Microrheometric upconversion-based techniques for intracellular
viscosity measurements
Paloma Rodríguez-Sevilla,
1
Yuhai Zhang,
2
Nuno de Sousa,
3,4
Manuel I. Marqués,
5
Francisco Sanz-
Rodríguez,
1,6
Daniel Jaque,
1,6
Xiaogang Liu,
2
and Patricia Haro-González
1
1
Fluorescence Imaging Group, Departamento de Física de Materiales, Universidad Autónoma de
Madrid, 28049 Madrid, Spain
2
Department of Chemistry, National University of Singapore, Science Drive 3, Singapore 117543,
Singapore
3
Departamento de Física de la Materia Condensada, Condensed Matter Physics Center (IFIMAC),
and Nicolás Cabrera Institute, Universidad Autónoma de Madrid, 28049 Madrid, Spain
4
Donostia International Physics Center (DIPC), Donostia-San Sebastián 20018, Spain
5
Departamento de Física de Materiales, Condensed Matter Physics Center (IFIMAC), and Nicolás
Cabrera Institute, Universidad Autónoma de Madrid, 28049 Madrid, Spain
6
Instituto Ramón y Cajal de Investigaciones Sanitarias, Hospital Ramón y Cajal, Madrid 28034,
Spain
ABSTRACT
Rheological parameters (viscosity, creep compliance and elasticity) play an important role in cell function and viability.
For this reason different strategies have been developed for their study. In this work, two new microrheometric
techniques are presented. Both methods take advantage of the analysis of the polarized emission of an upconverting
particle to determine its orientation inside the optical trap.
Upconverting particles are optical materials that are able to convert infrared radiation into visible light. Their usefulness
has been further boosted by the recent demonstration of their three-dimensional control and tracking by single beam
infrared optical traps. In this work it is demonstrated that optical torques are responsible of the stable orientation of the
Optical Trapping and Optical Micromanipulation XIV, edited by Kishan Dholakia, Gabriel C. Spalding,
Proc. of SPIE Vol. 10347, 103471S · © 2017 SPIE · CCC code: 0277-786X/17/$18 · doi: 10.1117/12.2275944
Proc. of SPIE Vol. 10347 103471S-1
upconverting particle inside the trap. Moreover, numerical calculations and experimental data allowed to use the rotation
dynamics of the optically trapped upconverting particle for environmental sensing. In particular, the cytoplasm viscosity
could be measured by using the rotation time and thermal fluctuations of an intracellular optically trapped upconverting
particle, by means of the two previously mentioned microrheometric techniques.
Keywords: Microrheometry, optical trapping, upconverting particle, optical torque, intracellular viscosity
1. INTRODUCTION
Cellular rheological parameters (viscosity, elasticity and creep compliance) role the different processes that take place
inside the cell. For example, viscosity controls the diffusion of molecules and organelles.
1, 2
Moreover, such parameters
are an indicative of the cell state, allowing to determine the appearance and onset of diseases.
3
Because of that,
rheological parameters are of great interest, and a wide variety of studies have focused on their characterization.
1, 4
For
such purpose, lots of different rheometric techniques have been developed. They can be classified in passive and active
methods. Passive techniques are based on the measurement of the thermal fluctuations of a probe, which could be
endogenous (organelle, vesicle or molecule) or exogenous (internalized particle o molecule).
1, 5-7
The Brownian motion
of the probe could be tracked by using video-based methods or by the analysis of its emission, in the case of luminescent
probes.
5, 8-10
On the other hand, active techniques exert a force over the cell to test its response to the applied stimulus.
11,
12
The action could be performed directly over the cell or by using a probe (usually polystyrene or silica beads). For
example, magnetic rheometers have been used for the determination of the viscoelastic character of the cytoplasm of
living cells.
13
The spinning of cylindrical particles, induced by the action of an external magnetic field, have been used to
measure the intracellular viscosity. This study evidences the usefulness of micromotors to test intracellular rheological
parameters. Active techniques should exert forces on the range of piconewtons in order to not compromise cell viability.
In this sense, optical trapping (OT) has been evidenced as the perfect tool for such purpose, since the strength of optical
forces lie in that range.
14, 15
OT takes advantage of the optical force and torque produced by the light-matter interaction
that stably trap the particle in a certain orientation inside the optical trap (i.e. focused laser beam).
16
Optical forces are
responsible for the confinement of the particle within the beam focus, while optical torques drive a rotation of the
particle towards the stable orientation. Thus, based on this premise, optically driven motors can be also used as
rheometers.
17
For such purpose, the real time orientation of the rotating particle must be characterized. If the particle is
large enough, video-based techniques can be used. However, in most of the cases, the size of the probe is reduced in
order to enhance spatial resolution, thus they cannot be optically resolved. Therefore an alternative approach is
mandatory. As introduced above, luminescent particles can be used as sensor agents. In particular, upconverting particles
(UCPs) have been widely used in different biological studies.
18-21
They present the capability to produce high energy
radiation (ultraviolet or visible) after the absorption of low energy photons (infrared), through a process known as
upconversion.
22
This ability have been already proven to be of great interest for the development of cancer treatments,
Proc. of SPIE Vol. 10347 103471S-2
(a) (b)
980nm laser beam
Linear polarizer
Spectrometer
Microscope
Objective
CMOS
Microchannel /
microscope slide
such as the photodynamic therapy or controlled drug release.
23-26
Moreover, the upconversion luminescence of non-
spherical UCPs is polarized.
27, 28
This leads to a dependence of the upconversion luminescence on the orientation of the
particle.
29
Thus, the analysis of the UCP luminescence can be used to monitor its orientation during the optically-driven
rotation.
Based on the previous idea, two microrheometric techniques are here presented. Both methods take advantage of the
analysis of the polarized emission to determine the orientation of the upconverting particle inside the optical trap. The
active technique relies on the laser-induced rotation of trapped hexagonal upconverting particles to test the dynamic
viscosity of the cytoplasm. On the other hand, the passive method takes advantage of the thermal vibration of the trapped
particle when it reaches its stable orientation inside the optical trap.
2. SAMPLE AND EXPERIMENTAL SETUP
β-NaYF
4
:Er
3+
,Yb
3+
UCPs were synthetized by a hydrothermal procedure as explained elsewhere.
30
These particles
present a hexagonal shape, as shown in Figure 1a, with a mean diameter of 800 nm and a thickness of 400 nm. As
mentioned in the introduction, they present polarized emission which have been characterized by using the experimental
setup depicted in Figure 1b. It is composed by a modified confocal microscope which allows to both spectroscopic
characterization and optical manipulation of the particles. Briefly, the radiation coming from a fiber-coupled diode laser
is focused by using a microscope objective. This radiation is used for excitation of the particle luminescence and its
optical manipulation. The luminescence of the particle is collected with the same microscope objective and directed
towards a high sensitive Si charge-coupled device camera (Synapse, Horiba), attached to a monochromator (iH320,
Horiba), for its real time analysis. For polarized spectroscopy, a linear polarizer (LPNIR050-MP2, Thorlabs) was placed
at the entrance of the detector to select the polarization state of the luminescence. As a results of the upconversion
process, β-NaYF
4
:Er
3+
,Yb
3+
UCPs showed green and red upconversion emission when excited with 980 nm radiation.
31
Only red emission were analyzed, since this emission band was evidenced to have a higher polarization degree than the
green emission band.
29
Optical images of the optical trap area where recorded using a CMOS camera.
Figure 1. (a) SEM image of the β-NaYF4:Er
3+
,Yb
3+
UCPs. (b) Schematic representation of the experimental setup. White and green
discs represents mirrors and filters, respectively. Reprinted with permission from Nano Letters 2016, 16 (12), 8005-8014. Copyright
2016 American Chemical Society.
Proc. of SPIE Vol. 10347 103471S-3
(a)
(b)
Horizontal configuration
6
'z
656 664
630 645 660 675
Wavelength (nm)
(C)
2.5
2.0 -
1.5 -
1.0 -
a
m
0.5 -
"Z 0.0-
ie
. 0.5 -
1.0 -
1.5 -
2.0 -
2.5
Vertical configuration
7L
656 664 656 664
690 645 660 675 690 645 660 675 690
Wavelength (nm)
Wavelength (nm)
a
o o
180
2.5 -
2.0 -
1.5-
1.0 -
a
S 0.5-
0.0 -
N
m
0.5-
1.0 -
1.5-
2.0 -
2.5 -
180
3. POLARIZED SPECTROSCOPY
3.
1 Dried sample characterization
A diluted suspension of the UCPs (2.3 x10
7
UCPs/cm
3
) was dried on a microscope slide by drop casting and the
luminescence of isolated particles were analyzed. The excitation/detection direction was perpendicular to the substrate in
such a way that two different UCP configuration were assessed (see Figure 2a). In the horizontal configuration, the
Figure 2. (a) Schematic representation of the two UCP configurations. The three main polarization states are included. The
propagation direction of the light is indicated by vector k, while its polarization is represented by vector E. (b) Emission spectra
for the three main polarization states. (c) Polar plots of the intensity ratio of the two peaks indicated in (b) as a function of the
polarization state. Left for horizontal configuration, and right for vertical orientation. Reprinted with permission from Nano
Letters 2016, 16 (12), 8005-8014. Copyright 2016 American Chemical Society.
Proc. of SPIE Vol. 10347 103471S-4
particle lies with its hexagonal facet perpendicular to the excitation/detection direction, while, in the vertical
configuration, the longitudinal axis of the particle is parallel to the excitation/detection direction. Polarized spectra were
analyzed for both particle orientation by rotating the linear polarizer and detecting the emission spectra for different
polarization states. At this point we state that, based on a previously published work,
27
the studied UCPs are uniaxial
crystals with their optical axis perpendicular to their hexagonal facet. This assumption implies that the two orientations
of the particle in respect to detection direction allow to assess three polarization states: σ, α, and π. In the σ polarization
state (Figure 2a, left) the luminescence propagates along the optical axis (z axis) of the crystal and, consequently, with a
polarization perpendicular to it. This polarization state corresponds to the particle being in the horizontal configuration.
On the other hand, in the vertical orientation, light propagates perpendicularly to the optical axis of the crystal with a
polarization that can be perpendicular (α, Figure 2a, center) or parallel (π, Figure 2a, right) to it, thus α and π
polarization states are accessible. The emission spectra detected for the three polarization states are show in Figure 2b.
As can be seen, σ and α polarization states are analogous, whereas π polarization state present a completely different
spectral shape. It is worth mentioning that this spectral modulation was expected for uniaxial crystals. In particular, the
most significant spectral change take place at 656 and 664 nm emission peaks. For a better characterization of the
spectral modulation, polar plots of the intensity ratio (I
656
/I
664
) of these two peaks as a function of the polarization state
are shown in Figure 2c. In the horizontal configuration (Figure 2c, left), the emission spectrum do not change with the
polarization state since only σ polarization state is accessible, thus a circular diagram is obtained. However, in the
vertical orientation (Figure 2c, right), the emission spectrum changes from that associated to α polarization state to π
polarization state when the polarization state is detected perpendicularly or parallel to the optical axis of the UCP. Polar
graphs in Figure 2c evidence that the orientation of the particle can be elucidated through the analysis of the
luminescence spectra.
3.2 Optical manipulation
3.2.1 Experimental results
The former analysis can be used for the determination of the UCP orientation within the optical trap. It is expected that,
when it is in the horizontal configuration, the non-polarized (no polarization state selected) emission spectrum is that
associated to the σ polarization state. However, when the particle is in the vertical orientation, the emission spectrum is a
combination of that obtained for α and π polarization states.
Based on this idea, the orientation of the UCPs within the optical trap was elucidated by analyzing the non-polarized
emission spectrum of the trapped particle. For such purpose, a diluted aqueous suspension of particles (2.3 x10
7
UCPs/cm
3
) were injected into a microchannel (µ-Slide I 80106, Ibidi Inc.). Then, a single particle was optically trapped
and its emission spectrum detected. Figure 3a shows the non-polarized spectrum obtained for a single optically trapped
UCP. As can be seen, the spectral shape of the emission is a combination of that associated to α and π polarization. This
unequivocally indicates that the stable orientation within the optical trap is the vertical configuration, as schematically
represented in the inset of Figure 3a. For a better characterization, the emission spectra as a function of the polarization
state was also analyzed. The polar plot of the intensity ratio is represented in Figure 3b. As expected, the dependence of
Proc. of SPIE Vol. 10347 103471S-5
