| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Pediatrics, Microbiology & Immunology and Radiology, Stanford School of Medicine, Stanford, California, USA;
* Picosecond Free Electron Laser Center, W. W. Hansen Experimental Physics Laboratory, Stanford University, Stanford, California, USA;
Xenogen Corporation, Alameda, California, USA; and
Department of Biomedical Engineering, Vanderbilt University, Nashville, Tennessee, USA
1 Correspondence: Department of Pediatrics, Stanford School of Medicine, 300 Pasteur Dr., Stanford, CA 94035-5208, USA. E-mail:ccontag{at}cmgm.stanford.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: laser–tissue interactions • luciferase • Hsp70 • thermal stress
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Investigation of transcriptional activation of the 70 kDa heat shock protein (hsp70 for the gene and Hsp70 for the protein) in response to elevated temperatures, including hyperthermal cell death, have been performed in cancer therapy (2
, 3)
and burn injury studies (4
, 5)
. In these studies, the response of cells to temperatures a few degrees above physiological levels (40°–47°C) for periods of a few minutes to a few hours was investigated. In burn-related studies, where higher temperatures and shorter durations may be more relevant, the response of cells exposed to temperatures as high as 68°C for 1 s was examined (6)
. hsp70 gene expression has been used in laser–tissue interaction models as a sensitive indicator of the response of cells to thermal stress after in vivo laser ablation in the liver. Fugitomi and co-workers (7)
demonstrated that elevated expression of Hsp70 24 h after thermal stress correlated with cell recovery; heat shock of tissue >43°C induced Hsp70 but thermotolerance was not observed.
In pulsed laser ablation under conditions of thermal confinement, extreme temperatures are achieved for very short periods. The explosively vaporized tissue reaches temperatures as high as 340°C. Although this material is ejected, surrounding tissue will be exposed to temperatures significantly above physiological temperatures for durations on the order of milliseconds. Elevated temperatures in biological tissue produced intentionally or as a side effect during laser treatment is important in determining the extent of tissue damage. To approximate the size of the damage zone after laser ablation and other tissue treatments, we developed a reporter gene strategy that would reveal spatial and temporal profiles of cellular injury and death. The promoter for a murine heat shock 70 (hsp70A1) gene was fused to the coding sequence of a modified firefly luciferase (luc) in order to provide a continuous indicator of the cellular response to heat shock. Since luciferase activity can be measured in living cells (8)
and tissues (9
, 10)
and can provide a readout for increases and decreases in transcription, it is an ideal real-time transcriptional reporter for indicating levels of hsp70 transcription. Using this reporter gene construct, we have shown that luciferase expression was induced in cells treated with a Holmium:YAG laser at a pulse energy of 65 mJ/pulse (total energy 1.95 J; total radiant exposure=6 J/cm2) and that cells died at 103 mJ/pulse (total energy 3.09 J; total radiant exposure=9.6 J/cm2) (11)
.
To determine whether additional changes in cell physiology could be identified between maximum induction of the hsp70 gene and cell death, we measured reporter gene responses to temperatures ranging from 42 to 64.6°C and assessed cell viability. Luciferase activity levels and cell viability measurements identified a transition zone where viable cells do not express the reporter. Cells were treated within this zone at 50, 50.5, 51.5, and 53.2°C for 30 s, then mRNA and protein levels (for Luc and Hsp70) as well as ATP levels were determined. These measurements indicated that reduced live cell luciferase activity was likely due to lower ATP and protein levels despite high levels of mRNA. Cells treated at these times/temperatures could fully recover their ability to express the reporter gene 30 h after the initial stress. The physiologic changes that affect reporter gene expression and cell viability under conditions of thermal stress define five zones that may be useful in describing collateral tissue damage after thermal ablation studies.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This Hsp70-luc cell line was subjected to different sources of thermal stress. To determine base line expression levels of the reporter gene, 
1 x 105 cells were grown in each well of 96-well dishes, sets of cells were treated at 42°C in an incubator for a duration of 20 min. In a second set of experiments, cells were removed from the cell culture dishes by treatment with 0.05% trypsin/EDTA, washed in phosphate buffered saline (PBS) and placed in 0.2 mL conical thin-walled thermocycler tubes (E and K Scientific, Campbell, CA, USA) at 1 x 105 cells/ tube in 100 µL of DMEM medium and heated in these tubes using a gradient thermocycler (PTC-200; MJ Research, Waltham, MA, USA). Cells were exposed to temperatures ranging from 50 to 64.6°C for 5–35 s.
Cells were removed from the conical thin-walled tubes and placed into 96-well cell culture plates and imaged as live cultures using a cooled CCD camera in a light tight box (IVIS, Xenogen Corp., Alameda, CA, USA). 30 µg/µL of Luciferin (1 µL) was added just before each image acquisition. Images were collected every 2 h for 10 h and then again at 24 h and analyzed using LivingImage software (Xenogen Corp.) as an overlay on the image analysis program, Igor Pro (Wavemetrics, Lake Oswego, OR, USA). Images were integrated over 2 min with pixel binding of 5 x 5. Cell viability was determined using flow cytometry and a Molecular Probes (Eugene, OR, USA) cell viability stain (see below).
The 15 and 30 s treatments at 50, 50.5, 51.5, and 53.2°C were chosen for further study at 4, 8, and 30 h post-treatment because cell viability studies revealed a transition at these temperatures, where reporter activity was not detectable but cells were still viable (90%); temperatures beyond that threshold led to reduced viability (50% or less). A luciferase assay was performed on cell lysates (Promega, E1500) 4 h after the 30 s thermal exposure and luminescence measured by an EG&G Berthold Lumat LB 9507 luminometer to determine whether functional luciferase was produced at temperatures beyond those that cells could produce light in vivo. We then performed an ATP assay (Roche ATP Bioluminescence Assay Kit CLS II) 4 h after the 30 s thermal exposure to determine whether the higher temperature treatments were associated with decreases in cellular ATP levels.
mRNA levels
Real-time PCR was performed on 20 ng of total RNA from the 30 s thermal treatments, prepared with the QuantiTectTM SYBR® Green RT-PCR kit to determine mRNA levels for the reporter and native gene at each temperature (4, 8, and 30 h after thermal stress). A PE Biosystems GeneAmp® 5700 Sequence Detector was used to amplify the RNA; data analyzed with GeneAmp® SDS Software (v1.3). mRNA levels for luciferase were quantified against a known luciferase standard and reported as number of molecules. hsp70 mRNA levels were determined and reported as fold induction, representing an increased level of induction relative to control. Data were transformed on a log scale and analyzed using SAS statistical software (SAS Inc., Cary, NC, USA) to calculate a correlation coefficient and P value.
Protein levels
Western blots analyses were performed on cell lysates collected 4 and 8 h after 30 s thermal treatments. Proteins were extracted using Pierce, M-Per Mammalian protein extraction kit; 25 µg of total protein was run on a 7.5% SDS-PAGE gel and electroblotted to nitrocellulose filters, split, and probed separately with anti-Hsp70 and anti luciferase antibodies (Stressgen, Victoria, BC, Canada) at ratios of 1:500 and 1:50, respectively. The Hsp70 filter was blocked in 5% and probed in 2% nonfat powdered milk; the luciferase filter was blocked in 10% and probed in 5% nonfat powdered milk. The appropriate secondary antibodies (Santa Cruz Biotechnology) were used at 1:2,500, incubated for 45 min, followed by an enhanced chemiluminescence detection system (Amersham). The filters were subsequently stripped and reprobed with a GAPDH loading control, blocked in 5%, and probed in 2% nonfat powdered milk with primary antibody concentration at 1:500 and the secondary at 1:2,500.
A luciferase standard was made by adding 500, 100, and 1 pg of luciferase protein to nontransformed NIH3T3 cells and loading 25 µg of these samples along with Hsp70 and luciferase samples. Hsp70 and luciferase treatments were compared with the standard as a relative measure of fold induction.
Flow cytometry
At 4, 8, and 30 h after thermal stress, medium was removed, then cells were resuspended in PBS and stained for 45 min for viability using Molecular Probes LIVE/DEAD viability and cytotoxicity stain (L-3224) containing calcein AM (which stains live cells green) and ethidium homodimer-1 (which stains membrane-compromised cells red). Cells were analyzed by flow cytometry and FlowJo© Analysis Software.
Pretreatment and response
Additional 51.5°C and 53.2°C 30 s treatments were subjected to a secondary stress 30 h post-initial treatment to determine whether these cells would recover the ability to produce light. These cells were treated at 50°C for 30 s and imaged 4 h post-secondary stress. In a second set of experiments, Hsp70 was induced by exposing cells to 42°C for 20 min 4 h before the 30 s thermal treatments. Message levels and viability measurements were taken 4 h after secondary thermal treatment and viability measurements at 30 h.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) and at slightly higher temperatures (45°C, 20 min) expression peaked after 6–8 h (data not shown). We evaluated reporter gene expression patterns after exposing cells to higher temperatures for shorter periods. A range of temperatures (50–64.6°C) and time (5–35 s) was evaluated using a heating block with a programmable thermal gradient of increasing thermal exposures (Fig. 2
). Luciferase activity revealed patterns of hsp70 induction that were dependent on temperature and length of exposure. We observed a range of responses that did not result in activation of the reporter gene (e.g., 50°C for 5, 10, and 15 s) to times/temperatures that apparently killed the cells (e.g., 61.8°C for 10, 15, 20, 25, 30, and 35 s). Temperature settings for 30 s treatments were confirmed using a thermistor (data not shown). In general, long exposure (e.g., 20 s) to the lower temperatures (e.g., 50°C) resulted in elevated expression that peaked at 4 h; shorter exposures (e.g., 5 s) at higher temperatures (e.g., 64.6°C) resulted in expression peaks at 6–8 h. Reporter gene expression appeared to define a zone along this thermal gradient that preceded cell death. To confirm this, we measured cell viability in cultures across the gradient of thermal exposures.
|
|
As expected, viability assays indicated that wells treated at the times/temperatures that activated the reporter gene and those below this stress level were 100% viable. However, at temperatures just beyond those where luciferase was induced, where we expected that the losses of signal were due to loss of viability, cells were nearly 100% viable (90% and greater viability); significant loss in viability occurred only at two temperature increments above the temperature where luciferase expression peaked (Fig. 3
A, B). For example, after a 30 s exposure to 50.5°C, luciferase expression peaked at 8 h with nearly 100% viability. In contrast, after 30 s exposure to the next higher temperature (51.5°C), luciferase activity was barely detectable later; however, there was only a slight reduction in viability. This pattern was consistent across all thermal exposure levels. These results suggested that hsp70-luc expression defines five physiologic zones (Fig. 2)
. The first zone is where thermal stress is not sufficient to induce expression; the second is where expression peaks at 4 h, the third where expression peaks at 8 h; the fourth is where there is no expression and yet cells are viable; and the fifth is where cells die. To further define the time/temperature window where cells are still viable but lose the ability to produce light (the fourth zone), we focused on 15 and 30 s treatments at temperatures around this window.
|
To determine whether loss of signal predicted that the cells would die later than those where luciferase expression was assayed, we assayed viability at 30 h in addition to 4 and 8 h. We saw no difference in viability between 4 and 8 h. At 30 h, cells in the 15 s treatment groups (50.5°C, 51.5°C, and 53.2°C) continued to proliferate, but cells in the 30 s at 53.2°C treatment group showed no growth relative to the 4 and 8 h time points (Fig. 3A
). Therefore, loss of bioluminescent signal at defined temperatures and times did not correlate with loss of viability and must be due to changes in cell physiology.
Since it is possible that loss of signal in the fourth zone was due to an inability to transcribe and translate the luc gene, we measured luciferase activity in cell lysates. Luciferase assays on cell lysates revealed that cells treated at 30 s have functional luciferase at 50–50.5°C, but there was a fivefold decrease in functional luciferase enzyme in cells treated at 51.5°C and no activity in cells treated at 53.2°C. A similar pattern was observed in the 15 s treatment group. Therefore, loss of signal intensity in live cell assays was in part due to a reduction in functional luciferase levels in the fourth zone. This fivefold reduction could not account for the total elimination of signal, so ATP levels were measured for these treatment groups. Results from the ATP assay indicated a significant reduction in ATP levels in cells treated at 51.5°C, where there were viable cells, and no luciferase activity as well as at 53.2°C where there was 50% loss in viability (Fig. 3C
). Cells treated at these higher temperatures were allowed to recover for 30 h, then cells in both groups showed a normal response to 50°C (30 s) thermal stress (data not shown). ATP levels were higher than normal levels in stressed cells that had strong bioluminescent signals (50 and 50.5°C for 30 s) (Fig. 3C
). We conclude that a combination of lower luciferase activity and reduced ATP levels can account for the reduced bioluminescent signals in the viable cells in the fourth zone.
To determine whether the reduction in luciferase levels in the fourth zone was due to inefficient transcription or translation and to validate the reporter construct Hsp70-luc as an indicator of hsp70 transcription, we measured mRNA for luc and hsp70. Real-time PCR measurements of luciferase mRNA levels paralleled those for hsp70 mRNA, but there were differences in magnitude (Fig. 4
). Measurement of mRNA levels in the 30 s treatment groups (50–53.2°C) revealed that luc and hsp70 mRNA levels at 4 h were highest in cells treated at 51.5°C (where luciferase activity was low); there was a further 10-fold increase in message levels at 8 h and a reduction to negligible levels at 30 h (Fig. 5
). Under conditions where 50% of the cells were killed by the thermal stress (30 s at 53.2°C), mRNA levels were low and similar to those with 50°C treatment. mRNA levels for luc and hsp70 dropped when there was a loss of viability.
|
|
Western blot analyses for Luc and Hsp70 protein indicated that each level approximately paralleled the luciferase activity levels in cell lysates. Neither protein was detectable in cells treated under conditions where there was cell death. At less intense stress conditions, the protein levels were higher than those observed in untreated controls (Fig. 6
) There were reduced levels of luciferase protein at temperatures of 51.5°C and higher, whereas Hsp70 protein did not diminish greatly from 50° to 50.5°C. Protein levels for Hsp70 and Luc were fivefold higher at 8 h than at 4 h.
|
To assess functional recovery of the cells after thermal stress, cells were pretreated at 42°C (for 20 min) 4 h before being subject to the various temperatures for 30 s. There were no differences in viability when measurements were made 30 h after the second thermal treatment, despite a fivefold increase in mRNA levels at 4 h for luciferase after 51.5°C treatment (Fig. 7
). At 30 h, this increase in message levels did not correlate with improved viability of the cells exposed to 53.2°C.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For the 30 s exposure, zone 4 existed in a narrow region between 50.5 and 51.5°C, where luciferase activity was not detectable but cells were still viable and could recover after 30 h. Despite extremely high levels of luc mRNA, luciferase protein levels were lower in zone 4 than in zone 3, and ATP levels showed a marked reduction in zone 4 (below normal levels). ATP levels were higher than the untreated controls in cells when stress conditions were below those of zone 4. The reduced levels of ATP and lower amounts of Luc protein most likely account for the results observed in zone 4. The relatively high mRNA levels and low protein levels in this zone indicate that transcription can proceed under thermal conditions where translation is inhibited. Given the complexity of translation relative to transcription (i.e., larger numbers of proteins, involvement of membranes and organelles and the need for macromolecular trafficking), it is not surprising that it is more thermally labile than transcription. Cells in zone 4 recover luciferase activity upon subsequent thermal stress if there is sufficient recovery time, which most likely restores ATP levels and the defects in translation.
Patterns of luciferase mRNA levels are highly correlated with hsp70 mRNA levels and match viability levels, making luciferase a good model for Hsp70 and thermal stress despite the loss of activity in the narrow temperature range that defines zone 4. The loss of activity in zone 4 may provide an important tool for determining the severity of thermal stress in laser–tissue ablation regimes by providing one more physiologic marker of stress. Patterns of Luc and Hsp70 protein levels were well correlated at 4 and 8 h for 50 and 50.5°C, although at 51.5°C luciferase protein levels were lower relative to Hsp70 at 4 h but higher than Hsp70 at 8 h.
These results correlated well with immunohistochemical measurements, where an inverse time/temperature relationship was observed, indicating a thermal tolerance at 60°C for 1 s and 55°C for 30 s (4)
. Our studies demonstrated that cells can withstand higher temperatures if the exposure times are reduced. In two separate studies, CO2 laser radiation was shown to be less damaging at lower fluences with repeated treatment as opposed to maximizing ablation volume at one time (12
, 13)
. Other studies have shown that stress events cause an increase in the amount of Hsp70 protein heat shock transcription factor (HSF1); therefore, cells that have been induced to express stress-related proteins may be more refractory, or able to produce Hsp70 quicker, than unstressed cells that have lower levels of Hsp70 (14)
.
This refractory pattern is reflected in our results as well: pretreated cells express luciferase earlier, likely due to the presence of transcription factors, which allow the cells to respond quicker and more intensely to stress (3-fold increase in mRNA levels in gradient treatments that had been pretreated relative to those that were not). It is also possible that there is leftover mRNA from the previous stress that can then be translated, but this would invoke a post-transcriptional mechanism of regulation. Nonetheless, a mechanism that regulates the response but allows the quicker secondary response must exist. Viability did not improve for the highest temperature treatment (53.2°C), possibly representing an upper threshold above which cells cannot recover despite overexpression of Hsp70. We suggest that at "more severe stress levels" (even though Hsp70 is produced), the lack of ATP will restrict the cellular response and Hsp70 expression may not protect them from dying.
The initial objective of this study was to use a Hsp70-luc reporter construct to define a zone of thermal induction for use in studies of laser–tissue interaction in cells and tissues. Thorough study of this process revealed that rather than three zones (where at lower temperature exposures the gene is not induced, an intermediate zone where hsp70-luc is activated, and a high exposure zone where cells die), there were actually five zones defined by this single reporter. The richness of this data set suggests that Hsp-70 marks several zones that could be used in thermal ablation models to identify the regions of tissue that underlie sites of thermal ablation where cell survival is likely. In future studies, this experimental design could also be used as an empirical validation of previous theoretical laser–tissue interaction models such as LATIS (15)
. Moreover, since the thermal induction of hsp70 has enabled targeted gene delivery, where the promoter for hsp70 has been used to direct expression of a therapeutic gene using focused ultrasound for local induction (16)
, defining the conditions that activate the promoter may enhance targeted gene delivery. The understanding gained by these studies will greatly accelerate the optimization of thermal ablation tools in biomedical research and medicine.
Results from this study, where several physiologic zones could be identified, suggest that the promoter for Hsp70 and other thermally induced genes may serve as useful markers of thermal stress and damage, especially when coupled to a short-lived reporter gene such as luciferase that can be assayed in living cells. Such strategies may provide more information about the effects of heat shock in various types of tissues than conventional assays, since activities can be measured in living tissues (9
, 10
, 17)
. An accurate measure of the cell response to thermal stress and cellular damage would enhance our understanding of thermal tolerance and serve as an indicator of wound healing. This in turn would lead to a more accurate design of various laser treatments of tissue in order to minimize collateral thermal damage.
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 19, 2003. Accepted for publication October 8, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. M. Houten, D. H. Volle, C. L. Cummins, D. J. Mangelsdorf, and J. Auwerx In Vivo Imaging of Farnesoid X Receptor Activity Reveals the Ileum as the Primary Bile Acid Signaling Tissue Mol. Endocrinol., June 1, 2007; 21(6): 1312 - 1323. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
