| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biophys J, April 2000, p. 2116-2126, Vol. 78, No. 4
and
*Section of Molecular and Cellular Biology, University of
California, Davis, Davis, California, Department of
Chemistry, State University of New York, Buffalo, New York,
Department of Anatomy, Physiology, and Cell Biology,
School of Veterinary Medicine, University of California, Davis, Davis,
California, USA
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
REFERENCES
|
|---|
The effect of temperature on the binding equilibria of calcium-sensing dyes has been extensively studied, but there are also important temperature-related changes in the photophysics of the dyes that have been largely ignored. We conducted a systematic study of thermal effects on five calcium-sensing dyes under calcium-saturated and calcium-free conditions. Quin-2, chlortetracycline, calcium green dextran, Indo-1, and Fura-2 all show temperature-dependent effects on fluorescence in all or part of the range tested (5-40°C). Specifically, the intensity of the single-wavelength dyes increased at low temperature. The ratiometric dyes, because of variable effects at the two wavelengths, showed, in general, a reduction in the fluorescence ratio as temperature decreased. Changes in viscosity, pH, oxygen quenching, or fluorescence maxima could not fully explain the effects of temperature on fluorescence. The excited-state lifetimes of the dyes were determined, in both the presence and absence of calcium, using multifrequency phase-modulation fluorimetry. In most cases, low temperature led to prolonged fluorescence lifetimes. The increase in lifetimes at reduced temperature is probably largely responsible for the effects of temperature on the physical properties of the calcium-sensing dyes. Clearly, these temperature effects can influence reported calcium concentrations and must therefore be taken into consideration during any investigation involving variable temperatures.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
REFERENCES
|
|---|
Fluorescent analogues of calcium chelators have
proven extremely useful in measuring the intracellular calcium
concentration of various types of living cells. Fluorescent probes that
have been widely used include Indo-1 (June and Rabinovitch, 1994
;
Sipido and Callewaert, 1995; Jacquemond and Allard, 1998; Sebille et al., 1998; Ju and Allen, 1998; Grynkiewicz et al., 1985), Fura-2 (Xu et
al., 1997; Sipido and Callewaert, 1995; Chen et al., 1996; Zicha et
al., 1996; Greco et al., 1996; Bhattacharya and Chakrabarti, 1998; Miao
et al., 1998; Grynkiewicz et al., 1985), Quin-2 (Tsien and Pozzan,
1989; Deber and Hsu, 1986; Jacob et al., 1987; Nijayaraghavan and
Hoskins, 1989), calcium green dextran (Zimprich and Bolsover, 1996;
Harris, 1994; McClellan et al., 1994; Tamm and Terasaki, 1994; Kono et
al., 1996; Bi et al., 1995; Carroll et al., 1994; Fetcho and O'Malley,
1995), and chlortetracycline (Tao et al., 1992; Jy and Haynes, 1984,
1987, 1988). Most experiments of this type are performed at constant
temperature (e.g., 37°C). However, if temperature is varied during
the investigation, the effect of that temperature change on the
fluorescent probe itself must be considered during data interpretation.
One important effect of temperature on the fluorescent calcium-sensing
dyes is a variation in the binding characteristics of the chelator. As
has been documented for Indo-1, Fura-2 (Shuttleworth and Thompson,
1991; Howarth et al., 1995), Fluo-3 (Lattanzio, 1990), and the calcium
chelators EGTA and BAPTA (Harrison and Bers, 1987), the binding
equilibria are shifted such that the dissociation constant
(Kd) increases at low temperature.
Fortunately, this type of error is easily avoided by using the
appropriate temperature-corrected Kd
for a given dye (Shuttleworth and Thompson, 1991). The calcium
affinities of chelators under various conditions can be measured using
Scatchard plot analysis, or calculated using the van't Hoff isochore
(Harrison and Bers, 1987, 1989). Further, computer programs, such as
MaxChelator (Bers et al., 1994), can be used to estimate
Kd at various values of temperature
and ionic strength. Because this type of error has been thoroughly
investigated, it will not be considered in the current study.
A second important consideration when temperature is varied during the
course of an experiment is the fact that the fluorescence itself is
affected by temperature, regardless of binding characteristics. In
general, quantum yield (Eastman and Rosa, 1968; Song et al., 1975;
Cornelissen-Gude and Rettig, 1998; Haynes et al., 1993), and thus
fluorescence intensity (Connors et al., 1998; Park, 1996; Law, 1994)
both increase as temperature decreases, and, in polar solvents, the
effect is even more pronounced (Waris et al., 1988; Bark and Force,
1993). In addition, excited-state lifetimes of fluorescent molecules
are strongly affected by temperature, showing, in most cases, longer
lifetimes at lower temperatures (Drain et al., 1998; Cornelissen-Gude
and Rettig, 1998; Kumke et al., 1997; Young et al., 1997; van den Zegel
et al., 1984). This effect of increasing fluorescence lifetimes is
largely due to a decrease in the rates of nonradiative decay at low
temperature (Lee and Robinson, 1984; Lam et al., 1998; Young et al.,
1997; Giri, 1992). Exceptions to the above trends have been noted,
however. For instance, certain tryptophan dipeptides show an increase
in quantum yield and intensity at elevated temperatures, due to an
equilibrium shift toward an unprotonated species with a larger quantum
yield (Brancaleon et al., 1995, 1997). Thus, temperature effects on fluorescence often cannot be predicted, and must be measured.
These two distinct consequences of temperature on the calcium-sensitive
dyes (i.e., effects on binding characteristics and effects on
fluorescence) are both crucial to the accurate and reliable use of
these probes for measuring free calcium concentrations, yet frequently,
neither are considered. Some groups avoid the problem of the
Kd effects by reporting the 
1/2
ratio instead of calcium concentration (Kenyon and Goff, 1998; Gambassi
et al., 1994). This is the ratio of fluorescence intensities
(I) for calcium-bound to calcium-free fluorophore
(I+Ca/I
Ca).
This method, however, still does not control for the effects of
temperature on the fluorescence intensity at those two wavelengths. In
fact, although consideration of the temperature effects on
Kd is fairly common (Puglisi et al.,
1996; Liu et al., 1991; Shuttleworth and Thompson, 1991), our
literature search revealed only one paper in which the effects of
temperature on fluorescence were considered (Zhao and Buhr, 1995).
In the current study, thermal effects on five calcium-sensitive dyes
were studied in the range of 5 to 40°C. All the dyes tested were
affected by temperature to some extent. Because fluorescence can be
strongly affected by certain physical characteristics of the system,
including pH (Martin and Jain, 1994; Ritucci et al., 1996; Rink et al.,
1982b; Szmacinski and Lakowicz, 1993), viscosity (Swaminathan et al.,
1997; Somogyi et al., 1994; Luby-Phelps et al., 1993; Busa, 1992), and
the presence of oxygen (Damdinsuren et al., 1995; Danielsen et al.,
1995; Lakowicz, 1983), each of these, as well as the influence of
temperature on the excited-state lifetimes, was studied as a possible
cause for the fluorescence changes with temperature. Regardless of the
causal mechanism, however, this study shows that the effects of
temperature on fluorescence, as well as on the binding characteristics
of the chelators, must be considered to obtain meaningful data with
calcium-sensitive fluorescent dyes when temperature is varied during
the course of the investigation.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
REFERENCES
|
|---|
Measuring fluorescence
Fura-2, Indo-1, Quin-2, and calcium green dextran were obtained from Molecular Probes, Inc. (Eugene, OR), and chlortetracycline (CTC) was obtained from CalBiochem (La Jolla, CA). The fluorescence intensities of the probes were measured at the following concentrations: 4 µM Fura-2 pentapotassium salt; 1 µM Indo-1 pentapotassium salt; 37 µM Quin-2; 75 µM CTC HCl; 0.5 µM calcium green-1 dextran potassium salt (MW 10,000). Probes were dissolved in phosphate buffered saline (PBS) (9.4 mM Na2HPO4, 0.6 mM KH2PO4, 100 mM NaCl, pH 7.2) at 20°C. Samples measured in the absence of calcium also contained 10 mM ethylene glycol-bis-(b-aminoethyl ether) N,N,N',N'-tetraacetic acid (EGTA), and samples containing calcium were adjusted to 1 mM CaCl2 with Tyrodes wash (148 mM NaCl, 2.5 mM KCl, 23 µM MgCl2, 5 mM glucose) containing 10 mM CaCl2.
Static fluorescence was measured with a Hitachi F-2000 fluorescence spectrometer controlled by a PC equipped with the F-2000 IC cation-measurement software (Hitachi, San Jose, CA). The excitation and emission values used were as follows (in nm): Quin-2: Ex 340, Em 490; CTC: Ex 390, Em 530; calcium green: Ex 488, Em 530; Indo-1: Ex 350, Em1 405, Em2 480; Fura-2: Ex1 340, Ex2 380, Em 510. Samples of 770 µL were loaded into 1.0-mL methacrylate UV/vis cuvettes (Fisher Scientific, Pittsburgh, PA) and stirred using a small magnetic stir bar and built-in stir plate. Excitation and emission maxima were obtained using the wavelength scan function of the F-2000 fluorometer, scanning at 1 nm/s.
Chilling samples
Two circulating water baths were connected by insulated tubing
to the jacketed cuvette holder and separated from each other by valves.
The heated water bath was held at 45°C, and the chilled water bath
was held at 
5°C. Sample temperature was monitored with an Omega
(Stamford, CT) microprocessor thermometer using a thin K-type
thermocouple wire threaded into the cuvette, but not disrupting the
beam. To cool samples, the heated bath valve was closed and the chilled
bath valve was opened. Reversal of this process warmed the samples.
pH and viscosity
A Corning pH/ion analyzer 255 was used to measure pH of the solutions and samples. Viscosity was monitored by measuring the flow rate of a given solution from a 100-mL reservoir in a buret through a borosilicate capillary tube (75 mm × 1 mm (I.D.)).
Fluorescence lifetimes
Multifrequency phase-modulation fluorimetry was used to measure
the excited-state lifetimes of all five dyes under both calcium-free and calcium-saturated conditions at 5 and 40°C. All time-resolved fluorescence intensity decay kinetics were measured in the frequency domain using an SLM-AMINCO 48000 multiharmonic Fourier (MHF)
spectrofluorometer (Spectronic Instruments). The instrument and its
capabilities have been described in detail elsewhere (Wang et al.,
1995). For these particular experiments, we used a cw argon-ion laser
(Coherent, Innova 90-6) operating at 488.0 nm as the excitation source
for calcium green and a 325.0 nm He-Cd laser (Omnichrome Series 74XA) for all other probes. In both cases, an appropriate interference filter
(Oriel, Stratford, CT) was placed in the excitation beam path to
prevent extraneous plasma-tube superradiance or Rayleigh scatter from
reaching the detection system. The sample fluorescence was monitored in
the typical L-format after passing through either a 515-nm (argon-ion)
or a 345-nm (He-Cd) longpass filter. The Pockels cell modulator was
operated at a 5 MHz base repetition rate. Typically, data sets were
acquired for 60 s between 5 and 150 MHz (30 total frequencies). At
least 10 discrete multifrequency data sets were acquired for each
sample under a given set of experimental conditions. For the
excited-state intensity decay measurements, we used a dilute solution
of rhodamine 6G dissolved in water or Me2POPOP in
ethanol as the reference lifetime standards with assigned lifetime
values of 3.85 ns (Heitz and Bright, 1995) and 1.45 ns (Lakowicz,
1983), respectively. Magic angle polarization conditions were used for
all excited-state intensity decay measurements to eliminate bias
arising from fluorophore rotational reorientation.
In the frequency domain, the experimental measurables for excited-state
fluorescence intensity decay experiments are the frequency-dependent phase angle (
) and demodulation factor (M). The
excited-state fluorescence lifetimes are recovered from the
frequency-domain data by using a commercially available nonlinear least
squares software package (Globals Unlimited, Urbana, IL) (Beechem et
al., 1991). In all data analyses, we used the true uncertainty in each datum as the frequency-dependent weighting factor. Background and
supplemental details on the theory of phase-modulation fluorescence can
be found elsewhere (Bright, 1995; Bright et al., 1990). The time-resolved fluorescence intensity decay I(t)
can generally be sufficiently described by a stretched exponential of
the form (Bright, 1995; Bright et al., 1990),
|
(1) |
i. For a
multiexponential decay (e.g., n = 2), the arithmetic
mean excited-state lifetime (
) can be calculated as
|
(2) |
| |
RESULTS AND DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
REFERENCES
|
|---|
Effect of temperature on fluorescence
The reported calcium concentration is roughly proportional to fluorescence intensity for the single-wavelength dyes, and roughly proportional to the ratio of fluorescence intensity at two wavelengths for the ratiometric dyes. If fluorescence is affected by temperature, it follows that variations in temperature would affect the calcium concentration value reported by the dye. In fact, fluorescence intensities of the single-wavelength dyes (CTC, Quin-2, and calcium green) increased by as much as twofold due to a shift from 40 to 5°C (Fig. 1). This result cannot be ascribed to the effect of temperature on Kd, because it is seen under both fully saturated (1 mM Ca2+) and calcium-free (10 mM EGTA) conditions. The extent of the fluorescence gain due to the temperature shift varied widely among the three dyes tested, with CTC and Quin-2 showing much more dramatic increases than calcium green. Calcium green, therefore, may be a more ideal choice for an investigation in which temperature is being changed than either CTC or Quin-2.
|
Although the effect of temperature on fluorescence intensity of the
single-wavelength dyes is dramatic, the resulting effect on
calcium-sensing is likely to be smaller than one might predict on the
basis of Fig. 1. Because both the bound and free forms of the dye are
affected by temperature, and because calcium determination is based on
a self-ratioing equation (Tsien and Pozzan, 1989), part of the
temperature effect will be controlled by this calculation. However, the
effects of temperature on the bound and free forms are not identical,
and a 20-30% difference remains between the Fmin/Fmax
ratios at 5 versus 40°C. Thus, the effect of temperature on
fluorescence intensity is one potential source of error in investigations involving variable temperatures. The other main source
of error is, of course, the effect of temperature on
Kd of the dye. As stated above, this
effect has already been thoroughly investigated, and thus is not
addressed in the current study. Nevertheless, both must be considered
to obtain reliable data.
Fluorescence intensities of the ratiometric dyes were also affected by temperature. Indo-1 and Fura-2 were examined in both calcium-saturated and calcium-free conditions by chilling and rewarming the samples. Indo-1 showed a decrease in the fluorescence ratio (Em 405/480) at low temperature under both conditions, although the effect was more pronounced in the calcium-free samples (Fig. 2 A). Nevertheless, a decrease of 10-25% in the fluorescence ratio could have a significant effect on the reported calcium concentration.
|
In most cases, Fura-2 showed a decrease in the fluorescence ratio (Ex
340/380) at low temperature as well (Fig. 2 B). A
complicating factor, however, was that, under calcium-saturating
conditions, the magnitude
and even directionof the change in
fluorescence ratio depended on the age and spectroscopic history of the
sample (which was not the case for Indo-1). This phenomenon is
illustrated in the inset to Fig. 2 B. A Fura-2 sample (1 mM
Ca2+) was repeatedly temperature-scanned over a
period of 10 days. On the initial scan, the fluorescence ratio
increased by 7% during the shift from 40 to 5°C. On all subsequent
scans, however, the fluorescence ratio decreased at low temperature,
and the magnitude of this effect became more pronounced with each
successive scan. This finding emphasizes the importance of careful
controls during variable temperature investigations. The degree to
which the fluorescence ratio varies with temperature depends on the
type of dye, the dye concentration, calcium concentration, and, in some
cases, age of sample.
These many and complex variables might help to explain the apparent
discrepancy of these results with one study that reported no change in
the fluorescence ratio during a temperature shift (Rink et al., 1982a).
Such effects have been seen previously, however. For instance, Lakowicz
notes that Fura-2 may undergo phototransformation producing
noncalcium-sensitive subpopulations (Lakowicz et al., 1994; Becker and
Fay, 1987). Clearly, the photobleaching has been documented to some
extent, but this effect has not been widely understood in the community.
As is the case for the single wavelength dyes, then, temperature has a
considerable effect on the fluorescence of the ratiometric dyes as
well, as evidenced by the temperature-induced trends in the 1/2
ratios. These effects must be considered, along with the effects of
temperature on dye Kd, to measure free
calcium accurately.
Possible mechanisms
To illuminate the causes of the effects of temperature on the calcium-sensitive dyes, we investigated several possible mechanisms using two different methods. One method involved mimicking the physical changes that occur in the samples during cooling while holding temperature constant. The alternate method was to check for specific changes in fluorescence characteristics that could be responsible for the effects of low temperature, and determine if they did indeed occur when temperature was reduced.
pH
The well-known phenomenon of increasing pH at decreasing temperature is responsible for important changes regarding the binding equilibria of fluorescent calcium chelators (Lattanzio, 1990; Lattanzio
and Bartschat, 1991; Shuttleworth and Thompson, 1991). To determine if
the pH changes caused by cooling were responsible for any of the
fluorescence changes as well, the pH change was mimicked in constant
temperature samples. During chilling, the pH of PBS increased from 7.0 (at 37°C) to 7.3 (at 5°C). When the pH of samples, held constant at
37°C, was shifted from pH 7.0 to 7.3 by the addition of base, there
was no significant change in fluorescence intensity for most of the
five dyes tested (Table 1). Of the
single-wavelength dyes, only CTC showed an increase in fluorescence
intensity with an increase in pH. However, the shift to pH 7.3 was not
sufficient to cause the full fluorescence rise seen during the shift to
5°C (~60% increase). Nevertheless, the elevation in pH could
contribute to the fluorescence intensity changes seen in CTC with
temperature. This trend is not an absolute, however, and each system
must be measured separately. Evidence for this comes from measurements
of CTC at higher concentrations. For instance, CTC when measured at 380 µM, showed the opposite trend (i.e., a decrease in fluorescence
intensity as pH increased), possibly due to the existence of aggregated
species.
|
Viscosity
Another important physical parameter of solutions that changes with temperature is viscosity. Because viscosity has been shown to be correlated inversely with oxygen quenching (Eftink and Ghiron, 1987)
and positively correlated with quantum yield (Eastman and Rosa, 1968),
we tested the calcium-sensitive dyes to determine if the increase in
viscosity at low temperature were responsible for the temperature-based
effects on fluorescence. The higher viscosity of the low-temperature
samples was mimicked by the addition of sucrose to the buffer. We
determined that a solution of PBS with 1.1 M sucrose had approximately
the same viscosity at 37°C as the PBS alone had at 5°C (1.5 cp
[CRC, 1984]) as described in Materials and Methods. The effects of
viscosity on fluorescence intensity are shown in Fig.
3.
In the case of the single-wavelength dyes (Fig. 3 A),
fluorescence intensities in the high viscosity PBS were not
significantly higher than in the PBS. In fact, in the case of CTC, the
intensity in the high viscosity PBS was significantly lower than in the
PBS alone. Clearly, viscosity is not responsible for the elevated
fluorescence intensity at hypothermic temperatures for the single
wavelength dyes.
|
|
1/2 fluorescence
ratio (as is seen in Fig. 2), the fluorescence intensity increase due
to viscosity would have to be greater for wavelength-2 than for
wavelength-1. In fact, the opposite is true. Thus, viscosity cannot be
responsible in the case of Indo-1. It has been noted (Poenie, 1990;
Busa, 1992) that high viscosity can reduce the 1/2 ratio for
Fura-2, but the viscosity at which these effects were seen was very
large (15 cp) compared to this study (1.5 cp), and, as stated, under
the current conditions, the effect of viscosity on Fura-2 was not
significant. However, aside from the effects on fluorescence mentioned
above, changes in viscosity can also affect the binding kinetics of the
dyes (Kao and Tsien, 1988). Thus, viscosity is similar to pH in that it
can have important consequences in regard to both binding
characteristics and fluorescence values.
Quenching
Fluorescence quenching due to molecular oxygen is directly proportional to temperature (Eftink and Ghiron, 1987). Therefore, it is
possible that the fluorescence increases seen at low temperature might
be due to a drop in oxygen quenching. Samples of all five dyes in the
presence or absence of calcium (1 mM CaCl2 or 10 mM EGTA) were tested in sealed cuvettes, either with or without a 20-min purge (by bubbling with N2 gas) to
eliminate dissolved oxygen (Saltiel et al., 1993). There were no
significant differences between the temperature effects of purged and
nonpurged samples for any of the dyes tested (Table
2). This is expected due to the short
excited-state lifetimes of the probes involved. Thus, modulation of
oxygen quenching cannot be responsible for the effects of temperature
seen on the calcium-sensitive dyes.
|
Spectral shifts
Changes in excitation or emission maxima with temperature have been noted for several fluorophores (Demchenko and Ladokhin, 1988; Law,
1994; Suelter, 1967). To investigate the possibility that such shifts
could be responsible for the effects of temperature on the fluorescent
dyes, we ran wavelength scans in the presence and absence of calcium at
40 and 5°C. Most of the maxima were not shifted by temperature (Table
3). However, a 6-nm red shift did appear
for the emission maximum of calcium-saturated Fura-2 (493 to 499 nm)
when the temperature was reduced from 40 to 5°C, as has been
previously described (Paltauf-Doburzynska and Graier, 1997). In
addition, a 4-nm red shift (469 to 473 nm) was evident for the emission
maximum of Indo-1 under calcium-free conditions resulting from the same
temperature reduction.
|
1)
emission maximum closer to the measured emission wavelength (510 nm).
This could theoretically have the effect of raising the fluorescence
intensity at the +Ca value, which would tend to increase the ratio
I+Ca/I
Ca at low temperature. Mimicking the
temperature-induced spectral shift at constant temperature by bringing
the value of the measured emission wavelength 6 nm closer to the +Ca
emission maximum (from 510 to 504 nm) caused a minimal change in the
fluorescence ratio (approx. +1% change, positive as predicted).
In the case of Indo-1, in contrast, the shift in emission maximum from
469 to 473 nm under calcium-free conditions brings the Ca maximum
closer to the Ca measured value of 480 nm. This would tend to
increase the fluorescence intensity at the Ca value, which would tend
to decrease the
I+Ca/ICa
ratio at low temperature. This is, in fact, what occurs when Indo-1 is chilled (see Fig. 2). Mimicking the spectral shift at constant temperature by reducing the measured Ca emission value from 480 to
476 nm, which brings it 4 nm closer to the Ca emission maximum, did
cause a significant (~10%) decrease in the fluorescence ratio. One
reason for the larger change seen for Indo-1 than for Fura-2 in this
regard is that, for Indo-1, the fluorescence maximum, which is shifted
by temperature, is the same one that is shifted by calcium binding
(emission). In contrast, for Fura-2, the emission maximum is shifted by
temperature, whereas the excitation maximum is shifted by calcium
binding. This could lead to a less pronounced effect of spectral shifts
for Fura-2. Nevertheless, this mechanism may contribute to the change
in the fluorescence ratio seen at low temperature for both ratiometric dyes.
Fluorescence lifetimes
There is a considerable body of literature showing that excited-state lifetimes of many fluorescent molecules are lengthened as temperature decreases (Drain et al., 1998; Kumke et al., 1997; Young et
al., 1997; Bark and Force, 1993; van den Zegel et al., 1984) in
correlation with a decrease in the rates of nonradiative decay (e.g.,
decreased coupling to vibronic bath modes) (Lam et al., 1998; Young et
al., 1997; Giri, 1992). This leads to a net increase in fluorescence
intensity (i.e., quantum yield), because, when decay back to the ground
state does occur, it is more likely through emission of a photon. To
determine if this trend held for the calcium-sensitive dyes as well,
the lifetimes of all five dyes were measured using multifrequency
phase-modulation fluorimetry, under both calcium-free and
calcium-saturated conditions, at 40 and 5°C (Fig. 4). In all cases,
the data were best described by a double exponential decay, as shown in
Table 4. Multiple exponential decays have
been reported before for Quin-2 and Indo-1 (Lakowicz et al., 1992,
1994; Szmacinski et al., 1993), and excited-state reactions were
suggested as a possible cause (Szmacinski et al., 1993).
|
|
; Miyoshi et al., 1991) stemming from its sensitivity to
calcium binding. Clearly, without detailed prior knowledge of its
temperature-dependent behavior, this probe would be a poor choice for
applications involving variable temperature.
The ratiometric dyes show a similar trend, in that reduced temperature
leads to increases in the fluorescence lifetimes. However, the effects
are much more pronounced in the calcium-free condition than in the
calcium-saturated condition. The increase in fluorescence intensity at
the calcium-free values would cause a decrease in the
I+Ca/ICa
ratio at low temperature, which correlates well with experimental
results for Indo-1. In the case of Fura-2, the effect of temperature
predicted by the exited-state lifetimes (decrease in fluorescence
ratio) contrasts with that predicted by the spectral shifts (increase
in fluorescence ratio). The balance between these two aspects of the
system may determine the overall effect of temperature on the
fluorescence ratio of Fura-2. The effect of temperature on the
excited-state lifetimes of the calcium-sensitive dyes is, therefore,
likely to be important in determining the overall effect of temperature
on fluorescence.
Can fluorescent dyes be used to measure Ca2+ at variable temperatures?
Although the calcium-sensitive dyes show a strong dependence of fluorescence on temperature, they can still be used to make reasonably accurate measurements of calcium concentration at variable temperatures, as long as the investigation controls for both the intensity and Kd effects. The most appropriate method for accomplishing this task involves conducting all data manipulation at the sample temperature of interest. For instance, Indo-1 and Fura-2 can both be used to accurately measure calcium at several different temperatures, as long as the maximal calcium reading (1 mM Ca2+) and the minimal calcium reading (10 mM EGTA) are taken at the same temperature as the measured calcium reading, and that the correct temperature-adjusted Kd is used for the calculation.
A second method that can be used to measure the calcium concentration
in samples at variable temperature is better suited for experiments
that require continuous measurement rather than measurements at
discrete time and temperature points. In such cases, it is important to
find a dye that is nontemperature-dependent in the range of interest.
For example, in the physiological range, the calcium concentration
reported by Indo-1 shows little temperature sensitivity between 20 and
5°C, although it is highly temperature-dependent in the range from 20 to 40°C (Oliver et al., 1999). Therefore, it is possible to estimate
continuous changes in calcium concentration as a function of
temperature between 20 and 5°C using Indo-1 (Oliver et al., 1999),
but would not be recommended in other more temperature-dependent ranges. Even under favorable circumstances of low sensitivity of
fluorescence to temperature, however, it is still critical to control
for the effects of temperature on Kd,
and determine that this type of artifact is not responsible for any
measured effect on intracellular calcium.
In summary, the fluorescence behavior for all five calcium-sensitive probes studied revealed some degree of temperature sensitivity. Increase in excited-state lifetimes along with reduced rates of nonradiative decay, are likely to contribute strongly to such effects. In addition, increasing pH may play a role in the effect of temperature on CTC, and spectral shifts probably also contribute to the effect of temperature on the ratiometric probes. Nevertheless, the intermolecular interactions in a fluorescent solution are extremely complex, and the effects of decreasing fluorescence intensity at elevated temperatures are likely to be the result of many different factors, including some not considered in the current study. Although the general effect of temperature on fluorescence intensity is well known to photophysics specialists, the use of the calcium-sensitive dyes is becoming extremely common for researchers specializing in many other disciplines as well. Therefore, we have attempted to describe some of the photophysical effects of temperature on these dyes in an effort to draw attention to the importance of considering this issue if temperature is used as an independent variable in experiments using the calcium-sensitive dyes.
| |
ACKNOWLEDGMENTS |
|---|
We thank Dr. Richard Nuccitelli for the generous gift of calcium green dextran.
This investigation was supported by National Institutes of Health grants HL57810 and HL61294.
| |
FOOTNOTES |
|---|
Received for publication 21 October 1999 and in final form 12 January 2000.
Address reprint requests to Ann E. Oliver, Section of Molecular and Cellular Biology, University of California, Davis, Davis, CA 95616. Tel: 530-752-1094; Fax: 530-752-5305; E-mail: aeoliver{at}ucdavis.edu.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
REFERENCES
|
|---|
-thrombin.
Biochemistry.
35:906-914[Medline]. current and Ca2+ transients in sensory neurones.
Cell Calcium.
24:35-48[Medline].-dimyristoylphosphatidylcholine vesicles.
Biophys. Chem.
20:333-345, w-diphenylpolyenylic carbanions.
J. Photochem. Photobiol. A: Chemistry.
102:163-172.
Biophys J, April 2000, p. 2116-2126, Vol. 78, No. 4
© 2000 by the Biophysical Society 0006-3495/00/04/2116/11 $2.00
This article has been cited by other articles:
 |
|
 |
|
  E. Guatteo, K. K. H. Chung, T. K. Bowala, G. Bernardi, N. B. Mercuri, and J. Lipski Temperature Sensitivity of Dopaminergic Neurons of the Substantia Nigra Pars Compacta: Involvement of Transient Receptor Potential Channels J Neurophysiol, November 1, 2005; 94(5): 3069 - 3080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. F. Lee, J. C. Callaway, and R. C. Foehring Effects of Temperature on Calcium Transients and Ca2+-Dependent Afterhyperpolarizations in Neocortical Pyramidal Neurons J Neurophysiol, April 1, 2005; 93(4): 2012 - 2020. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rubart Two-Photon Microscopy of Cells and Tissue Circ. Res., December 10, 2004; 95(12): 1154 - 1166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Boudreault and R. Grygorczyk Cell swelling-induced ATP release is tightly dependent on intracellular calcium elevations J. Physiol., December 1, 2004; 561(2): 499 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Protasi, A. Shtifman, F. J. Julian, and P. D. Allen All three ryanodine receptor isoforms generate rapid cooling responses in muscle cells Am J Physiol Cell Physiol, March 1, 2004; 286(3): C662 - C670. [Abstract] [Full Text]
|
|
|
|
|
|
M. Rubart, E. Wang, K. W. Dunn, and L. J. Field Two-photon molecular excitation imaging of Ca2+ transients in Langendorff-perfused mouse hearts Am J Physiol Cell Ph |
), chlortetracycline (
), and calcium green dextran (
).
Calcium green, for which a second y axis is shown at the
right, has a much smaller increase in fluorescence intensity than
either Quin-2 or chlortetracycline.
). Ratios are reported as percentages of the
40°C values. (B) Dependence on temperature of the
) and the 4th scan
(t = 7 h) (
) and 40°C (
) and 40°C (
). The solid
lines denote the best double-exponential fits to the experimental data.
These data were analyzed as described in Materials and Methods to
determine the excited-state lifetimes for each fluorophore (see Tables