Download Interaction of oxygen-sensitive luminescent probes Ru(phen) and

Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136–144
www.elsevier.com / locate / jphotobiol
Interaction of oxygen-sensitive luminescent probes Ru(phen) 21
and
3
21
Ru(bipy) 3 with animal and plant cells in vitro
Mechanism of phototoxicity and conditions for non-invasive oxygen
measurements
Jurek W. Dobrucki*
Laboratory of Confocal Microscopy and Image Analysis, Department of Biophysics, Institute of Molecular Biology, Jagiellonian University,
ul. Gronostajowa 7, 30 -387 Kracow, Poland
Accepted 15 October 2001
Abstract
Understanding the role of oxygen in the physiology, pathophysiology and radio- and chemosensitivity of animal cells requires accurate
and non-invasive measurements of oxygen concentrations in the range of 0–2310 24 M, in cells in vitro or in vivo. High resolution 3D
imaging techniques could be particularly useful in investigating tissue oxygenation in vivo and in model tissues (multicellular spheroids)
in vitro. The goals of this work were to develop microscopy techniques and (i) to define conditions under which two oxygen-sensitive
luminescent dyes, Ru(bipy) 21
(tris(2,29-bipyridyl)ruthenium(II) chloride hexahydrate) and Ru(phen) 21
(tris(1,10-phenanthro3
3
line)ruthenium(II) chloride hydrate) can be used to probe oxygen concentrations within viable cells in vitro, when no phototoxic effects
are evident, and (ii) to investigate the mechanism of phototoxicity once cell damage occurs. This report demonstrates that Ru(bipy) 21
and
3
Ru(phen) 321 do not pass through intact biological membranes, do not cause measurable photodamage to plasma membranes at a
concentration of 0.2 mM and, when loaded into endosomes, yield a strong luminescent signal. However, at an extracellular concentration
of 1 mM, in the presence of 457-nm light, detectable amounts of both complexes accumulate at the plasma membrane and cause a loss of
membrane integrity via a mechanism which may involve the generation of singlet oxygen.  2001 Elsevier Science B.V. All rights
reserved.
Keywords: Ruthenium bipyridyl; Ruthenium phenantroline; Oxygen; Cells; Phototoxicity; Confocal microscopy
1. Introduction
Oxygen is a known modifier of radio- and chemosensitivity of animal cells. Survival of mammalian cells
irradiated with low LET radiation in the absence of oxygen
is approximately three times higher than in the presence of
oxygen [1]. Oxygen may also modify the interaction of
drugs with cells [2]. Understanding the role of oxygen in
physiology and pathophysiology as well as in radio- and
chemosensitivity of animal cells requires accurate and
non-invasive measurements of oxygen concentrations in
the range of 0–2310 24 M, in cells in vitro or in vivo. As
hypoxic cells and tissue areas undergoing reoxygenation
may be very small, high resolution 3D imaging techniques
could be particularly useful in investigating these phenom*Tel.: 148-12-252-6382; fax.: 148-12-252-6902.
E-mail address: [email protected] (J.W. Dobrucki).
ena in vivo and in model tissues (multicellular spheroids)
in vitro. Measurements in single cells, at low partial
pressures of oxygen, seem particularly interesting as they
might aid studies of the role of hypoxic cells in tumor
regrowth and investigations of tissue reoxygenation.
A number of methods have been designed to measure
oxygen in solution. These rely on oxygen-sensitive electrodes, luminescence [3], EPR [4] and NMR [5]. Some of
these methods appear to be amenable for use in cell
cultures and tissues, even though several shortcomings
restrict their versatility. For instance, small oxygen electrodes consume oxygen themselves and the measurements
are invasive. EPR approaches use nitroxides or particulate
probes, like lithium phthalocyanine (LiPc), coal, soot or
even India ink [6]. While some of these agents exhibit no
apparent cytotoxicity (LiPc), others, like nitroxides, may
inflict damage on cells. EPR probes exhibit a spectrum of
dynamic ranges and sensitivities to dissolved oxygen, from
1011-1344 / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved.
PII: S1011-1344( 01 )00257-3
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
relatively low for nitroxides like CAT1, to very high, for
crystals of LiPc. The spatial resolution of EPR oxygen
images obtained in phantoms and tissues is too low for
imaging individual cells [4]. NMR approaches appear to be
non-invasive but also offer limited resolution [7]. In
contrast, approaches based on the use of luminescent
probes like ruthenium(II) complexes may allow the sensitive, high resolution and non-invasive analysis of oxygen
in living tissues and even individual cells [8].
Several factors need to be considered in order to select
the right luminescent probe for physiologically relevant
measurements of oxygen in viable cells. These factors
include intracellular concentration, localization and distribution of a probe, toxicity and phototoxicity, probe
stability, pH dependence of luminescence, and influence of
soluble proteins on luminescence properties. As demonstrated in this report, ruthenium(II) complexes cannot
freely pass through intact biological membranes. Therefore, they could be deliberately confined to the extracellular space or to endosomes. Moreover, luminescent signals
should be sufficiently strong to perform measurements of
phase shift and further calculations of luminescence lifetimes [8]. Measurements in primary endosomes seem
particularly interesting due to the rapid consumption of
oxygen by active NADPH oxidase. Moreover, luminescence of ruthenium(II) complexes appears to be independent of pH [9,10] within the range found naturally in
endocytic vesicles (i.e. pH 5.0–7.2). Ru(II) complexes do
not appear to be destroyed in lysosomes, at least not at a
rate detectable under the experimental conditions we used.
Obviously, due to vesicle fusion and formation of secondary lysosomes concentrations of Ru(phen) 21
complex3
es may vary between vesicles in the same cell and between
cells. Hence, fluorescence lifetime imaging microscopy
(FLIM) [8,11] provides the preferred technique for oxygen
determination in cells and cell organelles such as endosomes. Measurements of fluorescence lifetime are independent of probe concentration—within the range we can
expect in cells—and thus immune to probe redistribution,
destruction by metabolism or photobleaching and removal
from cells. Unfortunately, potential obstacles to intracellular FLIM measurements also exist, since oxygen sensitivity
of Ru(II) complexes bound to proteins or nucleic acids is
lower than in solution and may be difficult calibrate in live
cells. While luminescence lifetimes of Ru(bipy) 21
and
3
Ru(phen) 321 complexes determined in aerated solutions
were apparently only negligibly influenced by the presence
of serum proteins [8], the luminescence lifetimes of a new
oxygen probe [Ru(dpp(SO 3 Na) 2 ) 3 ]Cl 2 were shown to be
strongly influenced by proteins [12].
Luminescence of several ruthenium(II) complexes dissolved in water has been shown to depend on oxygen
concentration [13]. Luminescence lifetimes in aerated
solutions are shorter than in deoxygenated environments
by 36 and 51% for Ru(bipy) 21
and Ru(phen) 321 , respec3
tively [14]. New ruthenium complexes are also being
137
developed as better oxygen probes [15,16]. Ruthenium
complexes embedded in plastic were used for constructing
oxygen sensors [17–22]. Dissolved ruthenium(II) complexes were also applied in oxygen measurements in
solution, for studies of enzymatic reactions, and in cells
grown in vitro [8,23–25].
Most, if not all, experiments involving fluorescent
probes in contact with live cells are limited by the
phototoxicity of the dyes used [26]. This problem was also
encountered in our preliminary experiments to measure
oxygen in individual intact cells, using Ru(phen) 21
and
3
Ru(bipy) 21
as
luminescent
probes
[8].
Although
the
results
3
were promising, it became clear that knowledge of the
interactions between these dyes and viable cells in culture
would be required to further develop this technique. With
this goal in mind, the current report describes interaction of
two complexes of ruthenium, Ru(bipy) 21
and Ru(phen) 21
3
3 ,
with viable animal and plant cells maintained in vitro. It
attempts to define conditions under which these Ru(II)
complexes can be used as oxygen probes in microscopy
studies, with no measurable phototoxic effects and investigate the mechanism of phototoxicity under the conditions
when it does occur.
2. Materials and methods
2.1. Chemicals
Tris(2,29-bipyridyl)ruthenium(II) chloride hexahydrate
(Ru(bipy) 21
[27]
and
tris(1,10-phenanthroline)3 )
ruthenium(II) chloride hydrate (Ru(phen) 21
3 ) were obtained from Aldrich.
2.2. Cell cultures
J144 mouse macrophages were obtained from American
Tissue Collection and cultured in Dulbecco’s MEM
(Sigma) supplemented with 10% fetal bovine serum
(Sigma) and antibiotics. Cells were subcultured by washing off substratum with a stream of culture medium.
Normal human fibroblasts were cultured as described
previously [28]. Logarithmically growing cell cultures
were used for experiments (|48 h following seeding). For
imaging experiments, cell cultures were seeded onto round
¨
coverslips (22 mm diameter, Menzel-Glaser),
placed in
Petri dishes and maintained in a CO 2 incubator (Jouan) for
2 days. Subsequently, coverslips were mounted in a
microscope stage microincubator (Life Science Resources,
Cambridge, UK). Prior to seeding cells, coverslips were
selected for thickness of 0.1760.005 mm, degreased,
washed, wrapped individually in aluminum foil and sterilized by autoclaving. Integrity of plasma membranes was
tested by exclusion of propidium iodide or by hydrolysis of
fluorescein diacetate.
138
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
2.3. Incubation of cells with Ru( II) complexes
For incubation of cells with ruthenium complexes,
culture medium was supplemented with the drug immediately prior to experiments. Prior to adding to cell cultures,
solutions of Ru(II) complexes were sterilized using a
0.22-mm Millipore filter.
2.4. Fluorescence confocal microscopy and image
analysis
Images of macrophages exposed to Ru(II) complexes
were collected using a Bio-Rad MRC1024 laser scanning
confocal system (with an inverted microscope Nikon
Diaphot 300), using a 603 PlanApo NA1.4 oil immersion
lens. Excitation was 457 nm (100-mW Ar ion laser, ILT)
with a 3% neutral density filter in the excitation light path.
The dichroic splitter was 510 DLRP (VHS block); emission was collected using a 580 LP emission filter. Cells
growing on round glass coverslips were maintained in a
microincubator resting on a microscope stage, at 37 8C in
Dulbecco’s MEM supplemented with HEPES (pH adjusted
to 7.2), without Phenol Red. Microincubator facilitated
replacement of culture medium, adding dye solutions and
changing gas atmosphere during illumination and imaging,
without altering the position of the imaged area of cell
culture. Sequences of images (5123512 pixels, 256 gray
levels) were collected and analyzed using LaserSharp
TimeCourse software (Bio-Rad). Black level of the photomultiplier was adjusted so that noise values were within
1–10; gain was adjusted so that the values of red autofluorescence of intact cells were within a range of 5–50.
For timecourse recordings, horizontal cross-sections of
single cells (six to ten in one series) were selected and
followed over time. An equivalent area of dye-containing
culture medium was recorded as a control. Traces of a
mean of luminescence intensity in each selected cell and in
surrounding medium as well as time-series of images were
displayed live during the experiment and recorded simultaneously. Further analysis of images and traces were
performed using LaserSharp, Origin or SigmaPlot software.
2.5. Illumination of cells and simultaneous observation
of damage to plasma membrane
Preliminary measurements of phototoxicity of Ru(II)
complexes were conducted by illuminating cell cultures on
a stage of an inverted microscope (Nikon Diaphot 300),
with 450–490-nm light emitted by a high pressure mercury
arc lamp (Nikon). Cell cultures were maintained on plastic
Petri dishes at room temperature or on coverslips mounted
in a microincubator, at 37 8C. Following illumination,
accumulation of Ru(II) complexes in the regions of plasma
membranes was observed. Subsequently, entry of the dyes
into damaged cells occurred. Following different periods
of illumination, the dye was removed, standard medium
was introduced and cultures were maintained in a CO 2
incubator for 3–6 days. Subsequently, cells were counted
once a day and growth curves of individual cultures were
constructed in order to estimate cell damage (data not
shown). In further experiments described below, illumination of cells was achieved during confocal imaging, i.e. by
scanning a focused laser light beam in a raster pattern over
a selected area of a cell culture maintained in a microincubator. One scan over the whole imaged area of
1103110 mm took 1 s. Thus, one pixel was illuminated for
less than 4 ms in 1-s intervals. The beam delivered |0.1
mW (unless indicated otherwise) of 457-nm light focused
onto a diffraction limited spot in the sample. The fluence
rate was 50 kW/ cm 2 ; fluence was 0.16 J / cm 2 for one scan
over one pixel. Thus, the total light dose delivered to cells
was proportional to the number of illuminations (confocal
frames). The setting of a black level and gain of the
detector (see above) allowed accurate detection of the
onset of the penetration of the dye into damaged cells. At
these gain settings the luminescence of the dye accumulated in the nuclei of damaged cells eventually exceeded
the dynamic range of the detector.
Where indicated, during illuminations cells were submerged in physiological saline made in D 2 O (Institute of
Nuclear Physics, Swierk, Poland). Anoxia was achieved by
purging the saline solution with nitrogen.
3. Results
3.1. Penetration of Ru( II) complexes through biological
membranes
Penetration of Ru(II) complexes through cell plasma
membranes was investigated by incubating fibroblasts with
Ru(bipy) 21
(data not shown) or macrophages with
3
21
Ru(phen) 3 (0.2 mM; Fig. 1), under standard growth
conditions. As shown in Fig. 1B, following an 8-min long
incubation, luminescence of Ru(II) complexes was detected only in culture medium, outside cells. Following a
70-min incubation (in the dark) a growing number of
cytoplasmic vesicles loaded with Ru(phen) 21
was ob3
served (Fig. 1C). Apparently, in a process of endocytosis
Ru(phen) 21
was sequestered into membrane vesicles and
3
transported into the cell interior. The data presented in Fig.
1 indicate that Ru(phen) 21
has no capacity to cross intact
3
outer cellular membranes or membranes of endocytic
vesicles. Similarly, when algae (Clostridium) and a fern
leaf (Adiantum capillus veneris) were incubated in water
supplemented with Ru(bipy) 21
3 , the fluorophore did not
penetrate into cells [9].
3.2. Affinity to plasma membranes
When cells were incubated for 5 min with Ru(phen) 21
3
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
139
es, and the dyes subsequently removed, red luminescence
remained in the region of outer cell membrane (Fig. 2; see
also Fig. 1C). This observation indicates that, although
Ru(II) complexes themselves had no affinity for membranes, illumination with light did result in a measurable
accumulation of the dye in the region of plasma membrane. In this experiment, plasma membrane integrity was
not compromised as demonstrated by fluorescein diacetate
assay (data not shown) and the absence of Ru(II) complexes from cell nuclei (Discussion section). Thus, it is likely
that the dye accumulated at the outer surface of plasma
membrane.
3.3. Affinity to nucleic acids
Affinity of Ru(II) complexes to intracellular structures
exposed by fixation of cells with formaldehyde (1% at
room temperature for 10 min) was investigated. A dye
solution (2 mM) was added to fixed cells for 5 min and the
unbound dye was removed by rinsing cell preparations
three times with PBS. Microscopy observation revealed
that the dye accumulated primarily in cell nuclei and some
residual luminescence of the dye remained in cytoplasmic
regions (data not shown). Following photodamage to
plasma membranes a similar pattern of Ru(phen) 21
ac3
cumulation in cells was seen (see below). These observations were consistent with the reported affinity of
Ru(phen) 21
for nucleic acids [14,29–31].
3
3.4. Phototoxicity of Ru( II) complexes
Fig. 1. Excluding Ru(phen) 21
by viable, intact macrophages and entry of
3
the dye into cells by means of endocytosis. Images represent red
luminescence in a culture of macrophages. (A) Prior to adding
Ru(phen) 21
(0.2 mM). Some granular autofluorescence in cytoplasm can
3
be detected. Bar represents 10 mm. (B) Following 8 min of incubation
with Ru(phen) 21
3 . A confocal section through cells and the surrounding
medium showing that Ru(phen) 321 does not enter cells and remains
extracellular. (C) Following 70 min of incubation with Ru(phen) 21
and
3
removal of the dye-containing medium. Luminescence of Ru(phen) 21
is
3
detected exclusively in endocytic vesicles in the cytoplasm, and in plasma
membranes (arrowhead); no Ru(phen) 21
can be detected in nuclei.
3
or Ru(bipy) 21
and the dyes subsequently removed, no
3
measurable luminescence of Ru(II) complexes remained
associated with cells. However, if cells were illuminated
with 457-nm light during incubation with Ru(II) complex-
Although Ru(phen) 21
is excluded by cells during short
3
incubations in the dark, exposure to light may result in
phototoxicity of this fluorophore. The phototoxic effects
exerted by Ru(phen) 21
were measured in order to establish
3
the concentrations of the dye and the number of confocal
sections that could be collected if Ru(phen) 21
was to be
3
used as a fluorescent oxygen probe in direct contact with
live cells. Viable macrophages were maintained under
optimal growth conditions, in the microincubator resting
on the microscope stage. Immediately after adding
Ru(phen) 21
(0.2 mM) to culture medium a selected area of
3
the cell culture was exposed to 457-nm light. Under these
conditions no measurable endocytosis of Ru(phen) 21
took
3
place. When the number of illuminations exceeded 120
some deterioration of plasma membrane integrity became
evident and Ru(phen) 21
started to slowly enter cells (Fig.
3
3). This implies that illumination-induced phototoxic damage
compromises
membrane
integrity.
Initially,
Ru(phen) 21
which
passed
through
plasma
membranes
3
accumulated in cell nuclei (a representative sequence of
images can be viewed at http: / / helios.mol.uj.edu.pl). At
this stage the cytoplasm exhibited no measurable luminescence of the dye. Subsequent doses of light resulted in a
sudden collapse of plasma membrane integrity and a
massive entry of the dye followed by Ru(phen) 21
accumu3
140
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
Fig. 2. Accumulation of Ru(phen) 21
on plasma membranes of macrophages. A cross-section of a confocal image of one cell, recorded prior (A) and
3
following (B) 20 illuminations (457 nm, 0.4-mW laser beam) in the presence of 2 mM Ru(phen) 321 . Arrows depict position of plasma membranes.
lation in cell nuclei and cytoplasm. The intensity of
intracellular Ru(phen) 21
luminescence exceeded that of
3
culture medium. These high luminescence values resulted
partly from an increase of luminescence of Ru(phen) 21
3
upon binding to DNA [14] and partly from an accumulation of the dye bound to DNA [14], and possibly RNA, in
damaged cells. As a control, at the end of each experiment
the area of the culture, which had not been exposed to laser
illumination, was examined. Typically, no accumulation of
the dye was detected confirming that damage to plasma
membranes was a result of phototoxicity rather than any
inherent toxicity of the Ru(II) complex.
Fig. 3. Phototoxicity of Ru(II) complexes manifested by a loss of plasma
membrane integrity and entry of the dye into cytoplasm. The traces depict
intensity of luminescence of Ru(phen) 21
determined in culture medium
3
(a) and in the interiors (cytoplasm and nucleus) of six selected cells (b),
during illumination with 457-nm light. For clarity, in each trace only
every tenth reading of luminescence is plotted. Luminescence of the dye
accumulated in nuclei eventually exceeded the dynamic range of the
photomultiplier.
A summary of a series of similar experiments, depicting
phototoxic effects occurring in the presence of three
different concentrations of Ru(phen) 321 , is shown in Fig. 4.
The number of confocal sections that could be collected
without inflicting damage on plasma membranes was |20,
80, and over 120, at concentrations of 2, 1 and 0.2 mM of
Ru(phen) 21
3 , respectively.
Membranes of endocytic vesicles were also subject to
photodamage. In a similar experiment, macrophages were
exposed to Ru(phen) 21
for 90 min prior to illumination
3
and allowed to internalize the dye. Following this incubation the cytoplasm contained a large number of endosomes
loaded with Ru(phen) 21
(Fig. 5B). The subsequent illumi3
nation of cells resulted in breakage of the membranes of
Fig. 4. Photodamage exerted on plasma membranes by Ru(phen) 21
at
3
three different concentrations. Each point is a mean of intensities of
luminescence of Ru(phen) 21
entering eight selected cells. The values of
3
standard deviation were within the 10–60 range. The traces for individual
cells were collected as in Fig. 3. The curves were normalized to a level of
red autofluorescence of cytoplasm.
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
141
endosomes and release of Ru(phen) 21
3 . Consequently,
intact luminescing vesicles disappeared from images and
the liberated dye accumulated in cell nuclei (Fig. 5C).
When only selected regions of large cells (spindle-shape
human fibroblasts and some plant cells) were illuminated
in the presence of Ru(phen) 21
and Ru(bipy) 21
3
3 , the dyes
entered cells and accumulated exclusively in the regions
exposed to light. An example of this phenomenon is shown
in Fig. 6. This could be explained by the fact that plasma
membranes, and subsequently internal membranes, were
damaged exclusively in the illuminated areas. Subsequently, Ru(phen) 21
became bound to membranes in these
3
damaged regions.
3.5. Mechanism of phototoxicity of Ru( phen) 21
3
In order to understand the mechanism of phototoxicity
induced by Ru(phen) 21
3 , damage to plasma membranes
was measured in the presence and absence of oxygen.
Measurements were also carried out in heavy water, which
is known to lengthen the lifetime of singlet oxygen [32],
and in the presence of histidine (1 mM), a photoprotector
that acts by scavenging singlet oxygen [33]. Fig. 7
illustrates representative experiments in which luminescence of Ru(phen) 21
was followed in individual cells.
3
Under standard conditions, |100–150 illuminations could
be performed before loss of membrane integrity allowed
the dye to enter cells (Fig. 3). In the absence of oxygen
over 500 illuminations delivered to cells caused no damage
to plasma membranes (Fig. 7). In the presence of heavy
water |20–30 frames were sufficient to compromise
plasma membrane integrity. When histidine was added to
physiological saline made in D 2 O, the number of frames
resulting in damage to plasma membranes was again over
100. Accumulation of the dye in cells was significantly
slower (Fig. 7) and cell swelling was observed. Cells in
unilluminated areas did not accumulate the dye. These
observations indicate that singlet oxygen is involved in the
photodynamic damage exerted by Ru(phen) 21
3 .
4. Discussion
4.1. Interaction of Ru( II) complexes with cells
21
3
Fig. 5. Photodamage to endosomes loaded with Ru(phen) (2 mM). (A)
A transmitted light image of a group of macrophages (prior to incubation
with the dye). Bar represents 10 mm. (B) A confocal image of red
luminescence, collected after 90-min incubation with Ru(phen) 21
in the
3
dark and removal of the dye-containing medium. Endosomes are loaded
with Ru(phen) 21
3 . (C) Red luminescence of the dye following 110
illuminations. This illumination caused rupture of endosomes, release of
the dye into cytoplasm and subsequent accumulation in nuclei. The
photomultiplier gain was three times higher in C than in B.
This report demonstrates that selected Ru(II) complexes
are unable to cross intact biological membranes. This
behavior could be anticipated as they carry a positive
charge and are insoluble in lipids. In this respect Ru(II)
complexes behave similarly to a known ‘vital dye’,
propidium iodide (PI) [34]. PI is also a water soluble,
positively charged molecule and is known to remain
extracellular with respect to live cells. The analogy can be
taken further, since Ru(phen) 21
3 , just as PI, enters cells
142
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
Fig. 7. Photodamage exerted on plasma membranes by Ru(phen) 21
(0.2
3
mM), in the environment in equilibrium with air, in the absence of
oxygen, in the presence of D 2 O, and D 2 O with histidine. Each curve
depicts changes of intensity of luminescence of Ru(phen) 21
in interior of
3
six to eight selected cells. Data for individual cells were collected as in
Fig. 3.
Fig. 6. Photodamage in a selected region of a cell. (A) A group of
macrophages incubated with Ru(bipy) 21
(0.2 mM). The dye does not
3
penetrate into cells. Bar represents 10 mm. (B) Following illumination
with 457-nm light photodamage occurs, the dye enters cells, and
accumulates in nuclei and on internal membranes. (C) Following removal
of extracellular Ru(bipy) 21
3 , the dye remains in cell nuclei and membranes. (D) An image of an area adjacent to the region illuminated
previously. The border between the illuminated and unilluminated areas is
marked with a white line. In some cells extending across this border,
Ru(phen) 21
accumulated exclusively in the area illuminated previously,
3
while parts of the same cells that had not been illuminated remained
devoid of the dye.
with damaged plasma membrane and stains nucleic acids.
In other words, Ru(phen) 21
can be used as a ‘vital dye’ in
3
vitro. However, the luminescence of free PI is very weak
and increases 20–30-fold upon binding to DNA, while the
intensity of luminescence of unbound Ru(phen) 21
is
3
relatively strong and increases by less than 50% upon
binding to DNA [14,35]. Extracellular PI is usually not
detectable while unbound Ru(phen) 21
in culture medium is
3
strongly luminescent and can be used to visualize intracellular space between intact cells in multicellular spheroids
(J. Dobrucki, unpublished). In experiments involving
Ru(phen) 21
and live cells, penetration of the dye through
3
plasma membranes of intact cells can serve as a convenient
‘internal control’ for photodynamic damage to cells.
However, in an interesting report describing an attempt to
measure intracellular oxygen concentrations using
Ru(phen) 21
(1 mM) [24] potential phototoxic effects and
3
release of the dye from endosomes were not seen.
If Ru(phen) 21
enters cells by endocytosis it will become
3
trapped in endocytic vesicles. However, when it is free to
enter cells and diffuse throughout the cell interior, it
appears to accumulate in cell nuclei and, to a lesser extent,
in the cytoplasm. Since it is known that Ru(phen) 21
has
3
affinity for nucleic acids it appears most likely that it binds
to nuclear DNA and cytoplasmic RNA. A pattern of
binding in nuclei, which resembles the typical distribution
of heterochromatin and nucleoli, is consistent with this
hypothesis. This binding may also explain a higher local
concentration of Ru(phen) 21
in nuclei than in medium
3
surrounding damaged cells.
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
4.2. Potential use of Ru( II) complexes as cellular
oxygen probes
Potential use of Ru(II) complexes as oxygen probes in
direct contact with living cells will require a compromise
between the relatively high probe concentrations needed to
record a strong luminescence signal and the need to
minimize the dye concentration and light dose to limit
phototoxic effects. Under the experimental conditions used
here, phototoxic effects as expressed by plasma membrane
integrity were undetectable when Ru(phen) 21
concentra3
tion was 2310 24 M. Under these conditions cells could
tolerate a dose of light required to collect |100 confocal
frames, using |0.1-mW beam of excitation light used in a
confocal microscope. Preliminary measurements of
luminescence lifetimes of Ru(II) complexes in single cells
[8] and data presented here indicate that it may be possible
to optimize conditions to non-invasively detect luminescence of Ru(II) complexes in extracellular space or in
endosomes of single cells, and calculate extra- or intracellular concentrations of oxygen.
4.3. Mechanism of phototoxicity
Photosensitizing action of Ru(II) complexes bound to
DNA has been studied extensively. However, in the viable
cell system used here, the dyes remained extracellular and
had no direct contact with DNA. The first observable
phototoxic effect was damage to plasma membranes. The
influence of anoxia, heavy water and histidine suggests
that singlet oxygen was involved in the damage inflicted
on plasma membrane in the presence of Ru(phen) 21
3 .
Indeed, generation of singlet oxygen by illuminated Ru(II)
complexes has been reported [13,36]. Thus, one may put
forward the following hypothesis: upon illumination extracellular and membrane-bound Ru(II) complex generates
singlet oxygen. A local high concentration of singlet
oxygen causes a sequence of events that eventually lead to
plasma membrane damage, which is, in turn, manifested
by a loss of membrane integrity and entry of the dye into
cells.
5. Conclusions
The data presented here and preliminary results reported
previously [8] indicate that Ru(bipy) 21
and Ru(phen) 21
3
3
may be suitable for measurements of extracellular or
endosomal concentrations of oxygen, using FLIM. In case
of macrophages used in this study the concentrations of
Ru(bipy) 21
in the region of 2310 24 M ensured that
3
phototoxic effects were undetectable. At higher concentrations and prolonged illuminations phototoxic effects
occur, resulting probably from the generation of singlet
oxygen.
143
Acknowledgements
The author is grateful to the Foundation for Polish–
German Cooperation (Warsaw), the Wellcome Trust (London), and the Polish State Committee for Scientific Research (KBN, 6 P04A 057 21) for financial support,
Professor H. Acker, Dr N. Opitz (Max-Planck-Institute,
Dortmund) and Dr D. Jackson (University of Manchester)
for helpful discussions, Professor T. Sarna for critically
reading the manuscript, B. Czuba-Pelech, Eng. for excellent technical assistance, K. Kozlowski and I. Kiszka for
performing preliminary experiments, Professor S. Wieckowski for making a fluorimeter available, and Dr T.
Bernas for helpful discussions and editing the figures.
References
[1] T. Alper, in: Cellular Radiobiology, Cambridge University Press,
Cambridge, UK, 1979, pp. 50–70.
[2] R.E. Durand, Keynote address: the influence of microenvironmental
factors on the activity of radiation and drugs, Int. J. Radiat. Oncol.
Biol. Phys. 20 (2) (1991) 253–258.
[3] D.W. Lubbers, Oxygen electrodes and optodes and their application
in vivo, Adv. Exp. Med. Biol. 388 (1996) 13–34.
[4] R.K. Woods, J.W. Dobrucki, J. Glockner, P.D. Morse II, H.M.
Swartz, Spectral-spatial ESR imaging as a method of non-invasive
biological oximetry, J. Magn. Reson. 85 (1989) 50–59.
[5] M.C. Krishna, S. Subramanian, P. Kuppusamy, J.B. Mitchell,
Magnetic resonance imaging for in vivo assessment of tissue oxygen
concentration, Semin. Radiat. Oncol. 11 (2001) 58–69.
[6] H.M. Swartz, R.B. Clarkson, The measurement of oxygen in vivo
using EPR techniques, Phys. Med. Biol. 43 (7) (1998) 1957–1975.
[7] S.J. Dodd, W. Mangay, J.P. Suhan, D.S. Williams, A. Koretsky, Ch.
Ho, Detection of single mammalian cells by high-resolution magnetic resonance imaging, Biophys. J. 76 (1999) 103–109.
[8] H. Malak, J. Dobrucki, M. Malak, H.M. Swartz, Oxygen sensing in
a single cell with ruthenium complexes and fluorescence timeresolved microscopy, Biophysical Society Meeting, Kansas City,
MO, 1998, Biophys. J. 74 (2), A189 (abstr. Tu-Pos252).
21
[9] I. Kiszka, Interaction of Ru(bipy) 3 ion with animal and plant cells,
M.Sc. thesis, Jagiellonian University, Krakow, 1997.
[10] K. Kozlowski, Mechanism of phototoxic action of Ru(phen) 21
in
3
cultures of macrophages, M.Sc. thesis, Academy of Mining and
Metallurgy, Krakow, 1998.
[11] J.R. Lakowicz, E. Terpetsching, Z. Murtaza, H. Szmacinski, Development of long-lifetime metal-ligand probes for biophysics and
cellular imaging, J. Fluorescence 7 (1) (1997) 17–25.
[12] L. Li, H. Szmacinski, J.R. Lakowicz, Long-lifetime lipid probe
containing a luminescent metal–ligand complex, Biospectroscopy 3
(2) (1997) 155–159.
[13] J.N. Demas, D. Diemente, E.W. Harris, Oxygen quenching of
charge-transfer excited states of ruthenium(II) complexes. Evidence
for singlet oxygen production, J. Am. Chem. Soc. 95 (20) (1973)
6864–6865.
[14] A. Tossi, J.M. Kelly, A study of some polypyridylruthenium(II)
complexes as DNA binders and photocleavage reagents, Photochem.
Photobiol. 49 (5) (1989) 545–556.
[15] F.N. Castellano, J. Lakowicz, A water-soluble luminescence sensor,
Photochem. Photobiol. 67 (2) (1998) 179–183.
[16] A.I. Baba, J. Fergusson, B.G. Healey, D. Walt, Oxygen sensing
144
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
J.W. Dobrucki / Journal of Photochemistry and Photobiology B: Biology 65 (2001) 136 – 144
properties of a new ruthenium(II) compound, Anal. Lett. 30 (13)
(1997) 2289–2299.
K. Hashimoto, M. Hiramoto, T. Sakata, H. Muraki, H. Takemura,
M. Fujihira, Time-resolved measurements of luminescence of Ru(II)
complexes chemically or physically immobilized on semiconductor
or insulator particles in various solvents, J. Phys. Chem. 91 (24)
(1987) 6198–6203.
J.R. Bacon, J.N. Demas, Determination of oxygen concentrations by
luminescence quenching of a polymer-immobilized transition–metal
complex, Anal. Chem. 59 (1987) 2780–2785.
M. Kaneko, Photoluminescent copolymer-pendant Ru(bipy 21
3 )
grafted onto non-woven silk fabric and its application to oxygen
sensor, Makromol. Chem., Macromol. Symp. 59 (1992) 183–197.
T. Itoh, K. Yagegashi, T. Kosaka, T. Kinoshita, T. Morimoto, In
vivo visualization of oxygen transport in microvascular network,
Am. J. Physiol. 267 (1994) H2068–H2078.
J.A. Ferguson, B.G. Healey, K.S. Bronk, S.M. Barnard, D.R. Walt,
Simultaneous monitoring of pH, CO 2 , and O 2 using an optical
imaging fiber, Anal. Chim. Acta 340 (1997) 123–131.
R. Baveja, J.X. Zhang, M.G. Clemens, In vivo assessment of
endothelin-induced heterogeneity of hepatic injury perfusion, Shock
15 (3) (2001) 186–192.
M.G. Sasso, F.H. Quina, E.J. Bechara, Ruthenium(II) tris(bipyridyl)
ion as a luminescent probe for oxygen uptake, Anal. Biochem. 156
(1) (1986) 239–243.
H.C. Gerritsen, R. Sanders, A. Draaijer, C. Inee, Y.K. Levine,
Fluorescence lifetime imaging of oxygen in living cells, J. Fluorescence 7 (1) (1997) 11–15.
J.K. Asiedu, J. Ji, M. Nguyen, N. Rosenzweig, Z. Rosenzweig,
Development of a digital fluorescence sensing technique to monitor
the response of macrophages to external hypoxia, J. Biomed. Opt. 6
(2) (2001) 116–121.
F.W.D. Rost, Fluorescence Microscopy, Cambridge University Press,
Cambridge, UK, 1995.
[27] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. von
Zelewsky, Ru(II) polypyridine complexes: photophysics, photochemistry, electrochemistry, and chemiluminescence, Coord. Chem.
Rev. 84 (1988) 85–277.
[28] J. Dobrucki, Z. Darzynkiewicz, Chromatin condensation and sensitivity of DNA in situ to denaturation during cell cycle and
apoptosis—a confocal microscopy study, Micron 32 (2001) 645–
652.
[29] Ch.S. Chow, J.K. Barton, Transition metal complexes as probes of
nucleic acids, Meth. Enzymol. 212 (1989) 219–242.
¨
[30] H. Gorner,
A.B. Tossi, Cz. Stradowski, D. Schulte-Frohlinde,
21
21
Binding of Ru(bpy) 3 and Ru(phen) 3 to polynucleotides and
DNA: effect of added salts on the absorption and luminescence
properties, J. Photochem. Photobiol. B: Biol. 2 (1988) 67–89.
[31] A. Kirsch-De Mesmaeker, G. Orellana, J.K. Barton, N.J. Turro,
Ligand-dependent interaction of ruthenium(II) polypyridyl complexes with DNA probed by emission spectroscopy, Photochem. Photobiol. 52 (3) (1990) 461–472.
[32] C.S. Foote, F.C. Shook, R.B. Abakerli, Characterization of singlet
oxygen, Meth. Enzymol. 105 (1984) 36–47.
[33] I.B.C. Matheson, J.C. Lee, Chemical reaction rates of amino acids
with singlet oxygen, Photochem. Photobiol. 29 (1979) 879–881.
[34] H.A. Crissman, J.A. Steinkamp, Rapid simultaneous measurement
of DNA, protein and cell volume in single cells from large
mammalian cell populations, J. Cell Biol. 59 (1973) 766–771.
[35] Ch.V. Kumar, J. Barton, N.J. Turro, Photophysics of ruthenium
complexes bound to double helical DNA, J. Am. Chem. Soc. 107
(1985) 5518–5523.
[36] S.L. Buell, J.N. Demas, Heterogeneous preparation of singlet
oxygen
using
an
ion-exchange-resin-bound
tris(2,29bipyridine)ruthenium(II) photosensitizer, J. Phys. Chem. 87 (1983)
4675–4681.