Ann. N.Y. Acad. Sci. 1066: 222–242 (2005). © 2005 New York Academy of Sciences.
Thermally Induced Injury and Heat-Shock
Protein Expression in Cells and Tissues
MARISSA NICHOLE RYLANDER,a YUSHENG FENG,b JON BASS,b AND
KENNETH R. DILLERa
Department of Biomedical Engineeringa and Institute for Computational Engineering and
Science,b The University of Texas at Austin, Austin, Texas 78712
ABSTRACT: Heat-shock proteins (HSPs) are critical components of a cell’s de-
fense mechanism against injury associated with adverse stresses. Initiating in-
sults, such as elevated or depressed temperature, diminished oxygen, and
pressure, increase HSP expression and can protect cells against subsequent,
otherwise lethal, insults. Although HSPs are very beneficial to the normal cell,
cancer cells can also use HSPs in response to stresses associated with various
therapies (hyperthermia, chemotherapy, radiation), mitigating injury incurred
by these treatments. Hyperthermia is a common treatment option for prostate
cancer. HSPs can be induced in regions of the tumor where temperatures are
insufficient to cause lethal thermal necrosis. Elevated HSP expression can en-
hance tumor cell viability and impart increased resistance to subsequent che-
motherapy and radiation treatments, thereby promoting tumor recurrence. An
understanding of the structure, function, and thermally stimulated HSP kinet-
ics and cell injury for prostate cancer cells is essential to designing effective
hyperthermia protocols. Measured thermally induced cellular HSP expression
and injury data can be employed to develop a treatment planning model for
optimization of the tissue response to therapy based on accurate prediction of
the HSP expression and cell damage distribution.
KEYWORDS: thermal injury; damage model; heat-shock protein expression;
Mammalian species have developed numerous mechanisms to cope with stress.
Examples at the cellular level include temporary modifications in gene expression to
survive changing environments, as well as altering cellular structure and function to
deal with more permanent adverse conditions. An important cellular alteration in-
duced by stress involves the synthesis and function of heat-shock proteins (HSPs).
These proteins reside in all organisms from bacteria and yeast to humans and exist
in various forms. At the cellular level, HSPs exist in the endoplasmic reticulum, mi-
tochondria, cytosol, and nucleus at low levels to respond to everyday stresses within
Address for correspondence: Kenneth R. Diller, Chairman, Department of Biomedical Engi-
neering, The University of Texas at Austin, 1 University Station, C0800, Austin, TX 78712-
1084. Voice: 512-471-7167; fax: 512-471-0616.
223RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
the cell. HSPs are characterized into families according to their molecular weight.
Typically, HSP78, -75, -60, and -10 are found in the organelles and HSP110, -90,
-73, -72, and -20 are present in the nucleus and cytosol. Each HSP has many docu-
mented functions and can reside in various locations within the cell (TABLE 1).1
TABLE 1. HSP locations and specific functions
FamilyChaperone members Cellular compartmentsFunctions
stabilize inactive forms
of certain hormone
receptors until hor-
mone is present; inter-
act with certain
protein kinases to
assist their transit to
prevent aggregation of
unfolded structures for
assembly of immuno-
globins; target aged
proteins to lysosomes
for degradation; pro-
tein secretion; antigen
with certain immuno-
stabilize prefoled struc-
tures for folding/
assembly; re-export of
precursors to mem-
activity with HSP70 to
enhance ATPase rate
and substrate release
tolerance through sta-
ble roles in cell
αΑ and αΒ
224ANNALS NEW YORK ACADEMY OF SCIENCES
There is substantial evidence that HSPs play important physiological roles in nor-
mal conditions and in situations involving both systemic and cellular stress.1 HSPs
were first discovered in 1962 as chromosomal puffs in heat-shocked (exposure to el-
evated temperatures) Drosophila salivary gland cells.2 Since their discovery, it was
observed that elevated HSPs can also be triggered by a variety of stressful stimuli
including ischemia, hypoxia, pressure overload, heavy metals, free oxygen radicals,
protein kinase C, calcium increasing agents, ethanol, amino acid and glucose ana-
logues, inflammation, sodium arsenite, hormones, antibiotics, cytokines, and infec-
tion (FIG. 1).3 Increased HSP expression can be stimulated as a result of normal
physiological processes such as development and differentiation. Specifically,
HSP70 and -27 have been shown to be induced by elevated temperature, ischemia,
oxidative stress, and anticancer drugs.4
Elevated concentrations of HSPs due to exposure to stress have been shown to
provide protection for both cultured cells and animal tissues. One of the first physi-
ological functions associated with the stress-induced accumulation of the inducible
HSP was acquired thermotolerance, which is defined as the ability of a cell or organ-
ism to become resistant to heat stress after a prior sublethal heat exposure. The phe-
nomenon of acquired thermotolerance is transient in nature and depends chiefly on
the severity of the initial heat stress. In general, the greater the initial heat dose, the
greater the magnitude and duration of thermotolerance. The expression of thermo-
tolerance following heat will occur within several hours, with maximum expression
generally occurring 16–18 h following the initial thermal insult and may last 3–5
days in duration.5
FIGURE 1. Stimuli that induce heat-shock protein expression.4
225RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
The precise mechanisms for enhanced cellular thermotolerance in association
with increased HSP levels have not been delineated. However, HSPs have been
shown to be involved in preventing protein denaturation and/or processing denatur-
ated proteins for an assisted refolding. Supporting evidence for this scenario came
first from a set of in vitro experiments by Mizzen and Welch, who demonstrated that
heat stress results in translational arrest within a cell. Subsequent resumption of
translation resulted in HSP mRNA being translated into HSPs before the synthesis
of other proteins took place within the cell.6 Other researchers have confirmed that
the protection HSP provides is based on their ability to act as molecular chaperones
to inhibit improper protein aggregation and their capacity to direct newly formed
proteins to target organelles for final packaging, degradation, or repair (FIG. 2).1
HSPs play an important role in self-repair, self-protection mechanisms, and are
associated with the refolding of denaturated cellular proteins.1
In addition to their chaperone functions, HSPs have numerous other protective
roles (FIG. 3). Specifically, HSP70 and -27 have been shown to inhibit apoptosis and
thereby increase the survival of cells exposed to a wide range of lethal stimuli.7
Overexpresion of HSP70 elevates nitric oxide production as a result of cytokine
stimulation. Nitric oxide serves to protect cultured cells from TNF-α–induced cell
death by inducing HSP70. Overexpression of HSP27 can protect microtubules and
actin cytoskeleton in cardiac myocytes and endothelial cells after exposure to
ischemia.7 HSP27, -60, and -70 are important in the progression of cancer both
through angiogenesis and their role in apoptosis.
FIGURE 2. Stress-induced chaperone functions of HSP.1
226ANNALS NEW YORK ACADEMY OF SCIENCES
The exact mechanism for stress-induced HSP induction has not been verified. A
proposed mechanism, however, for HSP70 induction will be discussed (FIG. 4).1,9
Transcription of the heat-shock response is controlled by a heat-shock factor (HSF).
Under normal conditions, HSFs are bound to HSPs and are inactive. This HSF is re-
sponsible for recognizing a target sequence known as the heat-shock element (HSE).
The HSE consists of a chain of five repeats of the NGAAN sequences. In the pres-
ence of stress, such as heat shock, the HSFs separate from the HSPs. Protein kinase
or other serine/threonine kinases phosphorylate the HSFs, which cause them to form
trimers in the cytosol. The HSF trimers enter the nucleus and bind to the HSE of the
HSP gene. Following binding, the HSFs are further phosphorylated and HSP mRNA
is transcribed and exits the nucleus for the destination of the cytoplasm. Upon enter-
ing the cytoplasm, new HSPs are synthesized. The HSFs then return to the cytoplasm
and bind to the HSPs in the original orientation before exposure to the stress. Mul-
tiple HSEs exist on the promoter region of the HSP gene.1,9 In addition to the HSE,
a serum response element resides on the HSP promoter region. The SRE responds to
serum stimulation and is responsible for the presence of a basal level of HSP expres-
sion in cells.10 Humans possess multiple copies of the HSP gene. A special feature
of HSP DNA coding is that it lacks introns. The HSP mRNA and HSPs can be pro-
duced extremely rapidly under stressful stimuli because no RNA splicing is required
for HSP mRNA transcription.9
Although HSP expression has a myriad of documented protective functions that
are beneficial to healthy cells, cancer cells can also use these proteins, preventing ef-
fective destruction of tumors with existing therapies. HSPs have been implicated in
protective roles in neoplastic tissues including multidrug resistance,11 regulation
of apoptosis,12 and modulation of p53 functions13 in a wide range of tumors. This
paper focuses on the thermally induced HSP protection pertinent to hyperthermia
prostate cancer therapies. Thermally induced HSP27 and -70 have been shown to en-
hance tumor cell viability by preventing apoptosis and imparting resistance to radi-
ation and chemotherapy following thermal therapy. HSP27 overexpression is a poor
prognostic marker for invasive prostatic carcinoma, but the absence of HSP27 is a
reliable objective marker in early prostatic neoplasia.14 Elevated HSP27 levels have
been associated with enhanced tumor cell viability by inhibition of apoptosis.
FIGURE 3. Activation of HSP by specific stimuli and their protective effect.8
227RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
HSP27 modulates reactive oxygen species by means of a glutathione-dependent
pathway,15 providing protection for intracellular proteins and partially explaining
the resistance it confers to thermally stressed tumor cells against chemotherapeutic
agents.16,17 Elevated levels of HSP70 have been observed in several types of tumors
including breast and cervical cancers18 and may be involved in cell proliferation,
prognosis, and drug resistance.19 Increased HSP70 expression has been linked to the
synergistic effect of hyperthermia on radio and chemotherapies.20
Previously characterized thermally induced HSP27 and -70 kinetics data will be
employed to design optimized laser therapies through computational predictive
models. Prior to discussing the experimental and computational methods involved in
optimizing laser therapies, some specific information will be provided on the molec-
ular structure and chaperone activity specific for HSP27 and -70.
HSP27 MOLECULAR STRUCTURE
HSP27 (also known as HSP25 and HSP28) is the smallest molecular mass protein
that is induced by heat shock. At present, there is limited knowledge about the struc-
ture of HSP27; however, it is known that this protein possesses two homotypic inter-
acting domains located at the N-terminal and C-terminal regions (FIG. 5). The C-
terminal domain extends from residue 88–183 where the last 10 residues form a flex-
FIGURE 4. Schematic of a proposed mechanism of stress-induced HSP gene
228ANNALS NEW YORK ACADEMY OF SCIENCES
ible peptide composed of A- and B-crystallins.21 Through the interaction of two
binding domains, HSP27 oligermerization is possible. One of the binding domains
occurs between residue 94 and 178 which corresponds to the α-crystallin domain. In
several studies this domain has been shown to have a role as the building block of
the quaternary structure of other members of this family and has been suggested as
the region where major intermolecular interactions occur. It has been suggested that
the N-terminal and C-terminal domains are essential in the stable formation of
dimers and high-molecular weight multimers.22 The overall structure of HSP27 con-
sists of a large central cavity where the hydrophobic N terminus is hidden to provide
it protection from the environment. The interactions of the N-terminal domain in the
central cavity act to stabilize the supramolecular organization of the dimers formed
by the α-crystallin domain.22
HSP70 MOLECULAR STRUCTURE
The molecular structure of the HSP70 family consists of three domains: 44 kDa,
18 kDa, and 10 kDa (FIG. 6).23 The 44-kDa domain consists of amino acid residues
1–386 at the N-terminus and contains four domains forming two lobes with a deep
cleft between. It is the 44-kDa domain that contains an ATPase domain. The 18-kDa
peptide-binding domain consists of amino acid residues 384–543 and is composed
of two four-stranded anti-parallel sheets and a single helix. This domain contains the
peptide-binding domain responsible for binding folded and unfolded peptides.23 The
10-kDa domain contains the amino acid residues 542–646 at the C terminus. This
segment of the protein is predominately a helix followed by a glycine/proline–rich
aperiodic segment next to a highly conserved EEVD terminal sequence.24
FIGURE 5. Molecular structure of the HSP2 kDa family.22
229RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
HSP27 AND 70 CHAPERONE MECHANISMS
The specific chaperone mechanism is unique for each HSP; therefore this paper
will only focus on those associated with HSP27 and -70. HSP27 chaperone function
is proposed to behave similarly to highly documented HSP70 protein folding
(FIG. 7).25 HSP70 almost immediately detects improperly folded proteins by recog-
nizing a small stretch of hydrophobic amino acids on a protein’s surface. Aided by
a set of smaller HSP40 proteins, an HSP70 monomer binds to its target protein and
then hydrolyzes a molecule of ATP to ADP, undergoing a conformational change
that causes the HSP70 to clamp tightly on the target. After HSP40 dissociates, the
dissociation of the HSP70 protein is induced by the rapid rebinding of ATP after
ADP release. Repeated cycles of HSP protein binding and release help the target pro-
tein to refold.
FIGURE 6. Molecular structure of HSP70 kDa family.23
FIGURE 7. Protein-folding mechanism for HSP70.25
230ANNALS NEW YORK ACADEMY OF SCIENCES
HSP EXPRESSION AND THERMAL INJURY
Although HSPs permit protection against lethal thermal injury, their induction
and related thermotolerance suggests that significant injury has already occurred.
The mechanisms contributing to thermal injury vs. thermotolerance are poorly un-
derstood. Cells, tissues, and animals show similar kinetics of thermotolerance,
which suggests that the morbidity and mortality associated with whole-body heating
is due in part to the dysfunction of some critical target tissues.26–29 It was proposed
that the development of thermotolerance results from the improved thermotolerance
of the weakest organ and cell systems.30 For instance, the small intestine is capable
of generating thermotolerance31 and is also reported to be the tissue most sensitive
to heat damage.32 In some instances, HSP expression could be used as a biomarker
of cellular injury.29 In this scenario, cells or tissues most at risk could be detected as
the most likely to accumulate HSPs during stress. HSP accumulation could mark a
tissue for potential failure.
A better understanding of the thermally induced HSP27 and -70 kinetics and cell
injury in prostate tumors would permit prediction and improvement of the therapy
outcome by maximizing tumor injury and eliminating HSP expression. Laser thera-
pies are a common treatment option for causing thermally induced prostate tumor
destruction. Applied thermal stress tends to elicit offsetting effects of elevated HSP
expression and hyperthermia-mediated cell injury. Insufficient thermal damage in-
duces elevated HSP27 and -70 expression that can enhance tumor cell viability, in-
creasing the possibility of tumor recurrence. Thermal therapies are generally
employed in conjunction with radiotherapy, chemotherapy, and gene therapy to in-
crease therapeutic efficacy. Elevated HSP expression in surviving tumor cells fol-
lowing hyperthermia imparts increased resistance to subsequent therapies,
substantially hindering their effectiveness. We have developed a strategy to predict
the thermally induced HSP27 and -70 kinetics and cell injury by using measured
prostate cancer cell HSP kinetics data33 to develop computational treatment plan-
In this paper, we review basic experimental and computational methods that per-
mit measurement and prediction of thermally induced cell injury and HSP kinetics.
We will introduce the mathematical concepts for predicting the temperature profile
in the tissue and methods for estimations of the thermally induced damage and HSP
expression level corresponding to a certain temperature profile. These mathematical
formulations will then be integrated into a finite element model to enable prediction
of the tissue response to laser therapy.
TEMPERATURE PROFILE FOR LASER-IRRADIATED TISSUE
A controlled temperature increase in the desired tissue region can be achieved
through laser irradiation of tissue. In order to provide an accurate prediction of the
tissue response following laser therapy, a computational model was created to pre-
dict the temperature, damage, and HSP27 and -70 distributions. The light and ther-
mal distributions during laser heating were modeled using adaptive finite element
methods which are capable of minimizing numerical error to specified precision.
This model allows specification of the thermal distribution during laser heating and
231RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
study of the sensitivity of the thermal behavior to manipulation of individual source
The mathematical representation of the temperature distribution is based on the
Pennes bio-heat transfer equation, which accurately predicts the spatial dependence
of the temperature history.35,36 This expression states that the rate of temperature
change in a small volume of tissue is equal to the sum of the heat transfer from the
surrounding tissues (conduction) and the heat transfer from blood perfusion as
shown in FIGURE 8.37 Usually, a term (qm) due to metabolic heat generated by the
cells is included in the model. However, such a contribution is negligible for thermal
injuries that occur over a short period of time.
Within the present approach we introduce a term to account for the rate of ab-
sorbed laser energy per unit volume distributed within the tissue, which is given by
Q(x, y, z) = µaφ(x, y, z) where µa (m−1) and(ΦW/m2) are the irradiation absorption
coefficient (wavelength dependent tissue property) and fluence (light distribution
generated due to the laser source) respectively.38,39 External irradiation will be used
as the method of inducing a temperature rise and associated thermal injury. The flu-
ence distribution is typically determined by employing Monte Carlo methods where
the probability of absorption or scattering of an incident photon on the tumor sphere
generated by the laser source can be predicted.40 Therefore, the temperature profile
is obtained by solving the equation
∇ k T
Q x y z
FIGURE 8. Bio-heat transfer equation defines the rate of temperature change in a small
volume of tissue.37
232ANNALS NEW YORK ACADEMY OF SCIENCES
where ρt and ct are the density and specific heat of the tissue, respectively. The spe-
cific heat of blood and the arterial blood temperature are represented as cb and Ta,
respectively. The temperature-dependent thermal conductivity of the tissue and
temperature-dependent blood perfusion rate are denoted by k and ωb, respectively.
The nonlinear temperature dependence of thermal conductivity and perfusion were
included in the model to increase the accuracy of the temperature prediction follow-
ing laser heating. The mathematical expressions employed for the nonlinear effects
of the temperature-dependent blood perfusion in the tumor are shown in the follow-
A complete set of data for the temperature-dependent behavior of thermal prop-
erties of tissue does not exist. The thermal variation properties of water are well
known in the range of 20–100°C and are important because the thermal properties
of tissue are dependent on the water content. The temperature-dependent thermal
conductivity for water where λk is a dimensionless correction factor and w is the wa-
ter content of prostate tissue, .511, is shown in the following equation.42
The degree of nonlinearity associated with the temperature dependencies of density
and specific heat for the tissue and blood was found to be minimal, and therefore was
not considered in the model. Sensitivity analysis showed that the temperature depen-
dence of thermal conductivity and perfusion caused a 5% decrease in predicted tis-
sue temperature for the laser irradiation protocols. Parameters employed in the
simulation were derived for canine prostate (TABLE 2).
K(T) = 4.19(0.133 + 1.36λkw)∗10−1
where λk = 1 + 1.78∗10−3(T − 20°C).
TABLE 2. Optical and thermal properties of canine prostate
Density of tissue39
Density of blood39
Specific heat of tissue39
Specific heat of blood39
Arterial blood temperature
233RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
The boundary conditions were applied on all tumor surfaces composing the
boundary ΓB ∪ ΓM ∪ ΓS = Γ, where Γ is the boundary of the tumor (FIG. 9). Surfaces
ΓB and ΓM were specified insulating boundary conditions since the tumor is sym-
metric about these boundaries. A convective boundary condition was imposed on the
curved outer surface to account for flow of air over the tumor. The mathematical
expression employed for calculating the convection coefficient needed for quiescent
air surrounding tissue surfaces with convective boundaries is shown below:
where NuD,a is the average Nusselt number, k is the thermal conductivity of the tissue
and D is the diameter of the tumor. The radius of the simulated tumor was 6 mm since
typicalprostrate tumors (derivied from PC3prostrate cancer cells) grown on mice attain
a maximum spherical volume of 1 cm3 prior to exhibiting necrotic cores. The NuDa can
be calculated according to the following expression for flow over a sphere:
1 0.59 Pr
FIGURE 9. Boundary conditions applied to surfaces of spherical tumor model.
234ANNALS NEW YORK ACADEMY OF SCIENCES
where Pr is the Prandtl number and RaD is the Rayleigh number as defined below:
where g is the local acceleration due to gravity (m/s), β is the volume coefficient of
air expansion (1/K) evaluated at the mean value of the surface and air temperatures,
ν is the kinematic viscosity (m2/s), and α is thermal diffusivity (m2/s). The values
for Ts and T∞ were stipulated as 310 K and 295 K, respectively. The parameters Pr,
ν, and α were evaluated for air at 295 K with values of .849, 1.55 × 10−5 m/s2, and
2.18 × 10−5 m2/s respectively.43
INJURY AND HSP EXPRESSION MODELS
The computational models were based on measured thermally induced cell injury
and HSP expression for normal (RWPE-1) and cancerous prostate (PC3) cells in-
duced by water-bath heating. In order to acquire this data, PC3 and RWPE-1 cells
were cultured in phenolic culture flasks. Upon reaching confluence, the flasks were
submerged in a constant temperature water bath at a predetermined temperature and
duration in the range of 44 to 60°C and 1–30 min.33 The maximum experimental
temperature caused complete cell death for the shortest heating duration. Following
heating, the flasks were returned to an incubator heated at 37°C for subsequent man-
ifestation of damage and HSP elevation.
HSP 27 AND HSP70 EXPRESSION MEASUREMENT AND MODEL
After an incubation time of 16–18 h post-heating (shown to be an effective eval-
uation period for measuring maximum HSP70 expression33), cells were lysed (cell
membranes are broken with enzymes to extract protein) in buffer solution containing
protease inhibitors and 10% SDS. The HSP 27 and -70 expression levels in each
supernatant solutions were analyzed via gel electrophoresis and Western blotting.33
The measured HSP27 and -70 kinetics data enabled formulation of an empirical
model for prediction of HSP expression as described in previous work.33 The pro-
posed model describes HSP expression as a function of only temperature and heating
duration which is adequately supported by our experimental data.
Employing Maple® permitted a wide array of functions to be explored for deter-
mination of the most appropriate mathematical formulation to accurately fit the en-
tire data set for all measured temperatures and heating durations. HSP expression
induced by a transient temperature field was found to obey the following relationship
∂H t T
f t T
,() H t T
235RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
where f(t, T) is a general rate function that can take various forms. In our case, we
select f(t, T) = (α − β1tγ−1), which captures the characteristics of HSP expression
denoted as H = H(T, t). The parameters α,β1(= β·γ), and γ are are independent of
time, but may be dependent on temperature, with γ > 1. H(t, T) was found to be rep-
resented by the following function:
where A is a temperature-dependent constant. Since the basal value of H(t, T) = 1 at
t = 0 due to normalization, A = 1 for the measured data set. A least squares approach
was employed for parameter estimation. The model is valid for predicting both
HSP27 and HSP70 expression, but unique expression parameters, α,β1(= β·γ), and
γ describe HSP27 and HSP70 expression. The HSP expression model was integrated
into the finite element model to enable prediction of the thermally-induced HSP
response due to laser heating.
CELL INJURY MEASUREMENT AND CELL DAMAGE MODEL
Cell viability was assessed following 72 h post-heating (shown to be an effective
evaluation period for measuring the extent of cell death44). Cells underwent propid-
ium-iodide staining. Propidium iodide is permeable to only dead or dying cells, en-
abling injured cell populations to appear fluorescent. A flow cytometer was utilized
to detect the fluorescence levels for all thermally stressed cell samples to permit
quantification of cell viability.33
The availability of both measured cell injury and thermal history data enabled de-
termination of the constitutive parameter values for an Arrhenius damage model:45
where Ω is damage, defined as the logarithm of the ratio of the initial concentration
of healthy cells, Co, to the concentration of healthy cells remaining after thermal
stimulation, Cτ, for a stimulation duration of τ (s). A(1/s) is a scaling factor, Ea
(Jmol−1) is injury process activation energy, ℜ(Jmol−1K−1) is the universal gas con-
stant, and T(K) is instantaneous absolute temperature of the cells during stress,
which is a function of time, t(s). The cell viability values for PC3 cells 72 h following
hyperthermia were used to determine the damage parameters employed in the Arrhe-
nius damage model.33 The Arrhenius damage integral was fit to the cell injury data
to characterize the response to thermal stress temperature and duration. At each tem-
perature the threshold time (τ) was determined for Ω = 1 for which Cτ = 1/e of Co.
For isothermal stress conditions and when Ω = 1, the damage equation simplifies to
the logarithmic form,
The thermal damage kinetic coefficients of A and Ea were determined from the inter-
cept and slope respectively of the best-fit linear function to fit the experimental data.33
H t T
Ω τ ( )
EaℜT t ( )⁄()dt
τ ( )
236ANNALS NEW YORK ACADEMY OF SCIENCES
According to the Arrhenius formulation, increasing temperature and heating du-
ration will cause the damage value to increase indefinitely with complete cell dam-
age represented by infinity. In order to more meaningfully represent the damage,
another parameter was employed for determining tissue injury in the finite element
simulations. The damage fraction, FD of damaged to undamaged tissue is given by
Native tissue is represented by FD = 0 (Ω = 0) and fully denatured tissue is given
by FD = 1 (Ω = infinity).42
THERMAL, HSP, AND DAMAGE PREDICTIVE FINITE
Although HSP expression and damage were induced experimentally through
water-bath heating, the ultimate goal is development of laser protocols to minimize
HSP expression and maximize damage in the tumor region. Ultimately, HSP expres-
sion and damage data from laser-irradiated tissue will be employed to refine the ac-
curacy of the model. Current cellular data will demonstrate the power of the model
for predicting and optimizing HSP expression. Optimal heating protocols were iden-
tified by determining the most advantageous energy deposition pattern of a laser
source necessary to produce a temperature distribution resulting in the desired HSP
expression and tissue damage pattern. This involved determination of the appropri-
ate laser source parameters consisting of wavelength, number of probes, optical fiber
orientation in the tissue, power, and thermal exposure time. In the following simula-
tions, only the laser power and thermal exposure time will be varied when investi-
gating protocols. Determination of the desired temperature distribution was based on
optimal heating protocols (time/temperature) determined through cellular HSP ex-
pression and damage kinetic models. The defined energy dissipation parameters will
ultimately be applied to the laser source to produce destruction of the tumor while
minimizing HSP expression in the targeted regions.
All simulations model external irradiation incident on the tumor surface and em-
ploy a laser wavelength of 810 nm. The parameters employed for the first simulation
are laser power of 2 W and pulse duration of 1 min. The laser source was applied
externally because our current laser heating experiments in vivo initially employ ex-
ternal radiation in order to prevent HSP expression induced by insertion of the laser
probe. The model can accurately predict the temperature, damage fraction, and HSP
expression in the tumor for the given specified laser parameters (FIG. 10).
Significant elevations in temperature occurred at the tumor boundary in close
proximity with the laser source, but a substantial portion of the tumor experienced
minimal or no elevation in temperature. Nominal damage was induced throughout
237RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
the majority of the tumor volume, with the greatest injury occurring near the laser
probe and diminishing with increasing distance from the laser source. All HSP ex-
pression was normalized with respect to the basal level of HSP expression for un-
stressed conditions and is represented as 1 mg/mL. As a result, HSP expression
higher than 1 mg/mL represents an elevation in expression. There are three observ-
able zones of interest with regard to HSP expression in the tumor. The topmost blue
region represents a zone where significant damage was induced causing denaturation
of all proteins and rendering the cell machinery incapable of HSP expression. The
middle region with HSP expression greater than one represents the thermal regime
where temperatures caused considerable induction of HSP expression. This is the re-
gion of concern where HSP expression must be minimized to prevent tumor cell vi-
ability following treatment. The final bottommost region represents a zone where
temperatures were insufficient to elicit significant HSP expression with expression
levels at or near the basal level.
In order to evaluate the success of the therapy, it is essential to clearly define char-
acteristics of an optimal thermal treatment. The objective for designing a laser ther-
apy should be complete tumor destruction and preservation of healthy surrounding
tissue formulated according to the following mathematical criteria:
FIGURE 10. Finite element model prediction of (a) temperature (K), (b) damage, (c)
HSP27 (mg/mL), and (d) HSP70 (mg/mL) distribution for externally irradiated prostate tu-
mor with a laser power of 2W and pulse duration of 1 min.
238 ANNALS NEW YORK ACADEMY OF SCIENCES
(1) Complete tumor destruction: Tumor must experience maximum cell injury
(e.g., FD ≥ .99), and HSP expression must be diminished below its basal level
of 1 (HSP27,70 ≤ 1 mg/mL).
(2) Preservation of healthy tissue: Healthy tissue must receive minimal thermal
damage (e.g., FD ≤ .01) and induction of increased levels of HSP expression
(HSP27,70 > 1 mg/mL) to mitigate injury to subsequent thermal or chemother-
aphy and radiation treatments.
The simulations in this paper consider only a spherical solid tumor volume so
only criterion 1 is applicable. Evaluating this therapy outcome according to criterion
1 would permit us to realize that this is an unsatisfactory treatment due to insufficient
thermal damage incurred throughout the entire tumor and the presence of a large re-
gion of elevated HSP27 and -70 expression. The elevated level of HSP expression
would enhance the tumor cell viability, increasing the possibility of tumor recur-
rence and imparting tumor resistance to chemotherapy and radiation.
In order to achieve a more desirable treatment outcome, another therapy was con-
sidered in which the laser power was increased to 4 W and the pulse duration was
lengthened to 2 min (FIG. 11). This therapy incurred more substantial temperature
elevations and extensive thermal injury. Dramatic tissue injury extended to a greater
tumor depth corresponding to a larger zone of protein denaturation. The topmost
dark blue zone representing a region experiencing extreme thermal damage without
FIGURE 11. Finite element model prediction of (a) temperature (K), (b) damage, (c)
HSP27 (mg/mL), and (d) HSP70 (mg/mL) distribution for externally irradiated prostate tu-
mor with a laser power of 4W and pulse duration of 2 min.
239RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
elevation in HSP expression still exists, but has enlarged in size. Below this dena-
tured zone is a region of high HSP expression. This region has been shifted in depth
and narrowed in size, but still represents a zone of concern for tumor regrowth. With
increasing depth, a zone with minimal HSP expression is observed due to insuffi-
cient temperature elevation.This therapy is still considered unacceptable due to the
region of HSP expression since it does not adequately satisfy criterion 1.
The simulated cases have not yet identified an ideal therapy where HSP expres-
sion has been eliminated. The laser power is increased further to 5 W and the pulse
duration is lengthened to 3 min. If we achieve HSP expression reduction below its
unstressed basal level, the damage will have been so extensive causing complete tu-
mor destruction. As a result, HSP27 and -70 elimination is the strictest criterion, and
they will be the only quantities simulated. The HSP27 and -70 expression distribu-
tions were identical and therefore were represented by a single figure (FIG. 12). The
outcome of this therapy yields a tumor with not only a lack of HSP expression in-
crease, but also extreme denaturation of all proteins such that HSP expression is
nearly nonexistent. The complete elimination of HSP expression yields a tumor with
complete cell death so the cell damage plot is not shown. The tumor will not expe-
rience re-growth or thermally induced resistance to subsequent chemotherapy or ra-
diation. This therapy would be considered a success theoretically; however, we have
neglected the existence of surrounding healthy tissue that would have been damaged
due to elevated temperatures utilized to achieve this complete kill.
FIGURE 12. Finite element model prediction of HSP27 (mg/mL) and HSP70 (mg/mL)
distribution for externally irradiated prostate tumor with a laser power of 5W and pulse
duration of 3min.
240ANNALS NEW YORK ACADEMY OF SCIENCES
An accurate HSP expression model was developed based on measured cellular
data. This is the first study utilizing HSP expression computational models for pre-
diction of HSP expression for use in thermal therapy planning. The Arrhenius injury
model enabled the damage in the tumor to be predicted with precision based on a
given temperature profile in the tumor. Previous studies have demonstrated a much
lower thermal threshold for destruction of AT-1 canine prostate tumors in vivo com-
pared to their in vitro counterparts under similar conditions due to the presence of
the vascular network in vivo.46,47 Thus, it will be essential to investigate the hyper-
thermia-induced HSP expression kinetics and cell viability modifications for PC3
tumors in vivo before final dosimetry guidelines are developed for prostate cancer
The integration of thermal, damage, and HSP expression models into a single fi-
nite element model enabled prediction of the prostate tumor response to a given laser
therapy. After considering a wide range of laser parameters commonly employed in
surgical procedures in the finite element modeling simulations, definite zones of
HSP expression were identified in the irradiated prostate tumors after therapy. The
existence of these zones is of great concern, and minimization of HSP expression
should be considered in patient treatment design. In designing future therapies there
will be a tradeoff between complete tumor eradication and damage minimization to
healthy tissue. Further exploration into the role HSP expression in tumor survival is
essential in designing successful therapies. The model developed in this study will
enable the damage and HSP expression to be predicted for a wide range of laser pa-
rameters of the physician’s choice prior to surgery to improve understanding of the
tissue response and to enable optimized design of the patient therapy.
1. KIANG, J.G. & G.C. TSOKOS. 1998. Heat shock protein 70kDa: molecular biology, bio-
chemistry, and physiology. Pharmacol. Ther. 80: 183–201.
2. RITOSSA, F. 1962. A new puffing pattern induced by temperature shock and DNP in
Drosophila. Experientia 18: 571–573.
3. WEICH, H. et al. 1992. HSP90 chaperones protein folding in vitro. Nature 358: 169–
4. SCHLESINGER, M. et al. 1990. Stress Proteins: Induction and Function. Springer-Verlag.
5. WANG, S., K.R. DILLER & S. AGGARWAL. 2003. Heat shock protein 70 expression
kinetics. J. Biomech. Eng. 125: 794–797.
6. MIZZEN, L.A. & W.J. WELCH. 1988. Characterization of the thermotolerant cell. I.
Effects on protein synthesis activity and the regulation of heat-shock protein 70
expression. J. Cell Biol. 106: 1105–1116.
7. GARRIDO, C. 2001. Heat shock proteins: endogenous modulators of apoptotic cell
death. Biochem. Biophys. Res. Commun. 286: 433–442.
8. LATCHMAN, D. 2001. Stress proteins. Cardiovasc. Res. 51: 637–646.
9. MORIMOTO, R.I. & M.G. SANTORO. 1998. Stress-inducible responses and heat shock pro-
teins: new pharmacologic targets for cytoprotection. Nature Biotechnol. 16: 833–838.
10. WU, B.J. & R.I. MORIMOTO. 1985. Transcription of the human HsP70 gene is induced
by serum stimulation. Proc. Natl. Acad. Sci. USA 82: 6070–6074.
11. CIOCCA, D.R. et al. 1993. Heat shock protein hsp70 in patients with axillary lymph node-
negative breast cancer: prognostic implications. J. Natl. Cancer Inst. 85: 570–574.
241RYLANDER et al.: THERMALLY INDUCED INJURY & HSP EXPRESSION
12. TOMEI, L.D. & F.O. COPE. 1991. Apoptosis: The Molecular Basis of Cell Death. Cold
Spring Harbor Laboratory Press. Cold Spring Harbor, NY.
13. LEVINE, A.J. et al. 1991. The p53 tumor suppressor gene. Nature 351: 453–456.
14. CORNFORD, P.A. et al. 2000. Heat shock protein expression independently predicts
clinical outcome in prostate cancer. Cancer Res. 60: 7099–7105.
15. MEHLEN, P. et al. 1996. Human hsp27, Drosophila hsp27, and human αβ-crystallin
expression-mediated increase in gluthathione is essential for the protective activity
of these proteins against TNF-α-induced cell death. EMBO J. 15: 2695–2706.
16. OESTERREICH, S. et al. 1993. The small heat shock protein hsp27 is correlated with
growth and drug resistance in human breast cancer cell lines. Cancer Res. 53: 4442–
17. RICHARDS, E.H. et al. 1996. Effects of overexpression of the small heat shock protein
HSP27 on the heat and drug sensitivities of human testis tumor cells. Cancer Res. 56:
18. GEORGOPOULOUS, C. & W.J. WELCH. 1993. Role of the major heat shock proteins as
molecular chaperones. Annu. Rev. Cell Biol. 9: 601–634.
19. CRAIG, E.A. et al. 1994. Heat shock proteins and molecular chaperones: mediators of
protein conformation and turnover in the cell. Cell 78: 365–372.
20. VARGUS-ROIG, L.M. et al. 1997. Heat shock proteins and cell proliferation in human
breast cancer biopsy samples. Cancer Detect. Prev. 21: 441–451.
21. HALEY, D. et al. 1998. The small heat shock protein, αβ-crystallin, has a variable qua-
ternary structure. J. Mol. Biol. 277: 27–35.
22. LAMBERT, H. et al. 1999. HSP27 multimerization mediated by phosphorylation-
sensitive intermolecular interactions at the amino terminus. J. Biol. Chem. 274:
23. MORSHAUSER, R.C. et al. 1995 The peptide-binding domain of the chaperone protein
HSC70 has an unusual secondary structure topology. Biochemistry 34: 6261–6266.
24. HIGHTOWER, L.E. et al. 1994. Interaction of vertebrate HSC70 and HSP70 with
unfolded proteins and peptides in: the biology of heat shock proteins and molecular
chaperones. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY.
25. ALBERTS, B. et al. 1994. Molecular Biology of the Cell (3rd ed.). Garland Publishing.
26. URANO, M. 1986. Kinetics of thermotolerance in normal and tumor tissues: a review.
Cancer Res. 46: 474–482.
27. WESHLER, Z. et al. 1984. Development and decay of systemic thermotolerance in rats.
Cancer Res. 44: 1347–1351.
28. MOSELEY, P.L. 1997. Heat shock proteins and heat adaptation of the whole organism. J.
Appl. Physiol. 83: 1413–1417.
29. HALL, D.M. 2000. Caloric restriction improves thermotolerance and reduces
hyperthermia-induced cellular damage in old rats. FASEB J. 14: 78–86.
30. KREGEL, K.C. 2002. Heat shock proteins: modifying factors in physiological stress
responses and acquired thermotolerance. J. Appl. Physiol. 92: 2177–2186.
31. HUME, S.P. & J.C.L. MARIGOLD. 1980. Transient, heat-induced thermal resistance in
the small intestine of mouse. Radiat. Res. 82: 526–535.
32. HENSCHEL, A. et al. 1969. An analysis of the deaths in St. Louis during July 1966. Am.
J. Public Health 59: 2232–2240.
33. RYLANDER, M.N. et al. HSP27, 60, and 70 expression kinetics and cell viability in nor-
mal and cancerous prostate cells. Under review.
34. RYLANDER, M.N. et al. 2005. Optimizing HSP expression in prostate cancer laser ther-
apy through predictive computational models. J. Biomed. Optics. Under review.
35. PENNES, H.H. 1948. Analysis of tissue and arterial blood temperatures in the resting
forearm. J. Appl. Physiol. 1: 93–122.
36. WISSLER, E.H. 1998. Pennes 1948. Paper revisited. J. Appl. Physiol. 85: 35–41.
37. LEE, R. & W. DOUGHERTY. 2003. Electrical injury: mechanisms, manifestations, and
therapy. IEEE 10: 810–819.
38. STAR, W. 1995. Diffusion theory of light transport. In Optical-Thermal Response of
Laser-irradiated Tissue. A.J. Welch & M. Gemert, Eds.: 166–169. Plenum Press.
242ANNALS NEW YORK ACADEMY OF SCIENCES Download full-text
39. KIM, B. et al. 1996. Nonlinear finite-element analysis of the role of dynamic changes
in blood perfusion and optical properties in laser coagulation of tissue. IEEE J.
Selected Topics Quantum Electron. 2: 922–933.
40. JACQUES, S. & L. WANG. 1995. Monte Carlo modeling of light transport in tissues. In
Optical-Thermal Response of Laser-irradiated Tissue. Plenum Press. New York.
41. LANG, B. et al. 1999. Impact of nonlinear heat transfer on temperature control in
regional hyperthermia. IEEE Trans. Biomed. Eng. 46: 1129–1138.
42. ZHU, D. et al. 2002. Kinetic thermal response and damage in laser coagulation of tis-
sue. Lasers Surg. Med. 31: 313–321.
43. INCROPERA, F. & D. DEWITT. 1996. Fundamentals of Heat and Mass Transfer. John
Wiley and Sons. New York.
44. RYLANDER, M.N. et al. 2005. Correlation of HSP 70 expression and cell viability fol-
lowing thermal stimulation of bovine aortic endothelial cells. J. Biomech. Eng. 127:
45. HENRIQUES, F.C. 1947. Studies of thermal injury V. The predictability and the signifi-
cance of thermally induced rate processes leading to irreversible epidermal injury.
Arch. Pathol. 43: 489–502.
46. BHOWMICK, S. & J.C. BISCHOF. 1998. Supraphysiological thermal injury in Dunning
AT-1 prostate tumor cells. J. Biomech. Eng. 122: 51–59.
47. BHOWMICK, S. et al. 2004. In vitro thermal therapy of AT-1 Dunning prostate tumors.
Int. J. Hyperthermia 20: 73–92.