Drug delivery monitoring and tracking in the human body are two of the biggest challenges in targeted therapy to be addressed by nanomedicine. The ability of imaging drugs and micro-/nanoengineered drug carriers and of visualizing their interactions at the cellular interface in a label-free manner is crucial in providing the ability of tracking their cellular pathways and will help understand their biological impact, allowing thus to improve the therapeutic efficacy. We present a fast, label-free technique to achieve high-resolution imaging at the mid-infrared (MIR) spectrum that provides chemical information. Using our custom-made benchtop infrared microscope using a high-repetition-rate pulsed laser (80 MHz, 40 ps), we were able to acquire images with subwavelength resolution (0.8 × λ) at very high speeds. As a proof-of-concept, we embarked on the investigation of nanoengineered polyelectrolyte capsules (NPCs) containing the anticancer drug, docetaxel. These NPCs were synthesized using a layer-by-layer approach built upon a calcium carbonate (CaCO3) core, which was then removed away with ethylenediaminetetraacetic acid. The obtained MIR images show that NPCs are attached to the cell membrane, which is a good step toward an efficient drug delivery. This has been confirmed by both three-dimensional confocal fluorescence and stimulated emission depletion microscopy. Coupled with additional instrumentation and data processing advancements, this setup is capable of video-rate imaging speeds and will be significantly complementing current super-resolution microscopy techniques while providing an unperturbed view into living cells.
Drug delivery monitoring and tracking in the human body are two of the biggest challenges in targeted therapy to be addressed by nanomedicine. The ability of imaging drugs and micro-/nanoengineered drug carriers and of visualizing their interactions at the cellular interface in a label-free manner is crucial in providing the ability of tracking their cellular pathways and will help understand their biological impact, allowing thus to improve the therapeutic efficacy. We present a fast, label-free technique to achieve high-resolution imaging at the mid-infrared (MIR) spectrum that provides chemical information. Using our custom-made benchtop infrared microscope using a high-repetition-rate pulsed laser (80 MHz, 40 ps), we were able to acquire images with subwavelength resolution (0.8 × λ) at very high speeds. As a proof-of-concept, we embarked on the investigation of nanoengineered polyelectrolyte capsules (NPCs) containing the anticancer drug, docetaxel. These NPCs were synthesized using a layer-by-layer approach built upon a calcium carbonate (CaCO3) core, which was then removed away with ethylenediaminetetraacetic acid. The obtained MIR images show that NPCs are attached to the cell membrane, which is a good step toward an efficient drug delivery. This has been confirmed by both three-dimensional confocal fluorescence and stimulated emission depletion microscopy. Coupled with additional instrumentation and data processing advancements, this setup is capable of video-rate imaging speeds and will be significantly complementing current super-resolution microscopy techniques while providing an unperturbed view into living cells.
Optical spectroscopy
techniques such as near-infrared (NIR), Raman,
and Fourier Transform Infrared (FTIR) can potentially revolutionize
disease diagnosis and therapy as an alternative or adjunct tool providing
early diagnosis, surgical boundary identification, and therapeutic
response monitoring.[1−4] Novel treatment strategies involving personalized medicine and nanomedicine
approach can be effectively monitored using these techniques. In particular,
the convergence of pharmaceutical technology with nanotechnology in
the field of nanomedicine offers a wealth of nanoparticles for therapeutic
drug delivery[5] that can overcome biological
barriers. Imaging the interface between biology and the nonbiological
delivery system is important in providing, for example, the ability
of tracking cellular pathways of drugs and their carriers (micro-
and nanoparticles), monitoring therapeutic responses, especially with
respect to the biological building blocks, predicting antimicrobial
or antibiofilm performances,[6] and finding
means to overcome specific delivery problems in vitro and in vivo.
For example, recent studies show that only 0.7% injected dose of nanoparticles
actually reach the targeted solid tumor in vivo.[7] The bioavailability of many hydrophobic cancer drugs is
also limited and restricts efficient treatment.Understanding
the interaction of drugs and/or drug-loaded carriers
with cellular membrane and subcellular components can unveil the entry
route as well as the subsequent localization of therapeutic agents
within the cells. This can, in turn, provide important information
regarding improvement in the design of drugs, carriers, and internal/external
stimuli to improve therapeutic outcome. This had motivated researchers
to develop techniques that can image biological specimens and track
drug molecules at subcellular levels (say below 50 nm length scales).
Among these techniques, fluorescence microscopy has traditionally
played a key role in biointerface imaging due to the excellent specificity
and contrast that can be achieved. Furthermore, technological advances
in optical instrumentation, coupled with the development of chemistry-specific
fluorescent probes, have allowed for an explosion of methods to study
live cells at the micro- and nanoscales (e.g., confocal, stimulated
emission depletion (STED), stochastic optical reconstruction microscopy
(STORM), photo activated localization microscopy (PALM), structured
illumination microscopy (SIM)).[8−10] Fluorescence-based methods, however,
suffer from a major drawback, which is the absolute requirement for
labeling the features of interest, by using either fluorescent-labeled
antibodies against specific proteins within the cell or probes that
exhibit fluorescence upon binding to certain molecules.[11] A direct consequence of labeling is the introduction
of artifacts that can possibly alter the physiology of cells under
investigation and are not able to reveal direct biochemical processes
in real time due to photobleaching. The utilization of fluorophores
is not always possible; hence, there is a need for the development
of label-free techniques that can be used complementarily with fluorescence
microscopy and are capable of real-time chemical and structural imaging
under ambient conditions and high resolution.Spontaneous Raman
microscopy is a noninvasive technique that probes
the intrinsic vibrational signatures of molecules, allows label-free
imaging, circumvents the need for fluorescent or other extrinsic tags,
and permits the visualization of the distribution of specific molecules
with high sensitivity and specificity.[12] Unfortunately, despite the efforts conducted to develop Raman microscopy,
low signal levels limit this technique; hence, either a large number
of molecules or long acquisition times are required, presenting significant
limitations to the development of real-time studies.Over the
last 15 years, microscopy approaches using coherent Raman
scattering (CRS) as a contrast mechanism had emerged as powerful techniques
to address these limitations. Coherent Raman imaging, including stimulated
Raman scattering (SRS) and coherent anti-Stokes Raman scattering (CARS),
allows the enhancement of the weak Raman signal by means of nonlinear
excitation 5 orders of magnitude over traditional Raman scattering,[13] enabling the speeding up of the imaging speeds
to video rates.[14,15] However, the small penetration
depths due to the use of the NIR lasers, the optical photodamage due
to the two-photon absorption process, and the poor detection limit
(millimolar to micromolar) have limited the application of CRS techniques
mostly to in vitro applications.[16]Mid-infrared (MIR) microscopy is a powerful label-free technique
that derives its contrast from the intrinsic absorption of the biomolecules
at their vibrational fingerprint range and can provide a technical
solution to these limitations.[17−19] However, MIR instrumentation
stagnated due mostly to the trade-off between spectral-spatial resolution
and acquisition times. This has motivated researchers to look for
alternative IR sources. To overcome the low brightness of Globar (thermal)
sources, synchrotron sources (SRs) were used for FTIR microspectroscopy.
The high brightness and stability of broadband SRs made them excellent
for MIR imaging, and their coupling with focal plane array (FPA) detectors
has greatly expanded the capability of infrared microscopes in terms
of high rates of spectral acquisition at high signal-to-noise ratios.
The use of multiple SR beams in a wide-field detection scheme allowed
one to acquire diffraction-limited images at high spatial resolution
and high speed, which has considerably extended the potential of infrared
microscopy.[20,21] However, the limited access and
the high cost associated with the use of SRs restricted their use
and made them unpractical.Recently, new sources like quantum
cascade lasers (QCLs) have emerged
as an alternative to thermal or synchrotron sources,[22−24] and they are rapidly becoming practical mid-infrared sources for
a variety of applications, such as trace-chemical sensing and health
monitoring. However, their small tunability ranges (few micrometers)
limit their use to exciting one or two vibrational modes. Thus, a
good number of QCLs (half a dozen) are needed to exploit the whole
fingerprint spectrum of biological specimens that span from 500 to
4000 cm–1, corresponding to a range between 2.7
and 16.7 μm. By assuming a linear scaling factor, then the acquisition
time will be increased by another order of magnitude.[24] One way to overcome the tunability issue is by exploiting
the advantages of nonlinear optics to generate MIR laser beams that
cover the whole fingerprint spectrum.Benchtop MIR microscope
is still diffraction limited, with a resolution
that at best equals the used wavelength (λ),[25] whereas the use of synchrotron radiation with modifications
can extend this limit to 1/2λ, corresponding to a pixel size
of 0.54 μm × 0.54 μm.[20] For instance, it takes 2–4 h to acquire an area of only 30
μm × 30 μm as a fully diffraction-limited image using
a synchrotron source[26] equipped with a
conventional confocal system. Longer acquisition times imposed by
the MIR sources, in most practical cases, lead experimenters to choose
larger aperture and step sizes, thereby compromising the achievable
spatial resolution.[24] Thus, the development
of powerful, fast, and tunable MIR laser sources for microscopy will
push forward IR imaging as a viable tool for disease diagnostics and
medical research and provide real-time chemical and structural imaging
under ambient conditions.[27] Furthermore,
the implementation of a confocal microscope will offer high-resolution
imaging and potential three-dimensional (3D) imaging.In this
work, we present a high-speed, benchtop confocal MIR microscope
capable of providing label-free chemical imaging at subwavelength
spatial resolution. Our instrument is based on a high brightness compact
fiber laser source with a repetition rate of 80 MHz and 40 ps pulse
duration. A state-of-the-art fiber laser is used to pump a synchronously
pumped optical parametric oscillator (SP-OPO) built around a fan-out
MgO:PPLN crystal generating a signal and idlers beams varying from
1.5 to 2 μm and from 2 to 4.8 μm, respectively. The SP-OPO
outputs are synchronously mixed inside a nonlinear crystal in a difference
frequency generation (DFG) stage to generate a tunable mid-IR beam.
The tunability of this type of lasers in the mid-IR range depends
on the type of the nonlinear DFG crystal. Thus, our laser is tunable
up to 8 μm using a CdSe crystal and has the capability of extending
its tunability up to 20 μm using a AgGaS2 crystal.
The ability of tuning the laser source throughout the entire MIR range
paves the way that allows for a multiwavelength imaging of biological
samples, akin to multichannel imaging in fluorescence microscopy.
Such a development has an immense potential for therapeutic monitoring
and in vitro/in vivo diagnosis, including breast cancer, for which
an early detection and treatment with constant monitoring are required.As a proof-of-concept, we first performed in vitro imaging of docetaxel
(DTX)-loaded NPCs composed of synthetic polyelectrolytes.[28−30] Namely, the polyanion, poly(styrene sulfonate) (PSS), and the polycation,
poly(allylamine hydrochloride) (PAH), were used for the fabrication
of a nanoengineered multilayered shell onto the surface of sacrificial
calcium carbonate (CaCO3) microparticles loaded with the
chemotherapeutic drug, docetaxel.[31]NPCs are seen as very promising for a broad range of applications
in nanomedicine, such as drug delivery and biosensing, due to their
tunability and easy fabrication.[32−34] In this respect, the
possibility of their tracking and visualization in a label-free manner
could be useful to understand their interaction with cells and with
intracellular organelles,[35] in the view
of perspective clinical applications. Recently, it has been demonstrated
that NPCs can be functionalized with molecules as large as a membrane
protein complex (ba3-cytochrome c oxidase; ∼90
kDa) without perturbing their function.[36] Herein, we characterized the in vitro interaction of such NPCs,
loaded with docetaxel, with cancer cells by combining confocal and
super-resolution (STED) fluorescence microscopies. These two techniques
were used to validate the permeability variation of the NPCs by mainly
looking at the entrance of the dye molecules (rhodamine) from the
environmental solution into the NPC volume.[36]In summary, data presented in this paper show the capabilities
of our custom-made MIR microscope for chemical label-free imaging
at high speeds and subwavelength resolution and high throughput imaging
as well as its great potential in biomedical sciences. By adapting
more appropriate instrumentation (fast lock-in amplifiers, electro-optical
modulators), imaging at video rates will be possible. This will pave
the way for real-time monitoring of drug delivery and for understanding
its biological impact, which will support drug discovery and enhance
the therapeutics efficacy.
Results and Discussion
Synthesis of (PSS/PAH)
NPCs
NPCs represent an excellent
carrier for the encapsulation of drugs.[37] In this study, NPCs loaded with DTX were fabricated by the alternate
deposition of PSS and PAH onto the surface of docetaxel-preloaded
CaCO3 microparticles. A range of microscopic techniques
was applied to ensure the integrity of functionalized NPCs. Figure shows the scanning
electron microscopy (SEM) and optical images of CaCO3 cores
as well as the obtained (PSS/PAH)4 hollow NPCs. The SEM
images show clearly that the NPCs collapsed after the sample was dried.
Further coating of the NPCs with different moieties allows the NPCs
to find more applications in biosciences and provides important properties
to the NPCs such as (a) prolonged and sustained release (polymers),
(b) magnetic nanoparticles (targeting, hyperthermia), (c) fluorescent
nanoparticles and dyes (visualization), and others.
Figure 1
(Left) SEM images of
CaCO3 cores used for the creation
of PSS/PAH NPCs and (middle) (PSS/PAH)4 NPCs after the
removal of CaCO3 core.The NPCs collapsed after the sample
was dried. (Right) Corresponding wide-field optical image of PSS/PAH
NPCs, which display a typical size distribution of about 1–6
μm. Scale bar 20 μm.
(Left) SEM images of
CaCO3 cores used for the creation
of PSS/PAH NPCs and (middle) (PSS/PAH)4 NPCs after the
removal of CaCO3 core.The NPCs collapsed after the sample
was dried. (Right) Corresponding wide-field optical image of PSS/PAH
NPCs, which display a typical size distribution of about 1–6
μm. Scale bar 20 μm.
FTIR Spectroscopy
To facilitate the identification
and differential imaging of empty and DTX-loaded NPCs, we acquired
their FTIR absorption spectra to ensure that at least some spectral
differences were present due to the presence of DTX. Both empty and
DTX-loaded NPCs were kept in their solution state, and their FTIR
spectra in the range between 1000 and 4000 cm–1 were
acquired in attenuated total reflection (ATR) configuration. As shown
in Figure , a significant
difference was apparent in the region of 3000–3500 cm–1 (3.3–2.9 μm), indicating the successful encapsulation
of DTX in NPCs. We can see two distinct bands at 3200 and 3350 cm–1 that can be used to image both types of NPCs by tuning
our MIR pulsed laser to excite these vibrational bands.
Figure 2
FTIR spectra
of empty and DTX-loaded NPCs acquired in the ATR configuration
in the range 1000–4000 cm–1 with 0.5 cm–1 step.
FTIR spectra
of empty and DTX-loaded NPCs acquired in the ATR configuration
in the range 1000–4000 cm–1 with 0.5 cm–1 step.
Fast, High-Resolution Label-Free MIR Imaging
Imaging of NPCs
We initially acquired the images of
a calibration sample (silver thin film with microsized holes deposited
on a microscope slide) to calibrate our setup and to determine the
effective parameters and capabilities of our instrument such as imaging
speed, resolution, and tunability. The obtained data show the resolution
of our microscope to be 0.8λ, indicating the subwavelength resolution
imaging capability of our microscope (Figures S1 and S2). Then, empty PSS/PAH NPCs were dispersed on CaF2 coverslips and imaged using our MIR confocal microscope in
transmission mode. CaF2 coverslips were used as they are
transparent in the MIR region. The obtained MIR images are shown in Figure . The nonspherical
shape indicates that the NPCs collapsed after the sample was dried,
which is in good agreement with SEM data. The full width at half maximum
of the profile along the capsule is used to determine the size of
the NPCs.
Figure 3
MIR image imaging of empty (PSS/PAH)4 NPCs. (a) MIR
image taken at a wavelength of 3.125 μm corresponding to a vibration
at 3200 cm–1, image size 50 μm × 50 μm
and (b) 20 μm × 20 μm zoom corresponding to the green
square. The corresponding line profile (c) shows the size of the capsule
to be 4 μm. All images are 256 × 256 pixels and have been
acquired at 2 ms pixel/dwell time.
MIR image imaging of empty (PSS/PAH)4 NPCs. (a) MIR
image taken at a wavelength of 3.125 μm corresponding to a vibration
at 3200 cm–1, image size 50 μm × 50 μm
and (b) 20 μm × 20 μm zoom corresponding to the green
square. The corresponding line profile (c) shows the size of the capsule
to be 4 μm. All images are 256 × 256 pixels and have been
acquired at 2 ms pixel/dwell time.
MIR Imaging of Fixed Cells Incubated with DTX-Loaded NPCs and
Dual-Channel Imaging
After demonstrating the proof-of-concept
and capabilities of our microscope on a calibration sample and
NPCs (Figures S1 and S2), this was then
extended to an in vitro imaging of MCF7 cells incubated with 1 mg/mL
DTX-loaded NPCs for 48 h and then fixed in formaldehyde. To image
the MCF cells, we tuned our laser to 3.5 μm to excite the lipid
stretch at 2850 cm–1. A control sample was used
for comparison. We can see clearly that the IR intensity is homogeneous
throughout the control sample; however, the MCF7 cells incubated with
DTX-loaded NPCs show the presence of more intense features closely
attached to the cell membrane (Figure ).
Figure 4
MIR image of MCF7 cells incubated with DTX-loaded NPCs.
(Left)
Control cells without any NPCs and (middle) cells incubated with 200
μL of DTX-loaded NPCs at a concentration of 1 mg/mL for 48 h.
The circle on the right (bright white spots) indicates potential aggregation
of NPCs. Both images were obtained at 2850 cm–1 (λ
= 3.5 μm).
MIR image of MCF7 cells incubated with DTX-loaded NPCs.
(Left)
Control cells without any NPCs and (middle) cells incubated with 200
μL of DTX-loaded NPCs at a concentration of 1 mg/mL for 48 h.
The circle on the right (bright white spots) indicates potential aggregation
of NPCs. Both images were obtained at 2850 cm–1 (λ
= 3.5 μm).Because both NPCs and
cells membranes have a high absorption peak
around 2850 cm–1 as shown in the FTIR spectra, these
features are potential aggregates of DTX-loaded NPCs. Images are 256
× 256 pixels obtained at very high speed with a dwell time of
2 ms/pixel, resulting in a total acquisition time of 131 s. We note
that the imaging rate was limited by the acquisition parameters of
the lock-in amplifier (time constant = 1 ms) and the beam modulation
(0.5 Hz). Higher speeds with 1 μs pixel dwell time are possible
using fast lock-in amplifiers and electo-optical modulators, thus
allowing imaging rates 3 orders of magnitude higher than the current
speeds.As our laser is monochromatic and only one wavelength
can be used
at the time, revealing the chemical contrast between different biological
specimens can be achieved by wisely selecting two or three wavelengths
and recording the images in sequence. Our laser features a mechanical
tunability, allowing thus instantaneous change to the desired wavelength.
This multiwavelength approach and fast tunability (Figure S2) represent an alternative solution to the slow broadband
imaging and permit chemical imaging at different frequencies and will
considerably accelerate the chemical imaging mode while conserving
subwavelength resolution. To achieve this, we tuned our laser to two
specific wavelengths of 3.4 and 2.95 μm to image the cells and
DTX-loaded NPCs, respectively. These wavelengths correspond to the
C–H (2940 cm–1) and N–H (3400 cm–1) bonds in cells and DTX-loaded NPCs, respectively.
The obtained images are presented in Figure . The two images were overlaid to obtain
a label-free dual-color image showing the attachment of the DTX-loaded
NPCs (in green) to the cell membranes (in red). The obtained image
is of the same quality as the confocal fluorescence images, yet without
any labeling.
Figure 5
Dual-color high-resolution imaging of MCF7 breast cancer
cell line
incubated with DTX-loaded NPCs. The MIR laser was tuned to 3.4 μm
(C–H bond) to image cells (red) and to 2.7 μm (N–H
bond) to image NPCs (green). The overlay of the two images (right)
shows the attachment of the DTX-loaded NPCs to the cell membranes.
Image size is 256 × 256 pixels acquired with 2 ms pixel dwell
time, giving a total acquisition time of 131 s. Scale bar 20 μm.
Dual-color high-resolution imaging of MCF7 breast cancer
cell line
incubated with DTX-loaded NPCs. The MIR laser was tuned to 3.4 μm
(C–H bond) to image cells (red) and to 2.7 μm (N–H
bond) to image NPCs (green). The overlay of the two images (right)
shows the attachment of the DTX-loaded NPCs to the cell membranes.
Image size is 256 × 256 pixels acquired with 2 ms pixel dwell
time, giving a total acquisition time of 131 s. Scale bar 20 μm.Our MIR microscope features a
dual-analogue data acquisition interface
(DAQ, Physik Instrument), which allows us to record only two-dimensional
images. Thus, we were not able to record 3D images as in confocal
microscopy and to find out whether the NPCs were attached to the outer
side of the cell membranes or have crossed the cell membrane. To validate
our MIR findings, we used the more established confocal and STED microscopies.
Confocal and STED Microscopy of Living Cells
To assess
the effect of empty and DTX-loaded NPCs on living cells, we embarked
on a series of more established confocal and STED microscopies. As
both techniques require exogenous labeling, and to visualize both
cells and NPCs, we have consequently chosen two fluorescent probes
with distinct peak excitation/emission wavelength pairs (i.e., emission
of one probe not overlapping with the excitation of another). Thus,
cells were stained with Hoechst 33342 (Invitrogen, Paisley, U.K.),
a dye emitting around 450 nm, to visualize the nuclei, whereas the
four layers of PAH/PSS NPCs were labeled using covalently functionalized
PAH-ATTO 590 at the third layer. Atto 590 (ATTO-TEC GmbH, Siegen,
Germany) is a dye that has its absorption peak at 593 nm and an emission
peak at 622 nm. Figure shows time-lapsed imaging taken at 0, 2, and 4 h after incubation
with empty and DTX-loaded NPCs. A control sample was used to assess
the cytotoxicity of the NPCs. The obtained images show that even after
4 h, the control sample shows healthy cells (Figure , panels a, d, and g); however, cells incubated
with empty NPCs start to die after 2 h, and almost all of the cells
were dead after 4 h. We observed that the toxicity effect suggested
by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) assay (Figure S3) is confirmed microscopically,
where high concentrations of (PSS/PAH)4 NPCs have a toxic
effect and already exhibit some cell killing properties (Figure , panels b, e, and
h). Furthermore, the presence of DTX accelerated cell death, indicating
that the functionalization of NPCs does not alter the properties of
DTX (Figure , panels
c, f, and i). This finding suggests that these hollow NPCs can be
used as therapeutics carriers, and further functionalization of their
shell will make them very promising for targeted therapy.
Figure 6
Wide-field
fluorescence and transmission time-lapsed imaging of
living HeLa cells in culture at different time points. Control cells
without NPCs are shown in the panels (a), (d), (g) whereby cells behave
normally, whereas those with empty and drug-loaded NPCs exhibit cytotoxic
effects (panels (b, e, h) and (c, f, i), respectively). The gray images
were acquired in the transmission mode. The cell nuclei and NPCs were
labeled with Hoechst dye targeted against DNA (blue) and Atto 590
(red), respectively.
Wide-field
fluorescence and transmission time-lapsed imaging of
living HeLa cells in culture at different time points. Control cells
without NPCs are shown in the panels (a), (d), (g) whereby cells behave
normally, whereas those with empty and drug-loaded NPCs exhibit cytotoxic
effects (panels (b, e, h) and (c, f, i), respectively). The gray images
were acquired in the transmission mode. The cell nuclei and NPCs were
labeled with Hoechst dye targeted against DNA (blue) and Atto 590
(red), respectively.Then, the shape and morphology of the NPCs when exposed onto
the
cells were checked by both 3D confocal and STED microscopies, as shown
in Figure . We can
see clearly that the NPCs attach to the cell membrane, which is a
good step toward DTX delivery, by forcing the rupture of the capsule’s
shell, allowing thus an efficient delivery of DTX to the cells. The
NPCs do not adopt a position optimized enough for a timed, targeted
release of correct dosage. More work needs to be performed to optimize
the delivery process by controlling the opening of the NPCs as well
as their attachment to the cell membrane. In an ideal case, it would
be preferable for an individual capsule to contact a single cell or
at least reproducibly if the ratio is not 1:1. The optical and SEM
images of the capsule already indicate that a high monodispersity
could not be achieved within our experimental condition. Together
with advanced drug release modeling and pharmacokinetics, we anticipate
these parameters to be further optimized.
Figure 7
Multicolor 3D-STED images
of HeLa cells treated with DTX-loaded
NPCs. In panel (a), the cells were imaged as a single xy section with microtubules labeled with ATTO 647N-labeled antibody.
The corresponding (dotted line) xz section is shown
in panel (b), where the two ATTO 590-labeled NPCs are visible.
Multicolor 3D-STED images
of HeLa cells treated with DTX-loaded
NPCs. In panel (a), the cells were imaged as a single xy section with microtubules labeled with ATTO 647N-labeled antibody.
The corresponding (dotted line) xz section is shown
in panel (b), where the two ATTO 590-labeled NPCs are visible.These data are only the beginning
of a larger series of investigation
into label-free biological samples that can be exploited by our instrument.
As our view of biology has thus far been mostly shaped by fluorescence-based
microscopy techniques, label-free super-resolution MIR microscopies
herald a new era whereby commercial probe availability will no longer
restrict interesting studies. We expect MIR microscopy to be initially
subject to complementary imaging for cross-validation and multimodality-based
understanding into unique sample environments.
Conclusions
We have developed a benchtop MIR microscope based on high repetition
and fast tunable OPO laser. We used our custom-made microscope to
image MCF7cancer cells incubated with PSS/PAHDTX-loaded NPCs in
a label-free manner. The combination of MIR imaging with super-resolution
methods proposed in this work allows for the investigation of drug
carrier interaction with the cell membrane and the subsequent delivering
of anticancer drugs. The ability of directly visualizing individual
NPCs and cell organelles at subwavelength resolution and at high speeds
without adding extraneous tags will yield crucial information on biochemical
processes like the interaction of drug carriers (NPCs) with cells
as well as intracellular delivery of drugs. Complementing MIR with
super-resolution microscopy will provide more information about the
interaction of nanoparticles with cells, the intracellular trafficking
of drugs, their localization within the cells, and their subsequent
therapeutic effect. In this framework, the multimodality IR with super-resolution
microscopy represents a powerful tool toward the in vitro monitoring
of therapeutics delivery, as well as translating this work toward
in vivo applications.
Materials and Methods
Polyelectrolyte Capsule
Synthesis and FTIR Spectroscopy
Cationic PAH (Mw 70 kDa, Sigma-Aldrich)
and anionic PSS (Mw 70 kDa, Sigma-Aldrich)
were used as synthetic polyelectrolytes for shell deposition. NPCs
were fabricated following a well-established procedure.[28−30,36] Specifically, four polyelectrolyte
bilayers were deposited onto DTX-preloaded (1% w/v) CaCO3 microparticles (6 μm in diameter) (PlasmaChem GmbH). Ethylenediaminetetraacetic
acid (EDTA) (Sigma-Aldrich) was used as a complexing agent for the
removal of the CaCO3 template. PSS and PAH were prepared
in pure water at a concentration of 2 mg/mL, pH 6.5, and their adsorption
time was 10 min. Four bilayers were deposited onto the surface of
the particles; after each deposition step, the dispersion of the covered
particles was centrifuged (2500 rpm for 5 min) and the precipitated
covered particles were separated from the solution. These particles
were washed three times in pure water. The CaCO3 cores
were then dissolved by their dispersion in 0.1 M EDTA, pH 5, followed
by three washings in pure water and redispersed in 0.1 M phosphate
buffered solution (PBS) at pH 7.4.FTIR spectra of empty and
DTX-loaded NPCs were acquired in the ATR configuration (Perkin-Elmer
Spectrum 100) in the solution state without using the crystal, by
averaging 100 individual spectra for each. The instrument was blanked
with PBS. Both optical microscopy and scanning electron microscopy
(SEM) were used to determine the size of individual particles and
their integrity.
Cell Culture and Incubation with NPCs
MCF7 breast cancer
cells and HeLa cervical cancer cells were cultured in Dulbecco’s
modified Eagle’s medium with fetal bovine serum and 1% penicillin–streptomycin
at 37 °C under a 5% CO2 atmosphere. Before delivery,
NPCs were sterilized through UVC irradiation (λmax = 254 nm) as installed in a biological safety cabinet for 2 h. The
integrity of NPCs before and after UV sterilization was checked using
UV–visible spectroscopy. The obtained data showed that UV irradiation
of NPCs did not induce any structural changes, which was also confirmed
through confocal laser scanning microscopy of NPCs loaded with rhodamine.
This observation is in contrast to previously reported shrinking of
the polymeric shell.[38] Then, we performed
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)
tetrazolium reduction assay[39,40]—known for assessing
cell metabolic activity and cytotoxicity—to determine their
degree of cytotoxicity (see the Supporting Information, Figure S3a,b).
MIR Microscopy
For label-free MIR imaging, a custom-made
tabletop MIR confocal microscope was used. The microscope is equipped
with a compact and high-repetition-rate (80 MHz) picosecond (30 ps)
optical parametric oscillator (OPO) (Laserspec, Belgium) pumped with
a picosecond fiber laser (40 ps, 1030 nm) (Multitel Ltd, Belgium).
The OPO is tunable from 2 to 8 μm, resulting in an observable
spectral range of 1250–5000 cm–1, and also
allows excitation/detection in both reflection and transmission modes.
To acquire images, the sample is scanned using a Nano-cube XYZ scanner
from PI (P-611.3 NanoCube XYZ Piezo Stage, Physical Instrument), which
is controlled via E-664 Controller. The signal is detected using an
mercury cadmium telluride (MCT) photoconductive detector (P-2748-42,
Hamamatsu) through a lock-in amplifier, with the laser beam mechanically
chopped at 500 Hz. The data acquisition is performed via our own image
acquisition software that allows image collection at the rate of 1
ms/pixel dwell time. Note here that this long pixel dwell time was
imposed by the time constant of the lock-in amplifier used to collect
and amplify the signal. For high-level signals, we achieved a 100
μs pixel dwell time by surpassing the use of a lock-in amplifier.
A calibration sample (Ag film calibration slide with 5 μm holes
and 500 nm Ag nanoparticles) was used to align the microscope and
to estimate the lateral resolution of the system.
Confocal and
Super-Resolution Microscopy
Confocal laser
scanning microscopy was performed to study the permeability variation
looking mainly at the entrance of the dye molecules (rhodamine) from
the environmental solution into the NPC volume. The images were obtained
using a Leica TCS SP5 STED-CW (Leica Microsystems, Mannheim, Germany)
inverted confocal laser scanning microscope equipped with a supercontinuum
laser covering the visible spectrum in the range between 470 and 640
nm. The images were collected using a Leica 1006HCX PL APO STEDorange
NA 1.40 oil immersion objective (Leica Microsystems CMS, Mannheim,
Germany) with an excitation at 561 nm and an emission between 570
and 620 nm, with no lines averaging at a speed of 1 kHz per line,
a pixel dwell time of 2 μs, and a pinhole size of 0.8 Airy.
Under this imaging configuration, typical confocal resolution is on
the order of 200 nm in the lateral direction and 500 nm in the axial
direction. The morphology of the NPCs in contact with cells was also
studied with multicolor 3D-STED super-resolution nanoscopy.
Authors: L-P Choo-Smith; H G M Edwards; H P Endtz; J M Kros; F Heule; H Barr; J S Robinson; H A Bruining; G J Puppels Journal: Biopolymers Date: 2002 Impact factor: 2.505