|
As the
exploitation of nanoparticles accelerates we are beginning to understand some
of their possible hazards as well as their enormous potential for economic
growth, improved quality of life, and medical diagnosis. However,
nanoparticle research driven by rational applications must also address the
management and elimination of hazardous particles in order to protect our
health and our environment. A word of caution must be noted particularly
because of the seductive quality of some intellectually very attractive
nanoparticles. Nanoparticles with different morphologies made of
non-biodegradable materials provided invaluable information on their
self-assembly and physicochemical properties, but were rarely tested in
vivo. Unfortunately these nanoparticles can constitute threats to living
cells, animals or humans. Working as multidisciplinary teams [1] we
will be better poised to select appropriate materials, modify nanoparticle
surfaces, identify optimal routes of administration, and understand the
pharmacodynamics which will allow a safe use of nanoparticles [2].
Currently, metal-containing nanoparticles, including quantum dots, are still
largely an attractive chemical, pharmacological and medical toolbox but not a
clinical solution for diagnosis or therapy.
Recent reviews
have provided instructive examples of nanotechnology application in basic
neurosciences and its use in addressing interesting biological questions [1-13].
|
Figure 1. Summary of different
classes of nanoparticles and their characteristics
|
3. Nanoparticles
3.1
Nanoparticles
containing metals
Nanoparticles
made of metals e.g. quantum dots, QDs or metallic particles linked with
polymers endows them with unique optical properties. This section reviews
some of their specific photo-physical and optical characteristics which are
exploited for long-term and multicolor imaging [14, 15] and provides examples of
metallic nanoparticles used as research tools in biology, pharmacology and
medicine (Figure 1).
QDs can be
excited at a single wavelength far removed from their emission maxima, which
are tunable by the nanoparticles composition and size. This allows for the
simultaneous detection of multiple color QDs upon illumination with single
light source. QD absorption spectra are broad, but emission spectra are
narrow without the red tail characteristic of organic dyes. QDs are rather
resistant to photo bleaching [4],
which is one of the major drawbacks of currently used fluorescent dyes. The
electron dense QD core allows their detection by electron microscopy.
Combined fluorescence and electron microscopy analyses using the same
multipotent probe provide both temporal dynamics and high-resolution
intracellular localization [16].
Furthermore, QDs have an efficient multiphoton absorption cross-section
making them suitable for in vivo imaging [17]. An
example of a multicolor imaging of live cells with organelle-selective dyes
and QDs is shown in Figure 2. QDs’ wide variety of emissions allows the use
of numerous combinations of fluorescent dyes and QDs for nanoparticle
tracking within cells. QDs that can emit infrared or near infrared light are
particularly suitable for deep tissue imaging because autofluorescence of
hair and tissues in this range is minimal [18, 19].
Several QDs emitting within 650-800 nm are commercially available (e.g. QD
655, QD 705; QD 800) and some of them can be obtained as conjugates with
polyethylene glycol (PEG). Such conjugates seem particularly useful; when
administered subcutaneously, they remain localized at the site of injection
for several days. If administered intravenously, PEG-QDs are not immediately
eliminated by the liver and their protracted circulation time allows their
fate to be followed using different imaging set-ups. Injected QDs with PEG
coating or functionalized QDs can be detected with common in vivo
imaging systems for several days when administered subcutaneously or
intravenously [14-16]. In
vivo monitoring of QDs is appealing because it can provide needed
information on time-dependent QD distribution and accumulation in tissues,
important in the evaluation of potential therapeutic applications. An
accumulation of QDs due to the enhanced permeability and retention (EPR)
effect in tumor tissue could enhance efficiency of phototherapy and, if the
luminescence is satisfactorily high, could coincidentally allow for
monitoring of tumor-size reduction or spread of metastases. Such studies have
been already initiated in several laboratories [20, 21, 22].
Data for biodistribution and pharmacokinetics of QDs are not currently
extensive and more systematic studies are needed to demonstrate how rapidly
these particles can be eliminated from the body, where they accumulate, and
what non-specific tissue damage they may eventually cause.
A variety of
ligands have been attached to QD surfaces, including thiol-containing
molecules
[23, 24]
peptides [25],
polymerized silica shells with polar groups, amphiphilic diblock or triblock
copolymers and phospholipids [26].
For tumor targeting, multiple specific ligands such as folic acid and others
especially suitable for different types of tumors can be conjugated to the QD
surface [27, 28].
Bioluminescent QD conjugates for in
vivo imaging were recently developed by So’s group [19].
Instead of fluorescent excitation light, these QD nanocrystals are exited by
bioluminescence resonance energy transfer and consequently tissue auto
fluorescence is significantly reduced [19]. This
complementary approach using bioluminescent QD conjugates could be especially
useful for small animal imaging.
There appear
to be no studies on the stability of QDs either at the immediate site of
administration or after longer times post injection at remote sites. Due to
its inorganic character the QD core is relatively resistant to photo
degradation, but prolonged exposure to UV light can affect its integrity [29, 30]. In
general QDs with ligands, which are resistant to photo oxidation are better
suited for long-term cellular or in
vivo imaging. Photo oxidation of surface ligands takes place upon
prolonged UV illumination of QDs. However, increasing the thickness and
packing density of the ligand shell can delay the onset of photo oxidation [30].
Taken together, the QD stability is essential to optimize their performance
as imaging tools and to ensure their compatibility with living cells (see
sections 3 and 4 for discussions).
|
|
|
Figure 2. QDs in vitro - principle of cell
labeling
Photomicrograph
of PC12 cells treated with QDs and stained with organelle-specific
fluorescent dyes. Schematic of cell organelles stained with Hoechst 33342, DAF and internalized QDs (red) (A)
Photomicrograph of QDs obtained by confocal microscopy (B)
Plasma membrane (green) labeled with DAF (C) Superimposed
signals from QDs (red) and plasma membrane (green) are seen as yellow
suggesting a presence of QDs associated with plasma membrane (D).
There are no QDs within nuclei (blue).
Abbreviations:
nucleus, NC; cytoplasm, CP; plasma membrane,
PM
|
Another class
of metal-containing nanoparticles includes that made of gold. Gold
nanoparticles have been traditionally used as labels for immunocytochemistry
and in conjunction with electron microscopy they allow detection of selected
proteins with high resolution. In such labeling studies the gold
nanoparticles are integral parts of the secondary antibody. This labeling is
usually achieved by attaching gold nanoparticles of different sizes to IgG
molecules which bind to the primary antibodies. Although this approach can
reveal specific intracellular location of up to three proteins at the same
time, the quantification of the small (5 nm), medium (10-120 nm) and large
(30-50 nm) gold nanoparticles associated with the individual proteins of
choice is rather tedious and therefore is not frequently used. In practice,
single protein gold labeling is one of the “gold standards” in life
sciences/cell biology.
We were
interested in developing gold-labeled micelles that would allow us to assess
the fate of block copolymer micelles in different cells (Figure 3). To this
end we used block copolymer micelles containing covalently bound
nanoparticles.
[31, 32]. A
schematic presentation of such a nanoparticle is given in Figure 1 and Figure
3. An advantage of gold-labeled micelles compared with other nonmetallic
nanoparticles (e.g. silica or styrenes) is that their distribution can be
detected with transmission electron microscopy (TEM), which provides
information about their location within the cells with high resolution. The disadvantage of EM is that the tissues or cells have
to be fixed. Moreover to obtain a dynamic picture of nanoparticle
distribution using TEM is rather time consuming and tedious.
Supraparamagnetic
iron nanoparticles (spions) labeled with fluorescent dye Cy5.5 (Figure 1 and
Figure 3) [33]
allow dual imaging. Live cells are imaged by confocal microscopy to gain a
dynamic picture of cell-nanoparticle interaction and fixed cells are imaged
by TEM to gain information on nanoparticle subcellular distribution at the
high level of resolution. Thus, iron and QD nanoparticles serve as
complementary tools to standard fluorescent dyes and provide a versatile
means of assessing their fate both by confocal and electron microscopy [14, 26, 34].
Spions without conjugated fluorescent dyes have been examined in vivo [35, 36].
|
|
|
Figure 3. Types of nanoparticles
and their location in cells
Gold and superparamagnetic iron nanoparticles in the cells.
Transmission electron micrograph of gold nanoparticle (A) and their size
distribution (B) Gold nanoparticles in lysosomes and endosomes (C)
Iron nanoparticles labeled with Cy5.5 taken up by cells can be visualized
by confocal microscope (D and E). Schematic of a
functionalized spion (F). (For details see ref 31 and
33
Abbreviations: ES-endosome,
VS-vesicle, LS-lysosome,
NM-nuclear membrane.
|
|
|
|
|
|
Figure 4. Relationship
between QD-induced stress,
duration, intensity and location
of signal and cell fate
Different QDs exert different degree of toxicity. QD655 coated
with polyethylene glycol and CdSe/ZnS with multiple layers of ZnS are non
toxic in PC12 cell cultures (A). Different cell types are
differently sensitive towards QD treatments. Human SH-SY5Y neuroblasoma
cells are more susceptible to QD treatment than rat pheochromocytoma cells
(PC12), human breas cancer cells
(MCF-7) and human lung cells (A549) (B). Among the nanoparticles, micelles (Mic) and cross-linked micelles do
not cause cell toxicity when incubated for 24 hours (PC12 cells, 24 hours) whereas cadmium chloride (200nM) is
highly toxic (C).
|
Fluorescent nanoparticles
To follow the
fate of a nanoparticle alone or nanoparticle containing fluorescent dyes or
drugs, several groups have used fluorescent labeled polymers in combination
with organelle selective fluorescent dyes [12, 37-40].
Although, fluorescent dyes are useful for visualizing micelles at selected
time points, the tracking of micelles, particularly in time-lapse experiments
using confocal microscopy is considerably more difficult because of dye
bleaching.
Relatively
little is known about the fate of fluorescent nanoparticles in the whole
animal. One of the reasons for this is that the quantities required for in vivo imaging are significantly
larger than those used for in vitro
experiments and some fluorescent dyes are expensive. Moreover the synthesis
of fluorescent polymers is not a trivial matter, and some dyes simply cannot
be conjugated to the polymer. Additional problems include the
autofluorescence of the tissue and a limited access to suitable
instrumentation for in vivo imaging. Finally, a factor complicating
the determination of the fate of biodegradable block copolymer micelles in
vivo is the interference associated with components from blood and other
biological molecules plus the polymer degradation products within individual
organs and cellular compartments. One of the first studies designed to assess
the fate of fluorogenic dyes incorporated into micelles in complex biological
media was recently reported [41].
The findings showed a gradual increase in fluorescence at the site of micelle
decomposition, which is due to the conversion of the fluorogenic dye to the
fluorescent product. Fluorogenic dyes, which could yield a product emitting
in the near-infra region, would be very useful in the quest for new
information on micelle fate in vivo.
The problem of
micelle stability can be partly circumvented by cross-linking the micelle
shell. However, complete control of drug release from such micelles becomes
an issue to be resolved for each individual drug. A significant contribution
in development and utilization of
shell-cross-linked nanoparticles for tumor imaging is has been made by
Wooley’s group [42].
Shell cross-linked knedel-like (SCKs) nanoparticles are core-shell
nanospheres inspired by biological structures (e.g. virus capsids) and
optimized for drug packaging. Stable SCKs are obtained by cross-linking the
shell layers of micelles made of amphiphilic diblock copolymers. The
cross-linking procedures yield SCKs with well-defined sizes, surface charges,
shapes and functionalities. SCKs can be labeled with fluorescent, radioligand
or dense core particles and decorated with different ligands for cell or
tissue targeting [43-46].
In summary,
using both fluorescently labeled agents and fluorescent polymers one can
distinguish micelle-incorporated agents from free agents, and obtain
information about their intracellular location [40].
Studies using new imaging approaches and tools are already beginning to
emerge [41, 47-51]. It
seems certain that the use of nanoparticles with high
fluorescence/luminescence in combination with advanced imaging equipment will
soon provide needed information relevant to the in vivo condition.
4. Cell death induced by nanoparticles
4.1. Nanoparticle properties and cell death
Once
cells internalize nanoparticles they often undergo degenerative
changes and eventually die. This is independent of the
mechanism of their entry (e.g. non-specific endocytosis [4] or
specific endocytosis [52]). Micelles with polyethylene
glycol (PEG) corona, purified to satisfy the stringent requirements for work
with biological systems, are mostly non toxic in low concentrations [12, 40]. In contrast, polyethylene
imines are often toxic (PEI) [2]. Non-functionalized
QDs mostly end up in the cytoplasm. However, functionalized QDs with the
Nuclear Localization Signal (NLS) peptide or the mitochondrial localization
signal peptide enter the nucleus and mitochondria [53].
Peptide modifications on QD surface make them relatively well tolerated by
most cells.
The toxicity
of QDs when tested in vitro, is
dependent on numerous factors associated with QD properties and cell status.
QD properties, which are critical for their biocompatibility, are: size,
surface charge, type of ligand attached to the surface, stability, and
concentrations. Properties determining the cell status are associated with
the cells themselves and their environment. Thus, the fate of cells exposed
to nanoparticles will depend on: their tissue of origin and differentiation
state, their proliferative status (doubling time) and their redox state. In
addition to this, plating cell density, composition of the medium, presence
and absence of serum or defined trophic factors in chemically defined media will
be also critical to the final outcome. The combination of these physical and
chemical factors of QDs, the cellular environment, and the status of the
cells determine cell fate: survival, death or differentiation (Figure 5).
Depending on
the kind of nanoparticles, the type of cells, the duration of exposure, the
concentration of nanoparticles and the conditions under which
cell-nanoparticle interaction takes place, different cells will either
“tolerate” nanoparticles or succumb to their “attack”. In a brief contact
with cells, particles with surfaces well-protected by polymers or proteins,
are generally not very harmful or are totally benign [15].
However, the breakdown of protective coating (e.g. by irradiation, low
lysosomal pH or possibly metabolic degradation) can lead to cell damage and
death [29]. The cellular
environment plays a critical role in influencing physiological status of
cells and has an impact on their response to QDs. For example, cells exposed
to QDs are more vulnerable when deprived of growth factors, whereas cells
cultured in the presence of serum will be more resistant to QDs induced cell
death. Moreover, certain types of QDs can interact with serum and other
biological fluid components reducing or increasing their damaging effects on
cells. In general, short-term exposure (1-4 hours) and low concentrations
(nanomolar range) of nanoparticles is well tolerated by living cells. Under
certain circumstances, nanoparticle degradation “on command” could be
desirable and a number of different types of such nanoparticles have been
prepared and tested. (e.g. nanoparticles responding to changes in pH, in
temperature or in redox potential). Significant advances have been made in
this field and a number of biosensors based on QDs and other nanoparticle
structures exist that can respond to various pathological or physiological
stimuli [54-56].
4.2. Types of cell death and roles of
individual organelles
Cells can die
by apoptosis, autophagy or by necrosis (“accidental but aggressive” death) [57].
The mode of cell death depends on the type of insult, its intensity and
duration. The most common mechanisms associated with nanoparticle-induced
cell death are apoptosis
and necrosis. For instance, the long-term exposure to
unprotected CdTe QDs, which are internalized by cells, will cause damage at multiple
cellular sites [58].
Cells in a nutrient enriched medium and in a low metabolic state will generally
handle small insults well and once the nanoparticles are removed they will
recover. In contrast, cells under starvation/serum deprivation conditions
will be sensitized to the QD insult and will die by apoptosis, autophagy
and/or necrosis. Looking at morphological changes during QD-induced cell
death one can often notice a transition from one mode of death to the other
(e.g. apoptosis to delayed necrosis).
Programmed
cell death is most frequently associated with the term “apoptosis”, but there
is emerging evidence for “programmed necrosis”. Several recent reviews cover
different aspects of classical and “nonclassical” types of cell death [59, 60] and
provide an overview of consensus terms to be used in research associated with
cell death and dying [61]. A
brief account of the damaging effects of some nanoparticle types (e.g. QDs)
on different organelles is provided in the following section.
Mitochondria play a
central role in energy metabolism and they are responsive to even small
stresses in multiple ways [62].
When cells are exposed to nanoparticles (e.g. QDs) which can induce
generation of ROS, mitochondria are among the first and most sensitive
organelles affected. In QD-insulted cells, reduction of mitochondrial
membrane potential and swelling of mitochondria can be detected |