Intranuclear binding kinetics and mobility of single native U1 snRNP particles in living cells.
ABSTRACT Uridine-rich small nuclear ribonucleoproteins (U snRNPs) are splicing factors, which are diffusely distributed in the nucleoplasm and also concentrated in nuclear speckles. Fluorescently labeled, native U1 snRNPs were microinjected into the cytoplasm of living HeLa cells. After nuclear import single U1 snRNPs could be visualized and tracked at a spatial precision of 30 nm at a frame rate of 200 Hz employing a custom-built microscope with single-molecule sensitivity. The single-particle tracks revealed that most U1 snRNPs were bound to specific intranuclear sites, many of those presumably representing pre-mRNA splicing sites. The dissociation kinetics from these sites showed a multiexponential decay behavior on time scales ranging from milliseconds to seconds, reflecting the involvement of U1 snRNPs in numerous distinct interactions. The average dwell times for U1 snRNPs bound at sites within the nucleoplasm did not differ significantly from those in speckles, indicating that similar processes occur in both compartments. Mobile U1 snRNPs moved with diffusion constants in the range from 0.5 to 8 microm2/s. These values were consistent with uncomplexed U1 snRNPs diffusing at a viscosity of 5 cPoise and U1 snRNPs moving in a largely restricted manner, and U1 snRNPs contained in large supramolecular assemblies such as spliceosomes or supraspliceosomes.
- SourceAvailable from: Aaron Christian Ponti[show abstract] [hide abstract]
ABSTRACT: Fluorescent speckle microscopy (FSM) is becoming the technique of choice for analyzing in vivo the dynamics of polymer assemblies, such as the cytoskeleton. The massive amount of data produced by this method calls for computational approaches to recover the quantities of interest; namely, the polymerization and depolymerization activities and the motions undergone by the cytoskeleton over time. Attempts toward this goal have been hampered by the limited signal-to-noise ratio of typical FSM data, by the constant appearance and disappearance of speckles due to polymer turnover, and by the presence of flow singularities characteristic of many cytoskeletal polymer assemblies. To deal with these problems, we present a particle-based method for tracking fluorescent speckles in time-lapse FSM image series, based on ideas from operational research and graph theory. Our software delivers the displacements of thousands of speckles between consecutive frames, taking into account that speckles may appear and disappear. In this article we exploit this information to recover the speckle flow field. First, the software is tested on synthetic data to validate our methods. We then apply it to mapping filamentous actin retrograde flow at the front edge of migrating newt lung epithelial cells. Our results confirm findings from previously published kymograph analyses and manual tracking of such FSM data and illustrate the power of automated tracking for generating complete and quantitative flow measurements. Third, we analyze microtubule poleward flux in mitotic metaphase spindles assembled in Xenopus egg extracts, bringing new insight into the dynamics of microtubule assemblies in this system.Biophysical Journal 09/2003; 85(2):1289-306. · 3.67 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: For more than 60 years, RNA has been detectable in fixed cells and tissues by relatively specific staining methods. More recently, it has become possible to study RNA in unfixed, live cells. This review article describes how the intracellular dynamics and localization of RNA in vivo can be studied by microinjection of fluorescent RNA into cells- an approach we have termed Fluorescent RNA Cytochemistry. Depending on the particular RNA species under investigation, Fluorescent RNA Cytochemistry can operate as a "stain" to reveal intracellular sites at which a given RNA resides, or as a "tracer" to allow movements of a dynamically translocating RNA to be followed in the living cell. Several examples of Fluorescent RNA Cytochemistry are presented, collectively illustrating the range of applicability this approach offers in the toolbox of gene expression, studied as in vivo cell biology.European journal of histochemistry: EJH 02/2004; 48(1):57-64. · 2.41 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: We present experiments in which single proteins were imaged and tracked within mammalian cells. Single proteins of R-phycoerythrin (RPE) were imaged by epifluorescence microscopy in the nucleoplasm and cytoplasm at 71 frames/s. We acquired two-dimensional trajectories of proteins (corresponding to the projection of three-dimensional trajectories onto the plane of focus) for an average of 17 frames in the cytoplasm and 16 frames in the nucleus. Diffusion constants were determined from linear fits to the mean square displacement and from the mean displacement squared per frame. We find that the distribution of diffusion constants for RPE within cells is broader than the distributions obtained from RPE in a glycerol solution, from a Monte Carlo simulation, and from the theoretical distribution for simple diffusion. This suggests that on the time scales of our measurements, the motion of single RPE proteins in the cytoplasm and nucleoplasm cannot be modeled by simple diffusion with a unique diffusion constant. Our results demonstrate that it is possible to follow the motion of single proteins within cells and that the technique of single molecule tracking can be used to probe the dynamics of intracellular macromolecules.Biophysical Journal 11/2000; 79(4):2188-98. · 3.67 Impact Factor
Molecular Biology of the Cell
Vol. 17, 5017–5027, December 2006
Intranuclear Binding Kinetics and Mobility of Single Native
U1 snRNP Particles in Living Cells□
David Gru ¨nwald,*†Beatrice Spottke,* Volker Buschmann,‡§
and Ulrich Kubitscheck*
*Institut fu ¨r Physikalische und Theoretische Chemie, Rheinische Friedrich-Wilhelms-Universita ¨t, D-53115
Bonn, Germany; and‡Max-Delbru ¨ck-Centrum, 13125 Berlin, Germany
Submitted June 27, 2006; Revised September 6, 2006; Accepted September 13, 2006
Monitoring Editor: Jennifer Lippincott-Schwartz
Uridine-rich small nuclear ribonucleoproteins (U snRNPs) are splicing factors, which are diffusely distributed in the
nucleoplasm and also concentrated in nuclear speckles. Fluorescently labeled, native U1 snRNPs were microinjected into
the cytoplasm of living HeLa cells. After nuclear import single U1 snRNPs could be visualized and tracked at a spatial
precision of 30 nm at a frame rate of 200 Hz employing a custom-built microscope with single-molecule sensitivity. The
single-particle tracks revealed that most U1 snRNPs were bound to specific intranuclear sites, many of those presumably
representing pre-mRNA splicing sites. The dissociation kinetics from these sites showed a multiexponential decay
behavior on time scales ranging from milliseconds to seconds, reflecting the involvement of U1 snRNPs in numerous
distinct interactions. The average dwell times for U1 snRNPs bound at sites within the nucleoplasm did not differ
significantly from those in speckles, indicating that similar processes occur in both compartments. Mobile U1 snRNPs
moved with diffusion constants in the range from 0.5 to 8 ?m2/s. These values were consistent with uncomplexed U1
snRNPs diffusing at a viscosity of 5 cPoise and U1 snRNPs moving in a largely restricted manner, and U1 snRNPs
contained in large supramolecular assemblies such as spliceosomes or supraspliceosomes.
Cell nuclei exhibit a high degree of spatial and functional
organization of their molecular components (Cremer and
Cremer, 2001; Misteli, 2005). The question how nuclear fac-
tors move and interact within this well-organized structure,
how molecular factors find their targets, and how trafficking
exerts a possible regulatory function is currently a focus in
cell biology (Bubulya and Spector, 2004; Pederson, 2004;
Gorski and Misteli, 2005). An exact quantification of inter-
actions and molecular mobilities is the prerequisite for a
detailed understanding of the formation of intranuclear
structures and the regulation of their function by modifica-
tions of interactions and transport (Gorski et al., 2006).
Eukaryotic pre-mRNA transcripts go through several
post-transcriptional modifications before their translocation
by the NPCs into the cytoplasm (Darzacq et al., 2005). Usu-
ally pre-mRNAs have noncoding sequences designated as
introns that must be removed from the sequence to yield
functional mRNA. This essential biochemical processing is
designated as pre-mRNA splicing, which is achieved by
intranuclear molecular pre-assembled complexes, the so-
called spliceosomes. Spliceosomes consist of more than 70
different proteins, many of which are part of the uridine-rich
small nuclear ribonucleoproteins (U snRNPs), which are
classified as U1, U2, U5, and U4/U6, according to their small
nuclear RNA (snRNA) content. With the exception of U6, the
snRNAs are synthesized in the nucleus by RNA polymerase
II and exported to the cytoplasm, where sets of common and
specific proteins bind to the snRNAs (Will and Luhrmann,
2001). After their cytoplasmic assembly U snRNPs are reim-
ported into the nucleus. Spliceosomes have been shown to
be subcomplexes of huge multicomponent nuclear RNP
complexes, so-called supraspliceosomes. Purified by density
gradient centrifugation they sedimented as 200S complexes
(Sperling et al., 1985; Spann et al., 1989) with a mass of 21
MDa (Muller et al., 1998). Three-dimensional image recon-
struction of isolated supraspliceosomes revealed a geometric
extension of 50 ? 50 ? 35 nm3(Sperling et al., 1997; Medalia
et al., 2002).
The spatiotemporal distribution of splicing factors within
cell nuclei is an important example of the functional orga-
nization of the cell nucleus (Lamond and Spector, 2003).
Fluorescence labeling of splicing factors such as ASF/SF2 or
U snRNPs reveals numerous irregular, punctuate structures
distributed on a more homogeneous background within cell
nuclei. These fluorescent structures are formed by the en-
richment of splicing factors in subnuclear structures such as
interchromatin granule clusters and perichromatin fibrils
collectively designated as splicing factor compartments or
This article was published online ahead of print in MBC in Press
on September 20, 2006.
rial at MBC Online (http://www.molbiolcell.org).
Present addresses:†Department of Anatomy and Structural Biol-
ogy, Albert Einstein College of Medicine, Bronx, NY 10461;§Pico-
Quant GmbH, Rudower Chaussee 29, 12489 Berlin, Germany.
Address correspondence to: Ulrich Kubitscheck (u.kubitscheck@
Abbreviations used: ASF, alternative splicing factor; EMCCD, elec-
tron multiplying CCD; FCS, fluorescence correlation spectroscopy;
FRAP, fluorescence recovery after photobleaching; MSD, mean
square displacements; SPT, single-particle tracking; snRNA, small
nuclear RNA; MSD, mean square displacements; U snRNP, uridine-
rich small nuclear ribonucleoprotein.
VThe online version of this article contains supplemental mate-
© 2006 by The American Society for Cell Biology5017
speckles. The functional role of speckles is still unresolved.
Possible functions include splicing factor reprocessing sites
or storage spaces regulating the level of free and active
factors. Another hypothesis is that in speckles splicing fac-
tors are assembled together with other components of the
transcription and RNA processing machinery into supramo-
lecular complexes, whereas the dispersed splicing factors
might represent active complexes involved in cotranscrip-
In the last few years large efforts have been undertaken to
gain insight into intranuclear mobility and interactions of
intranuclear molecular components—DNA-binding pro-
teins, splicing factors, and RNP particles. The experimental
approaches mostly used were fluorescence correlation spec-
troscopy (FCS; Brock et al., 1998; Politz et al., 1998, 2006;
Schwille et al., 1999; Wachsmuth et al., 2000), and fluores-
cence recovery after photobleaching (FRAP) or photoactiva-
tion (PA) combined with mathematical modeling (Seksek et
al., 1997; Houtsmuller et al., 1999; Kruhlak et al., 2000; Lukacs
et al., 2000; Phair and Misteli, 2000; Verkman, 2002; Carrero
et al., 2003; Braga et al., 2004; Phair et al., 2004; Beaudouin et
al., 2006). In addition to these established techniques, re-
cently single-particle imaging based on state-of-the-art
videomicroscopy has proven its power to visualize details of
trafficking within the cell nucleus in a sequence of publica-
tions (Goulian and Simon, 2000; Kues et al., 2001a, 2001b;
Seisenberger et al., 2001; Babcock et al., 2004; Shav-Tal et al.,
2004; Bausinger et al., 2006).
The results obtained so far must be discriminated accord-
ing to analysis of tracer molecule mobility and studies fo-
cusing on functionally active proteins or RNPs (see review,
Gorski et al., 2006). Inert tracer molecules usually show
diffusion coefficients within cell nuclei ?4–15 times smaller
than aqueous solution (Lang et al., 1986; Seksek et al., 1997;
Braga et al., 2004). Usually, biologically active molecules
were reduced in their apparent mobility by a factor of
10–100 compared with aqueous solution (reviewed by
Houtsmuller and Vermeulen, 2001; Verkman, 2002; Gorski et
al., 2006). The significant reduction of mobility was inter-
preted to indicate frequent, but transient interactions of the
examined molecular factors with numerous largely immo-
bile intranuclear structures. Obviously, the motion of nu-
clear proteins, RNA molecules, or RNP particles reflects
their intranuclear function. Surprisingly, the GFP conjugate
of ASF/SF2 showed the same mobility independently of
whether it was associated with speckles or dispersed in the
nucleoplasm (Kruhlak et al., 2000). In accordance to this
work, it was recently found that the mobility of poly(A)
RNA did not differ between speckles and nucleoplasm in
HeLa cell nuclei (Politz et al., 2006).
In the last few years single-molecule tracking (SMT) by
high-speed fluorescence videomicroscopy has evolved to a
routine method in the biosciences (Schwille and Kettling,
2001; Moerner, 2003; Sako and Yanagida, 2003; Tinnefeld
and Sauer, 2005). It still appeared debatable, however,
whether the time resolution attainable by high-speed cam-
eras was really high enough to follow the trajectories of
single protein molecules within the cellular interior. Re-
cently, we demonstrated the imaging and tracking of single
protein molecules in physiological buffer at frame rates of
?350 Hz (Grunwald et al., 2006). Analysis of the single-
molecule trajectories yielded the same diffusion constants as
control measurements performed by FCS. Considering the
generally higher intracellular viscosity, it was thus proven
that the tracking of proteins inside living cells is faithfully
possible if frame rates in the range of 100 frames per second
or higher can be achieved. Previously we used single-mol-
ecule imaging for analyzing the movements of a recombi-
nant ?-galactosidase protein (Kues et al., 2001b) and of the
splicing factor U1 snRNP (Kues et al., 2001a) in digitonin-
permeabilized cells with maximum frame rates of 35 Hz.
These cells represented a system, which still contained in-
tracellular structures as geometric constraints on mobility,
but presumably not the functionally intact DNA and RNA
In the current study we applied single-molecule tracking
to study the intranuclear dynamics of a biologically active
splicing factor in live cells. We microinjected fluorescently
labeled splicing factors U1 snRNPs into the cytoplasm of
living cells and therefore maintained their biochemical func-
tions, integrity, and the structure of the nuclei. The splicing
factors were imported into the nucleus by nucleo-cytoplas-
mic transport. Imaging was performed using a fast and
sensitive electron-multiplying CCD (EMCCD). Using this
camera we could follow the movements of U1 snRNPs in
real time, but could also focus on slow events by reducing
the imaging frame rate. We found that U1 snRNPs moved
with diffusion coefficients in the range of 0.5–8 ?m2/s. Al-
though long-distance movements of U1 snRNPs could
clearly be observed, transient binding to immobile sites was
the dominating process. The dissociation kinetics from these
binding sites was analyzed on different time scales ranging
from milliseconds to seconds. Our data provided new in-
sight into the molecular dynamics of a functional ribonucle-
oprotein particle within living cells.
MATERIALS AND METHODS
Reagents and Buffers
Cy5-labeled U1 snRNPs were prepared as described in Huber et al. (1998).
Purity, functionality, and integrity of labeled U1 snRNPs were confirmed as
described in Marshallsay and Luhrmann (1994). Before final use in microin-
jection experiments U1 snRNP-Cy5 were diluted and centrifuged in transport
buffer (50 mM HEPES/KOH, pH 7.3, 110 mM potassium acetate, 5 mM
sodium acetate, 2 mM magnesium acetate, 1 mM EGTA, 2 mM DTT). The
plasmid coding for fusion proteins of ASF/SF2-GFP was a kind gift from
Cell Culture and Transfection
HeLa cells were grown in DMEM supplemented with 10% FCS. For live cell
analysis cells were seeded on cover glasses. As reference label for the nuclear
speckles cells were transfected with plasmids coding for fusion proteins of
ASF/SF2-GFP (Sleeman et al., 1998) using an Effectene transfection kit (Qia-
gen, Hilden, Germany) 1 d after seeding. Microscopic analysis was performed
in a custom-built sample holder at room temperature 24 h after transfection to
allow expression of fusion proteins.
Single-particle tracking (SPT) experiments were performed using a custom-
built single-molecule microscope based on a Zeiss Axiovert 100TV equipped
with a 63? NA 1.4 oil immersion objective lens (Jena, Germany; Kubitscheck
et al., 2005). Green fluorescence was excited by an Ar?-Laser emitting at 488
nm, and red fluorescence was excited by a HeNe-Laser emitting at 632.8 nm.
Laser illumination was switched on only during image acquisition by means
of an acousto-optical tunable filter. For single-particle image acquisition we
used the iXon DV 860 BI camera (Andor Technology, Belfast, Northern
Ireland) in combination with a 4? magnifier yielding a pixel size in the object
space of 95.24 nm. Microinjection of U1 snRNPs was carried out with an
Eppendorf injection and micromanipulation setup using an injection time of
1 s at an injection pressure of 100 hPa and a holding pressure 15 hPa.
Single-particle imaging was started 10 min after microinjection of U1 snRNPs
into the cytoplasm to allow cells to recover. Cell recovery was monitored by
examining the cellular morphology in bright-field mode by digitally contrast-
enhanced imaging. Before acquisition of the movies a focal plane was
searched to optimize the contrast of the GFP-labeled nuclear speckles. After
taking an image in the green channel with a Zeiss Axiocam MRm, movies
recorded in the red channel illustrated the motion of U1 snRNP-Cy5 after its
nuclear import. Usually 1000 frames were recorded in a single movie, with
integration times of 5 and 10 ms and frame rates of kacq? 5, 10, 100, and 200
Hz. A total of 10 cells was examined, yielding more than 100 single movies.
The green and red fluorescence channels were scaled and aligned to each
D. Gru ¨nwald et al.
Molecular Biology of the Cell5018
other using images of dispersed, immobilized multicolor fluorescence beads
(TetraSpeck Microspheres, diameter 0.1 ?m, Molecular Probes, Leiden, The
Image Processing of Video Images
Identification and tracking of the single-molecule signals was accomplished
using Diatrack 3.0 (Semasphot, North Epping, Australia), a commercial image
processing program for the identification and localization of single-particle
signals and trajectories (Vallotton et al., 2003). For tracking a maximal dis-
placement of 10 pixels from frame to frame was allowed. The application of
the automated data analysis scheme to our data was problematic, because the
single-molecule data often displayed low signal-to-noise ratios. Therefore,
after Diatrack processing we verified the single-particle tracks in the original,
unprocessed data by visual inspection. Intracellular compartments (cyto-
plasm, nucleoplasm, or speckles) were marked in specific colors using IPLab
(Scanalytics Inc., Fairfax, VA), and this false color reference image was used
for compartment assignment of the individual tracks with the help of user-
written macros in Origin 7.5 (Microcal Software, Northhampton, MA). All
tracks within the cytoplasm including a 10-pixel border region near the
nuclear envelope were discarded to avoid evaluation of particles during their
import into the nucleus. Also, all tracks within a distance of eight pixels from
the image border were discarded.
Each U1 snRNP-Cy5 trajectory was defined as a set of coordinates (xi, yi) with
1 ? i ? N, where N denoted the total number of observations. In the case of
two-dimensional Brownian motion the mean square displacements, ?r2(tc)?,
are related to time and diffusion coefficient, D:
?r2(tc)? ? 4Dtc
Thus, a linear relationship between ?r2(tc)? and time indicates Brownian
motion. However, if the motion is not due to free diffusion but, e.g., to
confined diffusion or directed flow, the relation between the mean square
displacements (MSD) and time is nonlinear (Saxton and Jacobson, 1997).
Furthermore, the joint analysis of the trajectories of an ensemble of molecules
according to Equation 1 is not suitable when the population contains different
mobility fractions. Such heterogeneous populations are more appropriately
analyzed by a jump distance analysis.
Jump Distance Analysis
The probability that a particle starting at a specific position will be encoun-
tered within a shell of radius r and width dr at time t from that position is for
a single species diffusing in two dimensions given as follows (Crank, 1975):
p(r, t) dr ?
if we identify the starting position with the origin. Experimentally, this
probability distribution could be approximated by a frequency distribution,
which was obtained by counting the jump distances within respective inter-
vals [r, r ? dr] traveled by single particles after a given time lag. In cases of
particles with multiple diffusive species or particles changing the mode of
motion along their trajectory the jump distance distributions cannot satisfac-
torily be fitted by eq. 2 assuming a single diffusion coefficient. Such different
mobility fractions can be detected, and quantified by curve fitting taking
several diffusion terms into account. E.g., for a jump distance distribution
containing contributions from three species with differing diffusion constants,
p?(r, t) dr ??
where M is related to the number of jumps considered in the analysis, and f1,
f2, and f3designate the fractions with diffusion constants D1, D2, and D3,
For immobile particles the jump distance between two subsequent frames
corresponded to a maximum drmax, which was determined by the localization
precision alone. We defined as drmaxthe threefold localization precision
(drmax? 3?loc? 100 nm). All particles that did not jump farther than drmax
between subsequent frames were taken as immobile.
To determine the binding durations, we screened all trajectories for the
jump distances between subsequent frames and counted the number of
subsequent steps with jump lengths smaller than drmax. This number, n,
characterized the length of an immobile trajectory or trajectory segment. For
all immobile trajectory segments identified in this manner, the maximum
extensions in the x and y direction, ?xmaxand ?ymax, were determined. All
trajectories with either ?xmaxor ?ymax?150 nm were inspected visually to
decide whether they were produced by indisputable immobile particles with
a random distribution of the jump directions. This was done to exclude the
possibility that several smaller steps into the same direction would finally
lead to a significant, sliding movement beyond the localization precision. No
trajectory had to be rejected because of this criterion. Finally, n was translated
into a binding time, ?, by ? ? (n ? 1)/kaq. The single values of ? were used to
calculate a decay curve N(?) giving the number of particles, which were still
immobile after time ?.
Correction of Photobleaching
Photobleaching was quantified by plotting the average intensity in the cell
nucleus as a function of time. Because U1 snRNPs were not exported, the
observed fluorescence decay was due to photobleaching. The fluorescence decay
was fitted by a monoexponential function, which yielded a bleaching time
constant of ?bl? 120 ? 30 ms. Hence, 50% of individual U1 snRNPs were
bleached after the acquisition of 16 images at a single-frame integration time of 5
ms, corresponding to a continuous illumination of 80 ms. The decay curves N(?)
of bound particles was corrected for bleaching by N(?)corr? N(?) ? e?/?bl.
Single U1 snRNPs Can Be Visualized and Tracked in Real
Time in Live Cell Nuclei
U1 snRNPs labeled by Cy5 were microinjected at a low con-
centration into the cytoplasm of live HeLa cells transiently
expressing ASF/SF2-GFP. By using very gentle injection con-
ditions the internal structure of the nuclei was left unaffected
by the microinjection, and the particles could be examined
in an undisturbed intranuclear environment after their nu-
clear import. According to previous work (Kleinschmidt and
introduced into the cytoplasm was expected to be imported
into the nucleus after 10 min, which was the time when mea-
surements were started. The concentration of the injected fluo-
rescently labeled U1 snRNPs was chosen such that the concen-
tration in the nucleus was in the picomolar range. This
extremely low concentration of particles allowed imaging the
particles spatially separated from each other. Using a custom-
built single-molecule microscope equipped with efficient laser
irradiation and a fast, highly sensitive EMCCD imaging sys-
tem, single fluorescent RNP complexes could be visualized in
real time in the cellular interior. Thereby high-speed movies of
the functioning of single U1 snRNPs within living cells (see
Supplementary Movie 1, Online Supplementary Material)
tion of single-particle traces by dedicated image processing
tools (for details on particle identification and trajectory extrac-
tion, see Materials and Methods). An example of the trajectories
of single Cy5-labeled U1 snRNPs and their spatial positions
within a cell nucleus are shown in Figure 1. The corresponding
Supplementary Movie was recorded at 200 Hz. Figure 1A
shows trajectories of numerous single mobile and immobile
U1snRNPs. The RNP traces were overlaid to the green fluores-
cence image of the same HeLa cell nucleus expressing ASF/
SF2-GFP. The brighter spots marked by dark-gray lines indi-
cated the positions of splicing factor compartments or speckles.
Note that the ASF/SF2-GFP was not imaged by a confocal but
a normal fluorescence microscope. Therefore, the contrast of
the speckles was not as high as can be achieved by confocal
imaging (Lamond and Spector, 2003). Plots like this were pro-
duced for each acquired single-particle movie and represented
the first qualitative views onto the U1 snRNP dynamics within
living cell nuclei. Mobile U1 snRNPs as well as transiently
immobilized particles could well be distinguished and ana-
lyzed in the nucleoplasm as well as inside speckles. The white
box in Figure 1A marks a region containing a single trajectory,
which was analyzed in detail in Figure 1, B–D. Figure 1B
displays the sequence of single frames showing the RNP on its
trajectory. A magnified plot of the trajectory is displayed in
Single U1 snRNP Dynamics in Live Cells
Vol. 17, December 2006 5019
site, resided here for a short time, and then left the site again.
Figure 1D displays the time course of the fluorescence signal in
the attachment region. On approach of the particle, the signal
(A) Trajectories (black dots connected by black lines) of numerous
single U1snRNPs extracted from a single-molecule movie, which
was recorded at 200 Hz (Supplementary Movie 1, Online Supple-
mentary Data). The trajectories were plotted over an image of the
ASF/SF2-GFP, which was transiently expressed in the respective
HeLa cell nucleus. Here, all nuclear trajectories were shown; how-
ever, those close to the borders were not evaluated (see Materials and
Methods). The brighter spots marked by dark gray lines indicated
the positions of speckles. The approximate position of the nuclear
envelope was deduced from the limits of ASF/SF2-GFP fluores-
cence, and indicated by the white line. Mobile U1 snRNPs as well as
transiently immobilized RNPs could well be distinguished within
nucleoplasm and speckles. The white bar corresponded to 2 ?m.
The white box marks a region containing a single trajectory, which
was scrutinized in B to D. (B) The short image sequence of the
trajectory marked in A. The size of a single image was 8 ?m2. (C)
Magnified view of the positions, at which the U1 snRNP was
observed, revealing that the RNP moved to a specific site, resided
Trajectories of single U1 snRNPs in a living cell nucleus.
there for a certain time and then left the site again. Please note that
the complete field shown corresponds to 1 ?m2only. (D) Time
course of the fluorescence signal at the attachment region x ? 6.77 ?
0.05 and y ? 2.67 ? 0.05 ?m, which indicated that the particle
remained for 45 ms at the specific nucleoplasmic position before
(A) Magnified view of a specific region of a cell nucleus. At many
positions as marked by black arrows single U1 snRNPs were ob-
served for extended periods of time at specific sites; scale bar, 1 ?m.
(B) The distance distribution of the single positions of 100 randomly
selected immobile trajectories to their respective centers could well
be described by a Gaussian distribution with a SD of 30 nm (dashed
line). A slight deviation between fit and histogram occurred at
distances greater than 60 nm.
The majority of single U1 snRNPs was in a bound state.
D. Gru ¨nwald et al.
Molecular Biology of the Cell5020
for nine images. This illustrated that the particle remained for
45 ms at the specific position before leaving. In addition, this
plot demonstrated the excellent signal-to-noise ratio (SNR) of
single U1 snRNP-Cy5 imaging in the cell nuclei.
Single U1 snRNPs Were Predominantly Observed during
Single U1snRNP particles were often observed for longer
periods of time at specific sites such as shown in the example
above (Figure 1C). Figure 2A shows a magnified view of a
specific region of a cell nucleus. At numerous positions (see
black arrows) single U1 snRNPs were observed for extended
periods of time at specific spots. We examined the geomet-
rical extension of these spots by calculating the center of
mass of 100 arbitrarily selected spots in different cells and
plotted the distance distribution of the single positions to
their respective centers (Figure 2B). This distance distribu-
tion could satisfactorily be described by a Gaussian distri-
bution with a SD of ?exp? 30 ? 5 nm. It should be noted that
the position measurement of a single particle or molecule
necessarily has a limited precision, which is defined by the
SNR and other parameters, such as pixel size and magnifi-
cation of the microscope (Kubitscheck et al., 2000; Thompson
et al., 2002). The fundamental cause for the limited precision
is the photon noise of the single-molecule fluorescence emis-
sion. Hence, for a completely immobile molecule, the mea-
sured position has a specific variation upon repetition of the
measurement, which is designated as localization precision.
This localization precision can be estimated by theoretical and
experimental means and yielded for the SNR values of 5–10
obtained by us in the cellular interior a value of ?theo? 20–25
nm (Kubitscheck et al., 2000). Hence, the single U1 snRNPs,
which appeared repetitively at distinct positions within the
nuclei, showed a positional variation almost identical to that of
completely immobile particles. We can conclude that these
particles were firmly attached to geometrically well-defined,
specific sites over the respective observation time.
Dwell Times of Single U1 snRNPs at Their Binding Sites
Movies of the intranuclear U1 snRNP dynamics revealed
that binding durations of individual particles varied signif-
icantly ranging from milliseconds to seconds. However, sin-
gle-particle fluorescence could only be observed in a limited
number of frames before final photobleaching. The average
frame number over which single U1 snRNPs could be ob-
served was 12. Therefore, photobleaching made the direct
observation of long binding events with a high frame rate
difficult. To obtain a more complete view of the U1 snRNP
binding events, we acquired movies at different frame rates
kacqranging from 5 to 200 Hz.
snRNPs. (A–D) The curves quantify the disso-
ciation of U1 snRNPs from binding sites ob-
served at 200, 100, 10, and 5 Hz as indicated in
the graphs. The axis labels given at the left
hand axis refer to the nucleoplasmic binding
(?), and the labels at the right hand axis refer
to the binding in speckles (●). The lines show
the results of fits to the data using a sum of
two exponentials. Dotted lines, the fits to the
nucleoplasm data; solid lines, fits to the
speckle data. (E) The averaged decay times as
determined by the fits in A–D were deter-
mined and plotted as a function of the image
cycle time. Obviously the binding times were
related to the time scale, at which the binding
Binding duration of immobile U1
Single U1 snRNP Dynamics in Live Cells
Vol. 17, December 20065021
For immobile particles the lateral distance between two
subsequent observations was defined solely by the localiza-
tion precision. To account for all immobile particles, we
defined a maximum step size drmax, the threefold localiza-
tion precision (drmax? 3?exp? 100 nm). All particles that
did not move beyond drmaxbetween subsequent frames
were regarded as immobile, and 99.7% of all immobile par-
ticles were taken into account by this criterion.
For determination of the binding durations the jump dis-
tances between subsequent frames were screened, and the
number of subsequent steps with jump lengths smaller than
drmaxwas counted. This number n corresponded to the
length of an immobile trajectory segment. Finally, the num-
ber of jumps in sequence n for which the particles did not
move was translated into a binding time ?, with ? ? (n ?
1)/kacq. These data were used to construct a decay curve
N(?) quantifying the number of molecules, which were still
bound at a specific site after time ?. Such decay curves were
determined from all movies obtained with a given kacqfor
intranuclear binding sites within and outside speckles. Fi-
nally the decay curves were corrected for photobleaching of
the U1 snRNPs (see Materials and Methods).
Figure 3, A–D, shows the resulting decay curves N(?)
quantifying the dissociation of U1 snRNPs from the putative
binding sites observed at imaging rates of 200, 100, 10, and
5 Hz. The decay curves were determined for the nucleo-
plasm (?) and speckles (●). The decay kinetics did not
comply with monoexponential functions, but fits using dou-
ble-exponential decay functions yielded for all data sets
satisfactory results (full and dotted lines in Figure 3, A–D,
respectively). From the fits we calculated the weighted av-
erages of the respective decay times ?aveand plotted these as
a function of the cycle time (inverse of the frame rate) in
Figure 3E. Obviously, the average decay times for nucleo-
plasm and speckles did not differ much, indicating compa-
rable dwell times of U1 snRNPs in both nuclear domains.
However, it was remarkable that the dwell times were not
constant for different cycle times, but on the contrary de-
pended on the time scale at which the binding was analyzed.
Obviously the dissociation of U1 snRNPs from their intranu-
clear binding sites could not be described by a simple bimo-
lecular dissociation kinetics. The unusual kinetics is dis-
cussed thoroughly below.
Intranuclear Mobility of U1 snRNPs
The movie data (see Online Supplementary Material, Sup-
plementary Movie 1) and Figures 1 and 2 demonstrate that
U1 snRNPs were mostly observed in an immobile, obviously
bound state. However, the high-speed movies also show
numerous clearly mobile U1 snRNPs, from which direct
information on the mobility characteristics of intranuclear
U1 snRNPs could be obtained. Supplementary Movie 2
shows an example of a mobile U1 snRNP, the single frames
of the sequence that are shown and analyzed in Figure 4.
of the U1 snRNP was marked by an arrow. Sequence from a movie taken with 100 Hz and a frame integration time of 5 ms. Original images
were smoothed by a Gaussian kernel (2 pixel radius), and the background was subtracted (rolling ball, r ? 20) for display here; original and
filtered data to this particle are available online; see Supplementary Movie 2. Full field: (128 pixel)2corresponding to (12.2 ?m)2. (B) Line
intensity profile through the position indicated in frame 6 of the sequence. (C) Trajectory of the particle in a magnified view. Please note that
the complete field shown corresponds to 9 ?m2. (D) Mean square displacements of the single-particle trajectory as a function of time. A linear
fit to these data yielded a diffusion constant of D ? 6.3 ? 0.6 ?m2/s. The deviations from linearity for higher MSD values were due to the
fact that a trajectory formed by only 10 positions was evaluated.
Trajectory analysis of a typical mobile U1 snRNP. (A) The frames show the trajectory of a typical mobile U1 snRNP. The position
D. Gru ¨nwald et al.
Molecular Biology of the Cell5022
This mobile U1 snRNP was observed in 10 subsequent
frames in a sequence acquired at 100 Hz with an integration
time of 5 ms per frame. The particle was moving within the
nucleoplasm (for an overlay with the ASF-GFP image, see
Supplementary Figure 1 in the Online Supplementary Ma-
terial). Figure 4A shows the single frames, which were fil-
tered and background subtracted to enhance the particle
contrast. Figure 4B displayed the line profile of the particle.
The profile demonstrated that an unambiguous identifica-
tion of the particle above the background noise was straight-
forward. The complete trajectory of the particle is shown in
Figure 4C in a magnified view, whereas the MSDs of this
trajectory were plotted as a function of time in Figure 4D. A
linear fit to these data according to Eq. (1) yielded a diffusion
constant of D ? 6.3 ? 0.6 ?m2/s.
To obtain a more general characterization of the U1
snRNP mobility, we analyzed the distances covered by the
U1 snRNPs between subsequent frames, the so-called jump
distances (Smith et al., 1999). Figure 5 shows the jump dis-
tance distributions for all U1 snRNPs, which were observed
within the nucleoplasm in the movies acquired at 100 Hz.
Such distributions can be described by Eq. (2), if they are due
to particles diffusing with a single diffusion coefficient (see
Figure 6 in Kues et al., 2001b). However, the distribution
shown in Figure 5 could not at all be fitted assuming a single
diffusion coefficient (Eq. 2), in contrast to jump distance
distributions of molecules diffusing in buffer solution
(Kubitscheck et al., 2000; Grunwald et al., 2006). For the data
displayed in Figure 5 the minimum number of three diffu-
sion terms was required for an acceptable fit. The fitting
results were indicated by the lines in Figure 5. The major
fraction, f1, corresponded to particles, which moved with a
variance of (50 nm)2, a value close to the experimental
localization precision (see legend of Figure 2). Hence, these
particles did not show significant jumps beyond the local-
ization precision and corresponded to those particles, which
were identified above and discussed as particles attached to
binding sites. Besides the dominating immobile fraction, two
mobile fractions, f2and f3, could be identified, which moved
with diffusion coefficients of D2? 0.51 ? 0.05 ?m2/s and
D3? 8.2 ? 3 ?m2/s. The area under the curves corre-
sponded to the relative sizes of the three fractions, and
yielded 77.5, 15, and 7.3% for f1, f2, and f3, respectively.
The nucleoplasm corresponded to 90–95% of the observed
intranuclear area of the various analyzed cell nuclei. The
speckles had limited sizes corresponding to an average di-
ameter of 0.5–1.5 ?m, and in sum they covered 5–10% of the
available area in a typical nucleus. Because of their limited
geometrical extension and irregular shapes, it was very im-
probable to observe jumps covering distances greater than
0.5 ?m within these domains, although they occurred now
and then (see Figure 1A). Hence, a quantitative analysis of a
jump distance histogram was not reasonable. Above we
defined a maximum jump distance for immobile particles as
the threefold value of the localization precision, drmax? 100
nm. Therefore the fraction of mobile molecules can be esti-
mated by considering jumps over distance greater than
drmax. Within the speckles, we found 17% of all jumps to be
greater than drmax, whereas within the nucleoplasm, 24% of
all jumps were greater than drmax(see Figure 5). The reduc-
tion from 24 to 17% for nucleoplasmic space in comparison
to speckles can be attributed to the limited geometric exten-
sions of the speckles. Therefore, at this qualitative level there
was no indication of a mobility reduction of U1 snRNPs
within speckle domains.
In this study we analyzed the dynamics of U1 snRNPs
within live cells at the single-particle level. The used parti-
cles were isolated from HeLa cell nuclei and then labeled by
Cy5. Hence, our experiments refer to native RNPs, avoiding
the potential problems of in vitro–assembled particles,
which might be incomplete or dysfunctional. We microin-
jected Cy5-labeled U1 snRNPs at such low concentration
into the cytoplasm of living HeLa cells that we could visu-
alize single U1 snRNPs and track their movements within
the cell nucleus after their nuclear import. Real-time imaging
was achieved by a custom-built single-molecule microscope
equipped with laser illumination and a fast EMCCD for
fluorescence detection. The examined HeLa cells transiently
expressed ASF/SF2-GFP so that the particle dynamics could
be studied within and outside nuclear speckles. The SNR of
the U1 snRNP signals was in the range of 5–10, so that they
could be localized by SPT techniques with a spatial precision
of roughly 30 nm. Using automatic image analysis proce-
dures, we extracted the traces of numerous single mobile
and immobile U1snRNPs, which could be overlaid to the
ASF/SF2-GFP images of the corresponding cell nuclei.
The great majority of observed single U1 snRNPs was in
a bound state. Almost 80% of the U1 snRNPs observed in
two subsequent image frames did not move significantly.
This was in stark contrast to an inert, unfunctional tracer
cells. Jump distance distribution for all single U1 snRNPs, which
were identified and tracked within the nucleoplasm in the movies
acquired at 100 Hz. A minimum number of three diffusion terms
was required for an acceptable fit of the data. The largest fraction
(long dashes, 77.5%) corresponded to particles, which were immo-
bile. In other words, they performed only a virtual movement
caused by the limited localization precision. Here, we obtained a
value of ?immob? 50 nm. These jumps were done by those particles,
which were above identified as particles involved in a binding
process (Figures 2 and 3). Two particle fractions, f2and f3, corre-
sponded to fast diffusional motion with diffusion coefficients of
D2? 0.51 ? 0.05 ?m2/s (short dashes, 15%) and D3? 8.2 ? 3 ?m2/s
(dotted line, 7.3%). The sum of these fractions yielded an excellent
description of the data (full line). The inset shows the same plot with
a magnification of the y-scale in order to demonstrate the significant
number of large jumps due to particles moving with D3.
U1 snRNP mobility within the nucleoplasm of living
Single U1 snRNP Dynamics in Live Cells
Vol. 17, December 2006 5023
molecule inside cell nuclei (D. Gru ¨nwald, R. Martin, V.
Buschmann, H. Leonhardt, U. Kubitscheck, and M. C.
Cardoso, unpublished data), which showed a much greater
mobile fraction. Presumably, the immobilization was caused
by interactions with large, immobile molecular structures.
Obviously, being associated to large molecular structures is
the standard state of a splicing factor such as U1 snRNP. The
corresponding binding sites were very well defined, because
the particles did not move significantly beyond the experi-
mental localization precision. The bound RNPs were at-
tached at fixed sites, and they did not sway or move slowly
during their binding. Such small-range movements could be
excluded, because the immobilized single U1 snRNPs
showed a positional spreading only 5–10 nm greater than
that, which would theoretically follow from their SNR. This
insignificant increase was probably due to the intranuclear
fluorescence background. We assume that the observed im-
mobilization of the U1 snRNPs was often caused by the in-
volvement of the particles in ongoing splicing events. This
assumption was suggested by the fact that in a previous study
using digitonin-permeabilized cells, which were largely phys-
iologically inactive, a significantly smaller immobile fraction of
U1 snRNPs was found, namely only 22% (Kues et al., 2001a).
Pre-mRNA splicing is occurring often cotranscriptionally
(Melcak et al., 2000). This means that an extremely large DNA/
RNA–protein complex is formed, which would certainly not
have a notable mobility, but would rather represent a large,
anchored supramolecular complex.
The dissociation from the binding sites showed a surpris-
ing kinetics. Obviously the dissociation times were depen-
dent on the time resolution of observation. Analyzing bind-
ing at high frequency, we observed short binding durations,
whereas when observing binding at low frame rate, we
obtained long binding durations. This was consistent with
the fact that we obtained two decay times for each time
range analyzed. This already indicated that complex inter-
actions were observed. Altogether, dissociation times rang-
ing from 5 ms to 1400 ms were obtained. In case of a simple
molecular dissociation reaction one would expect a mono-
exponential decomposition of initially existing complexes
with a single dissociation time constant. Our data showed
that the interaction of U1 snRNPs with their intranuclear
binding partners was not a simple bimolecular interaction.
The dissociation of U1 snRNPs from binding sites occurred
over a wide time scale, which reflected the extremely com-
plex way, in which U1 snRNPs were interacting with addi-
tional molecular components. Possibly we perceived—be-
sides short nonspecific interactions, molecular trapping in a
chromatin network on the one hand and genuine splicing
events on the other hand—further processes, such as assem-
bly of spliceosomes before splicing and postsplicing process-
ing (Darzacq et al., 2005). We think that a decay kinetics on
many different time scales is a fundamental property of
recognition events and reactions of multistep molecular in-
teraction systems (Phair et al., 2004). In the splicing reaction,
a large number of different molecular components must act
together, which are preassembled in the form of spliceo-
somes and supraspliceosomes, which comprise U1 snRNPs
(Muller et al., 1998; Azubel et al., 2006). In addition, a great
number of different splicing reactions—simple and more
complex ones—are taking place simultaneously within a cell
nucleus. Therefore, a single dissociation constant could ac-
tually not be expected.
We quantified the kinetics of dissociation within and out-
side speckles. Speckles were defined on the basis of ASF/
SF2-GFP fluorescence, and distinct spots of strong GFP flu-
orescence were interpreted as speckles. Unfortunately, our
microscope was not an optical sectioning microscope. Non-
confocal videomicroscopy resulted in a rather diffuse ap-
pearance of the speckles and made the clear identification of
speckle borders difficult. On the basis of the chosen speckle
and nucleoplasm definitions, the dissociation kinetics on
short and long time scales did not differ significantly for
binding within the speckles compared with the remaining
nucleoplasm (Figure 3E). This finding supported the results
of Kruhlak et al. (2000), who did not detect major differences
in the dynamics of ASF/SF2-GFP in nucleoplasm and speck-
les. It could be concluded that the increase in splicing factor
concentration within the speckles was not due to an in-
creased dwell time of U1 snRNPs at intraspeckle sites. There
remain two possible explanations for the higher concentra-
tions of splicing factors within speckles. First, the on-rate of
the interaction is enhanced, which might be caused by a
enhanced accessibility of splicing factors to the speckles in
comparison to the remaining nucleoplasm. Second, the den-
sity of interaction sites is higher than in the remaining
nucleoplasmic space, whereas the interaction is of similar
nature. The latter hypothesis is supported by electron mi-
croscopic results (Puvion and Puvion-Dutilleul, 1996;
Lamond and Spector, 2003). We interpreted a part of the
observed binding sites as sites of on-going transcription.
Together with the above observation this would suggest that
splicing is taking place also in speckles, as was previously
found by various researchers (Wei et al., 1999; Melcak et al.,
2000; Shopland et al., 2002). However, a clear-cut statement
on this question is problematic, because the fluorescence
microscopic discrimination between interchromatin granule
clusters forming the speckled compartments and highly ac-
tive transcription sites with increased levels of pre-mRNA
splicing factors is problematic (Lamond and Spector, 2003).
Altogether, the reason for the complex immobilization
could not yet finally be resolved. However, the large differ-
ence in the size of the interacting U1snRNP fraction in live
cells compared with digitonin-permeabilized cells (Kues et
al., 2001a) underscores the requirement of live cell measure-
ments when studying such intricate physiological processes.
Further studies using inhibition of transcription respectively
splicing will provide more insight into the functional rele-
vance of immobilization events.
Finally, it should be noted that dissociation data like that
shown in Figure 3 are usually available only by a special
synchronization of the molecular complexes in the initial,
associated state. This often presents a problem, which is very
difficult or impossible to solve, especially when working in
vivo. Single-molecule detection, however, can elegantly re-
solve this problem. The accumulation of the data here took
advantage of a special feature of single-molecule research:
the observation of individual molecular interaction events
did not require a synchronization of a molecular ensemble
(Weiss, 1999), because the individual events could be
aligned in time a posteriori (Kubitscheck et al., 2005).
The presented data created an entirely new view of the
molecular dynamics of a functional molecular entity during
ongoing live processes. In almost all recent studies on the
mobility of functional molecules within cell nuclei binding
processes were postulated and accordingly modeled in or-
der to account for mobility data obtained by photobleaching,
photoactivation, or FCS techniques (Wachsmuth et al., 2000;
Houtsmuller and Vermeulen, 2001; Carrero et al., 2003;
Wachsmuth et al., 2003; Phair et al., 2004; Beaudouin et al.,
2006). However, often models of anomalous diffusion were
also able to explain the data (e.g., Wachsmuth et al., 2000). In
this study using single-particle tracking binding events
could unambiguously be observed and thus be proven and
D. Gru ¨nwald et al.
Molecular Biology of the Cell 5024
be discriminated from other modes of motion. We did not
only observe distinct binding events, but could also measure
the durations of individual interactions. Thereby we found
that the kinetics of dissociation cannot be described by a
single dissociation constant, but rather ranges over more
than three orders of magnitude.
Single-molecule microscopy is especially well suited to
follow molecular traces in time (Saxton and Jacobson, 1997).
Recently we demonstrated that a frame rate of 350 Hz was
sufficient to track single protein molecules such as antibod-
ies and streptavidin molecules in buffer exhibiting diffusion
coefficients as high as 40 and 80 ?m2/s, respectively. There-
fore we could assume that the maximum repetition rate of
200 Hz used in this study was high enough to track the U1
snRNPs with a molecular weight of 240 kDa in real time,
especially because a 5- to 10-fold mobility reduction in cells
compared with buffer solution could be expected according
to previous FRAP studies (Lang et al., 1986; Seksek et al.,
1997; Braga et al., 2004). By analyzing the jumps of single U1
snRNPs between subsequent frames, we obtained a general
insight into the dynamic behavior of the particles within the
cell nuclei. As shown already by previous studies using FCS
and SPT the mobility of single molecules within cell nuclei
cannot be characterized by simple Brownian motion
(Goulian and Simon, 2000; Wachsmuth et al., 2000; Kues et
al., 2001b). We could discriminate one immobile and at least
two mobile fractions. We want to emphasize here that the
dissection of the jump distance distribution into three frac-
tions represented a minimum number. Three fractions were
sufficient to fit the data in Figure 5. However, the fit could have
been further improved by assuming more than three fractions.
For the following discussion of the mobile fractions one should
also keep in mind that the fractions do not resemble different
particles with distinct mobilities. Rather—as could be noted in
the trajectory analyzed in Figure 1C—single particles switched
their mode of motion along their trajectory. Hence, the frac-
tions identified in the jump distance histograms represent dif-
ferent modes of motion of possibly identical particles. Alto-
gether, we suppose that distinct mobility fractions do not exist,
but rather that the U1 snRNP mobility ranges from 0.5 to 8
?m2/s in a continuous distribution.
In the mobility analysis particles with mobilities ranging
from 0.5 to 8 ?m2/s were observed. A diffusion coefficient of
8 ?m2/s is four- to fivefold lower than that expected for a
240-kDa protein in aqueous solution. A fourfold reduction in
mobility was also found in previous mobility studies of
tracer molecules within cell nuclei performed by FRAP (Sek-
sek et al., 1997). Therefore, we assume that the high mobility
was shown by uncomplexed U1 snRNPs, which moved
within nuclei as in a solution with an effective viscosity of 5
cPoise compared with aqueous buffer of 1 cPoise. In vivo all
U snRNPs are central components of preformed complexes
designated as spliceosomes. It has been shown that these
complexes occur in a structure termed supraspliceosome,
which contains four native spliceosomes, with a total mass
of 21 MDa and dimensions of 50 ? 50 ? 35 nm3(Sperling et
al., 1997; Muller et al., 1998; Medalia et al., 2002). Such com-
plexes, if moving by unrestricted Brownian motion, would
have a diffusion constant of about D ? 2 ?m2/s in a solution
of 5-cPoise viscosity. This value for supraspliceosomes lies
in the range of diffusion constants determined for mobile U1
snRNPs. It is very unlikely that large objects like supraspli-
ceosomes would move in an unrestricted manner within the
molecular crowded intranuclear space. Rather, they would
be prone to multiple collisions or interactions with chroma-
tin or other large structures, which would slow it down. This
has been reported for large dextran molecules with a molec-
ular mass of 580 kDa to 2 MDa. Their mobility was depen-
dent on the concentration of intracellular obstacles (Seksek et
al., 1997). Furthermore, it has been shown that chromatin
regions represent a significant obstruction for the accessibil-
ity of large probe molecules (Gorisch et al., 2003). Hence,
jumps corresponding to a mobility as low as D ? 0.5 ?m2/s
could well correspond to uncomplexed U1 snRNPs or to U1
snRNPs contained in spliceosomes and supraspliceosomes
tumbling in a hindered manner through the nucleoplasm.
The established way to analyze intracellular mobility is by
FRAP, which measures bulk mobility on a spatial scale of
several micrometers in a time window of a few to 50 s. On
the other hand, SPT quantifies mobility of individual mole-
cules on length scales significantly smaller than 1 ?m in time
windows of milliseconds to seconds. It is not straightfor-
ward to extrapolate from the single-molecule data to the
results of FRAP measurements. To accomplish this a respec-
tive simulation of bulk mobility on the basis of the single-
particle data would be required, which has not been done
yet. However, we can correlate our results with FRAP re-
sults in the following manner. FRAP detected a three- to
fivefold reduction in D for larger tracer molecules within the
nuclei and a significantly more pronounced reduction for
molecules with specific intranuclear interactions (Gorski et
al., 2006). Our SPT data also reveal a four- to fivefold reduc-
tion in D for mobile, noninteracting, and up to an 70-fold
reduction for presumably interacting U1snRNPs. Finally, we
obtained a very detailed and quantitative view to the inter-
actions of a U1snRNP with immobilizing binding partners
on a subsecond time scale.
Single-molecule microscopy and single-particle tracking
permits a completely new view to intracellular dynamics. It
shows that splicing factor dynamics inside living cells is
extremely complex. U snRNPs move freely, are incorporated
into huge supramolecular complexes such as supraspliceo-
somes, attach to binding sites for extended periods of time,
and are released again. Binding and dissociation occurs
obviously under widely varying kinetic conditions. A de-
tailed analysis of long single-particle trajectories, use of sev-
eral fluorescent labels in parallel, and the combination with
complementary techniques such as FCS and quantitative
photobleaching techniques will provide further insights into
complex in vivo processes such as RNA processing.
U.K. is heavily indebted to Reiner Peters for using numerous experimental
facilities in his lab, and thanks M. Cristina Cardoso for helpful discussions
and Jan-Peter Siebrasse for critical reading of the manuscript. U.K. gratefully
acknowledges financial support by the Volkswagen Foundation (Grant
I/78852) and the German Research Foundation (Grant Ku 975/3-3).
Azubel, M., Habib, N., Sperling, R., and Sperling, J. (2006). Native spliceo-
somes assemble with pre-mRNA to form supraspliceosomes. J. Mol. Biol. 356,
Babcock, H. P., Chen, C., and Zhuang, X. (2004). Using single-particle tracking
to study nuclear trafficking of viral genes. Biophys. J. 87, 2749–2758.
Bausinger, R., von Gersdorff, K., Braeckmans, K., Ogris, M., Wagner, E.,
Brauchle, C., and Zumbusch, A. (2006). The transport of nanosized gene
carriers unraveled by live-cell imaging. Angew. Chem. Int. Ed. Engl. 45,
Beaudouin, J., Mora-Bermudez, F., Klee, T., Daigle, N., and Ellenberg, J.
(2006). Dissecting the contribution of diffusion and interactions to the mobility
of nuclear proteins. Biophys. J. 90, 1878–1894.
Braga, J., Desterro, J. M., and Carmo-Fonseca, M. (2004). Intracellular macro-
molecular mobility measured by fluorescence recovery after photobleaching
with confocal laser scanning microscopes. Mol. Biol. Cell 15, 4749–4760.
Single U1 snRNP Dynamics in Live Cells
Vol. 17, December 20065025
Brock, R., Hink, M. A., and Jovin, T. M. (1998). Fluorescence correlation
microscopy of cells in the presence of autofluorescence. Biophys. J. 75, 2547–
Bubulya, P. A., and Spector, D. L. (2004). “On the move”ments of nuclear
components in living cells. Exp. Cell Res. 296, 4–11.
Carrero, G., McDonald, D., Crawford, E., de Vries, G., and Hendzel, M. (2003).
Using FRAP and mathematical modeling to determine the in vivo kinetics of
nuclear proteins. Methods 29, 14–28.
Crank, J. (1975). The Mathematics of Diffusion, Oxford: Clarendon Press.
Cremer, T., and Cremer, C. (2001). Chromosome territories, nuclear architec-
ture and gene regulation in mammalian cells. Nat. Rev. Genet. 2, 292–301.
Darzacq, X., Singer, R. H., and Shav-Tal, Y. (2005). Dynamics of transcription
and mRNA export. Curr. Opin. Cell Biol. 17, 332–339.
Gorisch, S. M., Richter, K., Scheuermann, M. O., Herrmann, H., and Lichter,
P. (2003). Diffusion-limited compartmentalization of mammalian cell nuclei
assessed by microinjected macromolecules. Exp. Cell Res. 289, 282–294.
Gorski, S., and Misteli, T. (2005). Systems biology in the cell nucleus. J. Cell
Sci. 118, 4083–4092.
Gorski, S. A., Dundr, M., and Misteli, T. (2006). The road much traveled:
trafficking in the cell nucleus. Curr. Opin. Cell Biol. 18, 284–290.
Goulian, M., and Simon, S. M. (2000). Tracking single proteins within cells.
Biophys. J. 79, 2188–2198.
Grunwald, D., Hoekstra, A., Dange, T., Buschmann, V., and Kubitscheck, U.
(2006). Direct observation of single protein molecules in aqueous solution.
Chemphyschem 7, 812–815.
Houtsmuller, A. B., Rademakers, S., Nigg, A. L., Hoogstraten, D., Hoeijmakers,
J. H., and Vermeulen, W. (1999). Action of DNA repair endonuclease ERCC1/
XPF in living cells. Science 284, 958–961.
Houtsmuller, A. B., and Vermeulen, W. (2001). Macromolecular dynamics in
living cell nuclei revealed by fluorescence redistribution after photobleaching.
Histochem. Cell Biol. 115, 13–21.
Huber, J., Cronshagen, U., Kadokura, M., Marshallsay, C., Wada, T., Sekine,
M., and Luhrmann, R. (1998). Snurportin1, an m3G-cap-specific nuclear im-
port receptor with a novel domain structure. EMBO J. 17, 4114–4126.
Kleinschmidt, A. M., and Pederson, T. (1990). RNA processing and ribonu-
cleoprotein assembly studied in vivo by RNA transfection. Proc. Natl. Acad.
Sci. USA 87, 1283–1287.
Kruhlak, M. J., Lever, M. A., Fischle, W., Verdin, E., Bazett-Jones, D. P., and
Hendzel, M. J. (2000). Reduced mobility of the alternate splicing factor (ASF)
through the nucleoplasm and steady state speckle compartments. J. Cell Biol.
Kubitscheck, U., Grunwald, D., Hoekstra, A., Rohleder, D., Kues, T.,
Siebrasse, J. P., and Peters, R. (2005). Nuclear transport of single molecules:
dwell times at the nuclear pore complex. J. Cell Biol. 168, 233–243.
Kubitscheck, U., Ku ¨ckmann, O., Kues, T., and Peters, R. (2000). Imaging and
tracking of single GFP molecules in solution. Biophys. J. 78, 2170–2179.
Kues, T., Dickmanns, A., Lu ¨hrmann, R., Peters, R., and Kubitscheck, U.
(2001a). High intranuclear mobility and dynamic clustering of the splicing
factor U1 snRNP observed by single particle tracking. Proc. Natl. Acad. Sci.
USA 98, 12021–12026.
Kues, T., Peters, R., and Kubitscheck, U. (2001b). Visualization and tracking of
single protein molecules in the cell nucleus. Biophys. J. 80, 2954–2967.
Lamond, A. I., and Spector, D. L. (2003). Nuclear speckles: a model for nuclear
organelles. Nat. Rev. Mol. Cell Biol. 4, 605–612.
Lang, I., Scholz, M., and Peters, R. (1986). Molecular mobility and nucleocy-
toplasmic flux in hepatoma cells. J. Cell Biol. 102, 1183–1190.
Lukacs, G. L., Haggie, P., Seksek, O., Lechardeur, D., Freedman, N., and
Verkman, A. S. (2000). Size-dependent DNA mobility in cytoplasm and
nucleus. J. Biol. Chem. 275, 1625–1629.
Marshallsay, C., and Luhrmann, R. (1994). In vitro nuclear import of snRNPs:
cytosolic factors mediate m3G-cap dependence of U1 and U2 snRNP trans-
port. EMBO J. 13, 222–231.
Medalia, O., Typke, D., Hegerl, R., Angenitzki, M., Sperling, J., and Sperling,
R. (2002). Cryoelectron microscopy and cryoelectron tomography of the nu-
clear pre-mRNA processing machine. J. Struct. Biol. 138, 74–84.
Melcak, I., Cermanova, S., Jirsova, K., Koberna, K., Malinsky, J., and Raska, I.
(2000). Nuclear pre-mRNA compartmentalization: trafficking of released tran-
scripts to splicing factor reservoirs. Mol. Biol. Cell 11, 497–510.
Misteli, T. (2005). Concepts in nuclear architecture. Bioessays 27, 477–487.
Moerner, W. (2003). Optical measurements of single molecules in cells. TRAC-
Trends Analyt. Chem. 22, 544–548.
Muller, S., Wolpensinger, B., Angenitzki, M., Engel, A., Sperling, J., and
Sperling, R. (1998). A supraspliceosome model for large nuclear ribonucleo-
protein particles based on mass determinations by scanning transmission
electron microscopy. J. Mol. Biol. 283, 383–394.
Pederson, T. (2004). The molecular cytology of gene expression: fluorescent
RNA as both a stain and tracer in vivo. Eur. J. Histochem. 48, 57–64.
Phair, R. D., and Misteli, T. (2000). High mobility of proteins in the mamma-
lian cell nucleus. Nature 404, 604–609.
Phair, R. D., Scaffidi, P., Elbi, C., Vecerova, J., Dey, A., Ozato, K., Brown, D. T.,
Hager, G., Bustin, M., and Misteli, T. (2004). Global nature of dynamic
protein-chromatin interactions in vivo: three-dimensional genome scanning
and dynamic interaction networks of chromatin proteins. Mol. Cell. Biol. 24,
Politz, J. C., Browne, E. S., Wolf, D. E., and Pederson, T. (1998). Intranuclear
diffusion and hybridization state of oligonulceotides measured by fluores-
cence correlation spectroscopy in livings cells. Proc. Natl. Acad. Sci. USA 95,
Politz, J. C., Tuft, R. A., Prasanth, K. V., Baudendistel, N., Fogarty, K. E.,
Lifshitz, L. M., Langowski, J., Spector, D. L., and Pederson, T. (2006). Rapid,
diffusional shuttling of poly(A) RNA between nuclear speckles and the nu-
cleoplasm. Mol. Biol. Cell 17, 1239–1249.
Puvion, E., and Puvion-Dutilleul, F. (1996). Ultrastructure of the nucleus in
relation to transcription and splicing: roles of perichromatin fibrils and inter-
chromatin granules. Exp. Cell Res. 229, 217–225.
Sage, D., Neumann, F. R., Hediger, F., Gasser, S. M., and Unser, M. (2005).
Automatic tracking of individual fluorescence particles: application to the
study of chromosome dynamics. IEEE Trans. Image Process. 14, 1372–1383.
Sako, Y., and Yanagida, T. (2003). Single-molecule visualization in cell biol-
ogy. Nat. Cell Biol. Ss1–Ss5.
Saxton, M. J., and Jacobson, K. (1997). Single-particle tracking: applications to
membrane dynamics. Annu. Rev. Biophys. Biomol. Struct. 26, 373–399.
Schwille, P., Haupts, U., Maiti, S., and Webb, W. W. (1999). Molecular dy-
namics in living cells observed by fluorescence correlation spectroscopy with
one- and two-photon excitation. Biophys. J. 77, 2251–2265.
Schwille, P., and Kettling, U. (2001). Analyzing single protein molecules using
optical methods. Curr. Opin. Biotechnol. 12, 382–386.
Seisenberger, G., Ried, M. U., Endress, T., Buning, H., Hallek, M., and
Brauchle, C. (2001). Real-time single-molecule imaging of the infection path-
way of an adeno-associated virus. Science 294, 1929–1932.
Seksek, O., Biwersi, J., and Verkman, A. S. (1997). Translational diffusion of
macromolecule-sized solutes in cytoplasm and nucleus. J. Cell Biol. 138,
Shav-Tal, Y., Darzacq, X., Shenoy, S. M., Fusco, D., Janicki, S. M., Spector,
D. L., and Singer, R. H. (2004). Dynamics of single mRNPs in nuclei of living
cells. Science 304, 1797–1800.
Shopland, L. S., Johnson, C. V., and Lawrence, J. B. (2002). Evidence that all
SC-35 domains contain mRNAs and that transcripts can be structurally con-
strained within these domains. J. Struct. Biol. 140, 131–139.
Sleeman, J., Lyon, C. E., Platani, M., Kreivi, J. P., and Lamond, A. I. (1998).
Dynamic interactions between splicing snRNPs, coiled bodies and nucleoli
revealed using snRNP protein fusions to the green fluorescent protein. Exp.
Cell Res. 243, 290–304.
Spann, P., Feinerman, M., Sperling, J., and Sperling, R. (1989). Isolation and
visualization of large compact ribonucleoprotein particles of specific nuclear
RNAs. Proc. Natl. Acad. Sci. USA 86, 466–470.
Sperling, R., Koster, A. J., Melamed-Bessudo, C., Rubinstein, A., Angenitzki,
M., Berkovitch-Yellin, Z., and Sperling, J. (1997). Three-dimensional image
reconstruction of large nuclear RNP (lnRNP) particles by automated electron
tomography. J. Mol. Biol. 267, 570–583.
Sperling, R., Sperling, J., Levine, A. D., Spann, P., Stark, G. R., and Kornberg,
R. D. (1985). Abundant nuclear ribonucleoprotein form of CAD RNA. Mol.
Cell. Biol. 5, 569–575.
Thompson, R. E., Larson, D. R., and Webb, W. W. (2002). Precise nanometer
localization analysis for individual fluorescent probes. Biophys. J. 82, 2775–
Tinnefeld, P., and Sauer, M. (2005). Branching out of single-molecule fluores-
cence spectroscopy: challenges for chemistry and influence on biology. An-
gew. Chem. Int. Ed. Engl. 44, 2642–2671.
Vallotton, P., Ponti, A., Waterman-Storer, C. M., Salmon, E. D., and Danuser,
G. (2003). Recovery, visualization, and analysis of actin and tubulin polymer
D. Gru ¨nwald et al.
Molecular Biology of the Cell5026
flow in live cells: A fluorescent speckle microscopy study. Biophys. J. 85,
Verkman, A. S. (2002). Solute and macromolecule diffusion in cellular aque-
ous compartments. Trends Biochem. Sci. 27, 27–33.
Wachsmuth, M., Waldeck, W., and Langowski, J. (2000). Anomalous diffusion
of fluorescent probes inside living cell nuclei investigated by spatially-re-
solved fluorescence correlation spectroscopy. J. Mol. Biol. 298, 677–689.
Wachsmuth, M., Weidemann, T., Muller, G., Hoffmann-Rohrer, U., Knoch, T.,
Waldeck, W., and Langowski, J. (2003). Analyzing intracellular binding and
diffusion with continuous fluorescence photobleaching. Biophys. J. 84, 3353–
Wei, X., Somanathan, S., Samarabandu, J., and Berezney, R. (1999). Three-
dimensional visualization of transcription sites and their association with
splicing factor-rich nuclear speckles. J. Cell Biol. 146, 543–558.
Weiss, S. (1999). Fluorescence spectroscopy of single biomolecules. Science
Will, C. L., and Luhrmann, R. (2001). Spliceosomal UsnRNP biogenesis,
structure and function. Curr. Opin. Cell Biol. 13, 290–301.
Single U1 snRNP Dynamics in Live Cells
Vol. 17, December 20065027