which were estimated by model calculation for different setups of cooling. Additionally, the real temperature development was directly measured. Such a cooling unit enables long-term application of high wIRA irradiances and large doses without any detectable warming of the irradiated samples, in single cell layers. Using such a setup, thermal and athermal effects can be compared and in addition to that quantified.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Exposure of uncovered skin to spectral parts of solar radiation reaching the Earth’s surface is a fundamental part of current scientific investigation with a focus on the effects and interactions of the single ranges (UV, visible [VIS] and IR) with biological material.

Infrared radiation is an important part of solar spectral irradiance at Earth’s surface. IR is divided into the sub-ranges IRA (780–1400 nm), IRB (1400–3000 nm) and IRC (3000–1 000 000 nm). The solar IR emission (IRA + IRB + IRC) represents about 45% and IRA about 30% of the whole solar irradiance (UV + VIS + IR). As the atmosphere virtually eliminates wavelengths longer than about 2500 nm, only a negligible part of radiation emitted in the sub-range of IRB reaches the surface of the Earth. Because of significant absorption in the atmosphere mainly caused by water, oxygen, carbon dioxide and aerosol, short-wavelength IR radiation at several wavelengths is significantly decreased or even completely eliminated. The resulting trimmed spectrum can be compared with the artificial “water-filtered infrared A” (wIRA).

Recently, several publications suggested direct effects of wIRA that are actually comparable with those directly induced by UVA/B irradiation (1,2). Indeed, the absorbance of pure water in the range from 200 to 400 nm is small compared with that shown in the short-wavelength sub-range of IR (IRA). Thus, thermal effects induced by absorbance of water in the UV range play a minor role compared with photo-chemical effects of interaction with DNA, proteins and lipids. In contrast, the absorption coefficient of water increases strongly for increasing wavelengths even within the sub-range of IRA; because of the fact that water is the most abundant component in living mammalian cells, compared with unfiltered IRA radiation and under the assumption of identical irradiance and geometry of exposure, thermal effectiveness of wIRA is significantly decreased (usually by a factor of 2 or more) (3) but not eliminated.

In the range from 780 to 10 000 nm, the absorption of water increases from a factor of 1 (at 780 nm) to about 106 m−1 (at 3000 nm) and shows values of about 105 m−1 at about 10 000 nm. Thus IRA, IRB, IRC or even broad-band IR application is always accompanied by a temperature increase of the irradiated sample which depends on dose and irradiance and which is accelerated with increasing water content of the irradiated object. This makes it very difficult to distinguish cellular responses or effects that are caused by hyperthermia or heat-shock response from effects that are in fact caused by a direct interaction of cellular molecules with the applied (w)IRA-irradiation.

The spreading application of wIRA in therapy or wellness makes a further investigation of wIRA more important. Therefore, several publications were focusing on the thermal or athermal effects of wIRA applied to mammalian cells, often with contradictory results. In different works, a variety of responses to (w)IR(A) irradiation has been described, like an induction of the metalloproteinases MMP-1 (1,4–7), -2 (8), -3 (9), -9 (6) and -13 (9), TGF-β1 (8), activation of the MAPK pathways was observed, release of cytochrome c and Smac/DIABLO (10) from mitochondria, induction of Bax translocation (10) from cytosol to nucleus, increase of HSP27 (10) and -70 (11), a general increase of ROS formation (1,12), decrease in cellular carotenoid concentration (13) and an induction of trypase (14) and p53 (15). However, other groups did not find any induction of proteins counteracting oxidative stress (like heme-oxygenase, NO-releasing enzymes, SOD, or heat-shock proteins) (16). An overview of the often contradictory experimental outcomes after (w)IR(A) irradiation of skin has been brought together by Piazena and Kelleher (3). Among other cellular reactions an induction of different metalloproteinases (3) by athermal wIRA effects is discussed. Unfortunately, most of the published work suggesting athermal effects of wIRA lacking an exact description of the experimental setup used to irradiate the samples, or do not mention, if and how the cells were cooled during wIRA-irradiation or even if the temperature during exposition was monitored.

In this work, we show an experimental setup that can be used to minimize temperature changes of the irradiated cells during wIRA exposure. The setup was first computed and refined using simulation software (Comsol Multiphysics 4.0a), then constructed and measured in a real application to verify the correlation between the computed simulation and the real one. The real model turned out to preserve temperature homeostasis even in long-term irradiations using extreme amounts of wIRA (up to irradiances of 300 mW/cm2 [or 3000 W/m2] until after 4 h of exposure a dose of 4320 J/cm2 was accumulated) (12). For comparison, at noontime in central Europe wIRA irradiances by sunlight exposure will not exceed about 200 W/m2 (3). Therefore, this study has two main intentions: the first is the investigation of the performance of a representative cooling systems often used in wIRA studies. The second is the development of a highly effective cooling unit that enables precise temperature homeostasis and control even in long-term wIRA irradiations at high irradiances.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

To achieve a highly effective cooling setup, various experimental setups and computer simulations were tested. From those experiences, we decided to construct a water cooler in combination with a glass bottom dish.

Thermal analysis via computer simulation.  For a physically accurate computer simulation of the different air- and water-cooling setups, the program “Multiphysics 4.0a,” version 4.0.0.982 from Comsol was used. For modeling of more complex structures of the simulated setups, the available “primitives” of “Multiphysics 4.0a” in combination with different Boolean operators were used. The physical material characteristics like heat conductivity and heat capacity are taken from the integrated material library of the software itself and were checked via physical reference books. The following parameters were used in our simulations for heath capacity (air: 1005, medium: 4184, borosilicate glass: 980 and polystyrene: 1800 J/kg·K), for heath conductivity (air: 0.026, medium: 0.604, borosilicate glass: 1.16, and polystyrene: 0.17 W/m·K), and for density (air: 1.2928, medium: 1000, borosilicate glass: 2230 and polystyrene: 1050 kg/m3). Changes of those parameters depending on temperature were computed by Comsol via different characteristic lines, fitting the actual physical properties of the simulated materials according to their current temperature.

Several different models of cooling units were compiled for the computer simulation. These models represent different combinations of water and air cooling with a common Petri dish with a diameter of 35.0 mm, a common dish with a diameter of 90.0 mm and a 35.0 mm glass bottom dish (from MatTek Corporation, Massachusetts, article “P35GC-1.0-14-C”) with a bottom hole of 14 mm capped by a thin glass-coverslip of 0.14 mm in thickness. Each of the dishes was 10.0 mm in high, while the walls of the used Petri dishes were defined to be 1.0 mm thick, which represents a good approximation of the producers’ data, showing slight aberrations from a constant wall thickness. The dish material was polystyrene and the glass coverslip of the glass bottom dish was defined as borosilicate glass (according to the manufacturer). Air cooling was simulated by a dish placed on a tempered plate surrounded by air, the water cooling by an almost completely submerged dish in a water bath of a constant temperature. In general, the room air temperature of both environments was defined as 298 K, the temperature of the water bath as 310 K. Energy input by wIRA absorption of the aqueous cell medium and the cell culture material was simulated by a continuous energy influx. As further experiments revealed that the wIRA absorption of the cell culture material is minimal (12), the main energy input into the system was considered to be due to the wIRA absorption of the aqueous cell medium, while only a low amount was attributed to the cell culture material in the used computer simulation. The cell medium was simulated by a cylinder on the bottom of the Petri dishes 2 mm in height. Further experiments showed an absorption of about 10% of the wIRA irradiation by such a water layer (12). Thus, the model system was defined to absorb a continuous energy input of 10% of the applied wIRA irradiance via the medium and additional 1% via the cell culture material.

Finally, the steady-state temperature of the whole system (Petri dish and cell medium) was computed, resulting from the temperature of the surrounding water and/or air and the applied amount of wIRA irradiation. The model did not contain a layer of single cells covering the (glass-)bottom of the Petri dish, because such an extreme complexity of the model was not necessary for our objectives. Rather it was assumed, considering the very small volume of a singed cell layer compared with the volume of the surrounding aqueous medium, that the cells would have the same temperature as the cell medium on the (glass-)bottom of the Petri dish.

Experimental setup.  Every compiled computer simulation was assembled as a real existing model, too, to allow a comparison of the computed analysis with the real setup. The air-cooling unit was a combination of a large transparent Petri dish with a diameter of about 20 cm that was in direct contact with permanently stirred cooling water at a temperature of 310 K. The irradiated sample dishes were placed on that large Petri and different colors (black and white) of the large dish were simulated by covering it with the colored paper. In this case, attention was paid both to a direct contact of paper and plate and a very planar bearing of it. The uncovered dish is denoted as “transparent plate.”

The water-cooling unit was realized by different Petri dishes (with and without glass bottom) that were almost completely submerged in the tempered and stirred cooling-water bath by a ring-shaped fixture. The used water bath was the Haake“SC 100.”

Lids of the Petri dishes have always been removed during wIRA irradiation, because neither air nor water cooling were able to avoid extreme heating of the samples during application of high wIRA irradiances resulting in complete cell loss, even after short-term exposures.

wIRA exposure and dosimetry.  The used wIRA source was a “Hydrosun 750” radiator from the manufacturer Hydrosun Medizintechnik (Müllheim, Germany), equipped with a 4 mm water filter. An additional longpass filter (“RG-780” from Schott AG, Mainz, Germany) filtered out wavelengths shorter than 780 nm. Sample dishes were irradiated vertically from above and the wIRA irradiance was regulated via the distance between lamp and sample. Maximal irradiances of 3000 W/m2 were applied. Applied wIRA irradiance was measured directly via the “HBM-1” radiometer (from Hydrosun, Müllheim, Germany).

Thermometry.  Direct temperature measurements were performed using a very fine micro-sensor (the Chromel-Alumel Typ K thermoelement “2AB Ac 025” from Thermocoax SAS, Flers Cedex, France) with a tip diameter of 250 μm, allowing precise thermometry in very small volumes. Data readout of the thermoelement were performed with the “K204 Datalogger” (from Voltcraft, Hirschau, Germany). For wIRA irradiation, Petri dishes were filled with a PBS layer 2 mm in high. The temperature in this layer was determined via that microsensor in three different levels: directly at the surface, in the middle and at the bottom of the PBS layer/at the surface of the Petri dish, respectively. Each of the three levels was measured from the center of the Petri dish to its outer border in 1 mm increments. The resulting data were presented as temperature gradients along the radius of the dish. In every case, radial symmetry of the measured temperature distribution was assumed, because the dish was irradiated with wIRA of constant and homogenous intensity across the surface of the exposed dish. The temperature curves shown in the following were achieved by moving the tip of the element in 1 mm steps from the center of the Petri dish to its outer border in the three different high levels of the PBS layer (surface, middle and bottom). According to the manufacturer, the precision of the microsensor is ±0.1 K. Contact-free thermometry on surfaces was performed via an “IR-Thermometer 63” infrared thermometer from Fluke, room temperature (RT) was measured via another thermoelement. For adjustment, temperatures have been measured with two different thermometers (both “IR-Thermometer 63” and the mentioned thermoelement “2AB Ac 025”) and compared.

Spectral irradiance of the used wIRA-lamp.  The spectral irradiance of the used wIRA lamp (a “Hydrosun 750” radiator from the manufacturer Hydrosun Medizintechnik, Müllheim, Germany) is depicted in Fig. 1.

image

Figure 1.  Spectral irradiance of the used wIRA lamp. This graph displays spectral irradiance of the used wIRA-lamp (a “Hydrosun 750” radiator from the manufacturer Hydrosun Medizintechnik, Müllheim, Germany) containing a 4 mm water filter and an additional longpass filter (“RG-780” from Schott AG, Mainz, Germany), that eliminates wavelengths lower than 780 nm.

Download figure to PowerPoint

Measurement of absorption in the IRA-range.  Spectral absorption of cell fractions and cell culture materials was performed using a double monochromator spectroradiometer (Spectro 320D; Instrument Systems, Munich, Germany) in the range from 400 to 1400 nm. The absorption of whole cells and the isolated cytosol did not show any significant difference (data not shown), as displayed in Fig. 2 (upper panel). Transmittance of the cell culture material (polystyrene) is depicted in Fig. 2 (lower panel).

image

Figure 2.  Spectral transmittances of some of the different materials involved in the experimental setup. The upper panel shows the spectral transmittance of PBS/full cell medium (dotted line) and of the isolated cytosol from human fibroblasts (dashed line). The wIRA range is highlighted in light gray. The lower panel shows the spectral transmittance of the cell culture material used (a polystyrene layer, 1 mm thick); the wIRA range is again in light gray.

Download figure to PowerPoint

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Interaction of cells and cell culture material with wIRA

As visible in Fig. 2 (upper panel), the transmittance of both PBS/full cell medium and cytosol is very high in the visible range (shown from 400 to 780 nm) and decreases in the wIRA range (780–1440 nm), especially for wavelengths longer than 900 nm, though only a comparatively low amount of the whole applied wIRA irradiance is absorbed by both medium and cells. However, the used cell culture material shows a high transmittance across the whole wIRA range (780–1440 nm), as shown in Fig. 2 (lower panel).

Effectivity of the air cooling in dependence of the plate color

It turned out that the steady-state temperature of the sample during wIRA exposure undergoing air cooling strongly depended on the color of the plate the irradiated samples are placed on. All samples showed a significant increase of the steady-state temperature from the center of the common Petri dish (no glass bottom) to its outer border. The temperature in the center (from r = 0 to 8 mm) of the dish showed only negligible gradients (the difference between the lowest and highest temperature found in the steady state of the irradiated sample dish during wIRA irradiation) but differed strongly from the desired value of 310 K, except for the white plate (see Fig. 3). The outer areas of the dish (from r = 8 to 16 mm) showed significant temperature gradients that differed significantly from the desired temperature of 310 K. Only the “white plate” provided the desired temperature of 310 K in the bottom’s center of the irradiated dish (from r = 0 to 8 mm) where the cells would be placed.

image

Figure 3.  Temperature distributions of an aqueous medium in wIRA-irradiated Petri dishes undergoing air cooling in dependence of the color of the cooling plate they are placed on in comparison with water-cooling system. For this panel, the wIRA-irradiated (irradiance of 240 mW/cm2) Petri dishes were placed on temperature-controlled (310 K) cooling plates of different colors: a transparent plate (first column from the left), a white (second column) and a black (third column) one. The right column shows the temperature distributions in a Petri dish undergoing water-cooling during wIRA irradiation. The Petri dishes were filled with a 2 mm water layer and its temperature was directly measured at its surface (1), in the middle of the layer (2) and at the bottom of the dish (3), from the center (r = 0 mm) to the outer border (r = 16 mm) of the Petri dish. Each graph represents the averaged results from three independent measurements at a room temperature of 298 K. The dashed line in every panel indicates the desired temperature of 310 K.

Download figure to PowerPoint

Use of our water-cooling setup in combination with a common Petri dish (right panel “water bath” of Fig. 3) massively reduced those gradients, but was still not capable of keeping the sample temperature at 310 K during irradiation, even at the bottom of the dish.

Because of the promising results of the “white plate” setup (second column from left in Fig. 3) that provided exactly 310 K in the center of the irradiated dish, both influence of wIRA irradiance and RT were investigated using this Petri dish.

Effectivity of air cooling in dependence on wIRA irradiance and room temperature

As shown in Fig. 4, the applied wIRA irradiance revealed a massive influence on the steady-state temperature of the irradiated dish and the sample within. These experiments were performed on a white cooling plate. The left column of Fig. 4 (240 mW/cm2 of wIRA applied) shows a central steady-state temperature of about 310 K at the surface, in the middle and at the bottom of the medium layer, that significantly increased in radial direction to the outer border of the dish up to values larger than 313 K. At the bottom of the dish, an area from r = 0 to 8 mm was actually at the desired temperature of 310 K (bottom image of the left column in Fig. 4). At the surface of the medium layer, this area was reduced to r = 0–4 mm. If the wIRA irradiance was reduced to 120 mW/cm2 using the same experimental setup (right column of Fig. 4), the shapes of the individual temperature distributions showed only minor changes compared with the distributions found at an irradiance to 240 mW/cm2, but are shifted along the temperature axis: as shown in the panels of the right column of Fig. 4, the temperature was decreased by about 3 K at the surface of the medium layer, by about 2.3 K in the middle and by about 1.1 K at the bottom of the medium layer. Furthermore, the areas with constant temperature as found in the left column were gone in the middle and the surface of the medium layer. Only at the bottom where the cells would be located, the temperature showed an approximately constant distribution in the range from r = 0 to 8 mm but was significantly lower than the desired temperature of 310 K (bottom image in the right columns of Fig. 4).

image

Figure 4.  Temperature distributions in an air-cooled Petri dish in dependence of the applied wIRA irradiance. For this experiment, a 35 mm Petri dish was placed on a temperature-controlled (310 K) white plate and subjected to air cooling during wIRA irradiation. The temperature of the 2 mm medium layer was directly measured at the surface (1), the middle of the layer (2) and the bottom of the dish (3) from the center of the dish (r = 0 mm) to the outer border (r = 16 mm) in increments of 1 mm. The left column shows the temperature distributions at a wIRA irradiance of 240 mW/cm2, the right column at an irradiance of 120 mW/cm2. The dashed line in every panel indicates the desired temperature of 310 K. Graphs represent averaged results from three different measurements at a room temperature of 298 K.

Download figure to PowerPoint

As a defined change of the RT in a broad range was not practicable, we simulated its effects with our computer simulation. An applied irradiance of 240 mW/cm2 was simulated and the temperature of the surrounding air was changed in increments of 5 K in the range from 288 to 308 K (see Fig. 5). The shape of the temperature gradient did not change but was shifted from a central temperature (at r = 0) of 307.7 K in the dish (at RT = 288 K) to 312 K (at RT = 308 K). The experimental result at a RT of 298 K matched very well the predictions of our computer simulation (see below). In this case, the experiment was performed under air-cooling via a white plate, too.

image

Figure 5.  Bottom temperature in an wIRA irradiated air-cooled Petri dish in dependence of the room temperature (wIRA irradiance = 240 mW/cm2). In this experiment, the Petri dish was placed on a temperature-controlled white plate (310 K) during wIRA irradiation. The temperature at the bottom of the dish was measured directly from the center (r = 0 mm) to the outer border (r = 16 mm) as displayed by the continuous line. The room temperature was 298 K (60% humidity; air ventilation did not show any effect on the experimental outcome, data not shown) and the line displays the averaged results of three different experiments. The dashed lines represent computer simulations of the temperature at the bottom of the dish at different room temperatures between 288 and 308 K.

Download figure to PowerPoint

These results indicate that the constant 310 K in the white plate (see Fig. 1) is the accidental result of the used irradiance, RT/conditions and cooling system efficiency, and by no means applicable for different conditions.

Radial temperature-distribution in air-cooled petri dishes of different diameters

Under same conditions as described above (wIRA irradiance of 240 mW/cm2, RT of 298 K, and a white plate using air-cooling of the irradiated samples), the diameter of the used Petri dish revealed a significant influence on the radial temperature distribution as shown in Fig. 6. A smaller dish (32 mm in diameter, left column of Fig. 6) showed both strong temperature gradients and large differences between the lowest (about 310 K at r = 0) and the highest temperatures detected (up to 313.5 K at r = 16 mm). At the position of irradiated cells at the bottom of the dish, the desired temperature of about 310 K was only obtained in the range from r = 0 to 8 mm. From r = 8 to 16 mm, the temperature strongly increased up to >313 K (bottom panel in the left column of Fig. 6). A larger dish (90 mm in diameter, right column of Fig. 6) showed lower differences between the lowest and the highest temperatures found, but a massive difference (up to 4 K) between the desired temperature of 310 K and the experimentally detected temperature (see Fig. 6, right column). The temperature at the position of cells was significantly increased (from 313.5 K at r = 0 mm to about 312 K at the outer border of the dish) compared of the desired 310 K (see Fig. 6, right column).

image

Figure 6.  Temperature distributions in small (r = 16 mm) and large (r = 42.5 mm) Petri dishes undergoing air cooling during wIRA irradiation. For irradiation, both plates were put on a white temperature-controlled cooling plate (310 K) and filled with aqueous medium. The small plate was filled with 2.0 mL of PBS resulting in a 2 mm layer. The large plate was filled with 24.45 mL, resulting in a 4 mm layer, because 12.72 mL that would have produced a 2 mm layer did not form a cohesive water layer. The results of the small plate (35 mm in diameter) are displayed in the left column, the results of the large plate (90 mm in diameter) in the right one. The temperature was directly measured in 1 mm steps from the center of the dish (r = 0 mm) to the outer border (r = 16 or 42.5 mm, respectively) at the surface of the medium (1), in the middle of the aqueous layer (2) and directly on the bottom of the dish (3). The dashed line in every panel indicates the desired temperature of 310 K. Every graph represents the average of three different experiments the room temperature was 298 K in every case.

Download figure to PowerPoint

Computer simulation of a water-cooled glass-bottom Petri dish

As the above-mentioned setup did not show sufficient stability and the resulting temperature of the cell layers depend on a multitude of parameters, we decided to use a more sophisticated setup. Here, we used a glass bottom dish, because the glass has better thermic conduction properties. To test a more efficient cooling we used also a water-cooling system. A cross-section of the used setup modulated in the computer analysis is shown in Fig. 7 as well as the steady-state temperature distributions found during a simulated application of 240 mW/cm2 wIRA. Directly at the upper side of the glass bottom, the temperature of the medium layer showed only very low divergences (≤0.1 K) from the desired temperature of 310 K.

image

Figure 7.  Computer simulation of the temperature distribution of a water-cooled glass-bottom dish during wIRA irradiation at an irradiance of 240 mW/cm2. The left panel shows a cross-section of the setup used for the computer simulation. The water surrounding the dish was defined to have a constant temperature of 310 K, the air temperature was defined to be 298 K. The temperature between the surface of the glass bottom, normally keeping the cells, and the medium directly above was defined to determine the temperature of the cells in a real irradiation. The temperature distribution returned by the computer simulation is displayed color coded in the right panel of this figure. The differences between the desired temperature of 310 K and the computed temperature distribution at the glass bottom were in the range of +0.1 K.

Download figure to PowerPoint

Figure 8 shows both the computed (dashed lines) and experimentally found (continuous lines) temperature distributions at irradiances of 120 (left column) and 240 mW/cm2 (right column) at the surface (1), the middle (2) and the bottom of the medium layer (3) in a water-cooled wIRA-irradiated Petri dish. The most important result is the surface temperature of the glass bottom, normally carrying the wIRA-exposed cells. Both in 120 and 240 mW/cm2 of wIRA irradiance the temperature of the glass bottom did not significantly differ (below 0.1 K) from the 310 K of the surrounding cooling-water (panel 3 in both columns). The experimentally found temperature distributions matched very well the outcomes of our computer simulation. Maximal differences between desired and actually measured temperature were in the range of ±0.1 K that is within the measuring precision of the used microsensor. According to the computer simulation, the medium temperature is not increased more than +0.1 K at the surface of the glass bottom compared to the desired 310 K. Only an insignificant temperature gradient very close to the outer border of the glass bottom (r = 7 mm) (see Fig. 8, panel 3 in both columns) was revealed by the simulation.

image

Figure 8.  Steady-state temperatures in the medium of glass-bottom Petri dishes at two different wIRA irradiances (120 and 240 mW/cm2) using water cooling. In this figure, the temperature gradients in a 2 mm medium layer from the center (r = 0 mm) of a wIRA-irradiated and water-cooled glass-bottom Petri dish to its outer border (r = 16 mm) at two different wIRA irradiances (left column: 120 mW/cm2, right column: 240 mW/cm2) is shown. The temperature was both directly measured (continuous line) at the surface of the medium layer (1), 1 mm beyond its surface (2) and at the bottom of the dish (3) and calculated by our computer simulation (dotted line). The horizontal dashed line shows the desired temperature of 310 K, set by the temperature of the water bath. The vertical dashed line shows the border of the glass bottom with a radius of 7 mm (Rb). The room temperature was 298 K in every case, the continuous lines show the averaged results of three different measurements.

Download figure to PowerPoint

Direct comparison of air and water cooling

In Fig. 9, both the air and water cooling are compared using the same glass-bottom dish filled with a PBS layer of 2 mm in high at different wIRA irradiances (0–300 mW/cm2). The steady-state temperature was measured in the center of the glass-bottom only. At an irradiance of 0 mW/cm2, the steady-state temperature of both systems was determined by both the used cooling setup and the RT (298 K). In both systems, a linear increase of the temperature dependent from the applied wIRA-irradiance could be found, though during water cooling the difference between the desired temperature of 310 K and the measured temperature was in a very small range (from −0.15 at 0 mW/cm2 to +0.10 K at 300 mW/cm2). If air cooling was applied, the differences revealed a much broader range (from −2.5 to +0.7 K).

image

Figure 9.  Sample temperature in dependence of wIRA irradiance using air- and water-cooling systems. In this figure, the efficiency of air (dotted line) and water cooling (dashed line) during wIRA irradiation is compared. In both cases, the temperature was directly measured at the center of the glass bottom of the same 35 mm glass bottom dish filled with a PBS layer of 2 mm. For air cooling, the dish was placed on a white plate. For water cooling, the setup displayed in Fig. 8 was used. Without wIRA exposure the measured dish temperatures resulted from both cooling and room temperature (298 K, 60% humidity; air ventilation did not show any effect on the experimental outcome, data not shown), as shown above. In both cases, a linear positive correlation of measured temperature and wIRA irradiance was detected, even though the maximal differences from the desired temperature of 310 K (horizontal continuous line) was significantly larger during air-cooling: −2.5 K without wIRA irradiation up to +0.7 K at the highest irradiance of 300 mW/cm2. The maximal differences under water cooling were −0.15 K (at 0 mW/cm2) and +0.1 K (at 300 mW/cm2), respectively. Each graph shows the averaged results from three different measurements.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This study clearly shows the importance of a careful choosing of the experimental setup if thermal and athermal effects of wIRA or other radiation that may induce warming of the exposed samples are investigated. As pointed out above, air cooling via a temperature-controlled plate, as often described in various publications, is not sufficient to provide a desired and constant temperature for an irradiated biological sample. One of the main technical problems in placing a Petri dish on a temperature-controlled plate are the so-called “stacking ring” at the bottom of the dish that prevents any direct contact between the bottom of the dish and the plate itself. Furthermore, a thin air layer of about 0.5 to 1.0 mm (depending on the dimensions of the stacking ring) is enclosed between dish and plate that serves as a very potent isolator, massively reducing heat transfer between dish and plate. But even if there were a direct contact between dish and plate, air filled microcaves at the contact area of both surfaces would reduce any heat exchange, too. At the same time, the resulting inhomogeneous heat transfer between plate and dish may result in patterns of different temperatures in the same sample.

A solution could be the air-free application of a liquid medium between both surfaces that fills these microcaves like heat-conductive paste that is used in electrotechnology. Another problem is the thermodynamic fact that the irradiated dish can be heated up by the applied irradiation, but it cannot be cooled by the plate beyond in the same way via irradiation. Most of the thermal energy induced by wIRA irradiation in a dish and the medium within will be dissipated by convection of the surrounding air. This explains at the same time the significant influence of different factors like air temperature, relative humidity and air ventilation in the room on the steady-state temperature of an irradiated dish as shown in Fig. 5. If the convection is disabled by the dishes’ lids during wIRA exposure, the medium temperature increases extremely even at low wIRA irradiances during air cooling.

Moreover, the use of common polystyrene Petri dishes is inappropriate, both in water and air cooling (see Figs. 3 and 4), because the heat conductance of this material is very low (0.17 W/(m·K)), causing significant temperature gradients between irradiated sample and cooling environment and thus an influence of thermal effects on the experimental outcome.

The presented approach with a water-cooled glass-bottom Petri dish is promising, because heat can be dissipated much better by a glass bottom of only 0.14 mm in thickness that has both a lower mass and a higher heat conductance (1.16 W/(m·K)) than a polystyrene bottom of about 1.0 mm. Moreover, the heat capacity of water is about four times higher than that of air, the thermal conductivity even about 22 times and the conduction of heat between a stirred cooling liquid and a hard surface is free of any inhomogeneities or microcavities, if air pockets are avoided.

However, even the described water-cooling setup is not able to keep irradiated samples at a desired temperature at arbitrary wIRA-irradiances. Our setup has been tested up to 300 mW/cm2; application of higher wIRA irradiances may actually result in significant hyperthermia/heat shock and the accordingly induced cellular responses. In addition, our setup was only used for wIRA exposure of single cell layers (12), but not larger tissue samples like whole skin samples with a volume of several cubic millimeter. In such a sample, that can be considered as a cluster of many thousands of cells, the ratio of surface to volume is considerably lower than that of single cells, resulting in a significantly reduced area that could dissipate heat from the sample.

In this study, we consider effects as “athermal” if the surrounding environment of the cells is maintaining the temperature during irradiation. This does not exclude the possibility of local temperature increase in microenvironments inside the cell. So, the causal mechanism of cellular change may not be strictly “athermal” even though the temperature in the surrounding is constant.

Therefore, the described experimental setup can be very useful in differentiation of thermal and athermal effects in investigation of singled cells at irradiances of up to 300 mW/cm2.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Acknowledgements—  We thank Dr. Eduard Wolf for his technical support and the “Dr. med. h.c. Erwin Braun Stiftung, Basel” for promotion of our theoretical and experimental work.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Schroeder, P., C. Pohl, C. Calles, C. Marks, S. Wild and J. Krutmann (2007) Cellular response to infrared radiation involves retrograde mitochondrial signaling. Free Radic. Biol. Med. 43, 128135.
  • 2
    Calles, C., M. Schneider, F. Macaluso, T. Benesova, J. Krutmann and P. Schroeder (2010) Infrared A radiation influences the skin fibroblast transcriptome: Mechanisms and consequences. J. Invest. Dermatol. 130, 15241536.
  • 3
    Piazena, H. and D. K. Kelleher (2010) Effects of infrared-A irradiation on skin: Discrepancies in published data highlight the need for an exact consideration of physical and photobiological laws and appropriate experimental settings. Photochem. Photobiol. 86, 687705.
  • 4
    Kim, M. S., Y. K. Kim, K. H. Cho and J. H. Chung (2006) Regulation of type I procollagen and MMP-1 expression after single or repeated exposure to infrared radiation in human skin. Mech. Ageing Dev. 127, 875882.
  • 5
    Schroeder, P., J. Lademann, M. E. Darvin, H. Stege, C. Marks, S. Bruhnke and J. Krutmann (2008) Infrared radiation-induced matrix metalloproteinase in human skin: Implications for protection. J. Invest. Dermatol. 128, 24912497.
  • 6
    Cho, S., M. J. Lee, M. S. Kim, S. Lee, Y. K. Kim, D. H. Lee, C. W. Lee, K. H. Cho and J. H. Chung (2008) Infrared plus visible light and heat from natural sunlight participate in the expression of MMPs and type I procollagen as well as infiltration of inflammatory cell in human skin in vivo. J. Dermatol. Sci. 50, 123133.
  • 7
    Buechner, N., P. Schroeder, S. Jakob, K. Kunze, T. Maresch, C. Calles, J. Krutmann and J. Haendeler (2008) Changes of MMP-1 and collagen type Ialpha1 by UVA, UVB and IRA are differentially regulated by Trx-1. Exp. Gerontol. 43, 633637.
  • 8
    Danno, K., N. Mori, K. Toda, T. Kobayashi and A. Utani (2001) Near-infrared irradiation stimulates cutaneous wound repair: Laboratory experiments on possible mechanisms. Photodermatol. Photoimmunol. Photomed. 17, 261265.
  • 9
    Kim, H. H., M. J. Lee, S. R. Lee, K. H. Kim, K. H. Cho, H. C. Eun and J. H. Chung (2005) Augmentation of UV-induced skin wrinkling by infrared irradiation in hairless mice. Mech. Ageing Dev. 126, 11701177.
  • 10
    Frank, S., L. Oliver, C. C. Lebreton-De, C. Moreau, M. T. Lecabellec, L. Michel, F. M. Vallette, L. Dubertret and B. Coulomb (2004) Infrared radiation affects the mitochondrial pathway of apoptosis in human fibroblasts. J. Invest. Dermatol. 123, 823831.
  • 11
    Kim, M. S., Y. K. Kim, K. H. Cho and J. H. Chung (2006) Infrared exposure induces an angiogenic switch in human skin that is partially mediated by heat. Br. J. Dermatol. 155, 11311138.
  • 12
    Jung, T., A. Hohn, H. Piazena and T. Grune (2010) Effects of water-filtered infrared A irradiation on human fibroblasts. Free Radic. Biol. Med. 48, 153160.
  • 13
    Darvin, M. E., A. Patzelt, M. Meinke, W. Sterry and J. Lademann (2009) Influence of two different IR radiators on the oxidative potential of the human skin. Laser Phys. Lett. 6, 229234.
  • 14
    Kim, M. S., Y. K. Kim, D. H. Lee, J. E. Seo, K. H. Cho, H. C. Eun and J. H. Chung (2009) Acute exposure of human skin to ultraviolet or infrared radiation or heat stimuli increases mast cell numbers and tryptase expression in human skin in vivo. Br. J. Dermatol. 160, 393402.
  • 15
    Frank, S., S. Menezes, C. C. Lebreton-De, M. Oster, L. Dubertret and B. Coulomb (2006) Infrared radiation induces the p53 signaling pathway: Role in infrared prevention of ultraviolet B toxicity. Exp. Dermatol. 15, 130137.
  • 16
    Applegate, L. A., C. Scaletta, R. Panizzon, E. Frenk, P. Hohlfeld and S. Schwarzkopf (2000) Induction of the putative protective protein ferritin by infrared radiation: Implications in skin repair. Int. J. Mol. Med. 5, 247251.