Sample Preparation Guidelines for Photo-activated
Localization Microscopy (PALM) Imaging
1
1.1
Introduction to PALM imaging
General considerations
This treatise is meant to give some guidelines to the preparation of samples suitable for photoactivated localization microscopy (PALM) as well as PALM imaging. It is not meant as an
exhaustive handbook but rather a starting point from which the user can refine his protocols
according to published methods. Sections preceding detailed protocols give some orientation
and rational behind the presented recipes and cover brief surveys of PALM and related
methods, the required instrumentation as well as data acquisition and analysis. The following
detailed protocols and recipes for buffer solutions used in PALM imaging will give step by
step guidance how to prepare the sample before mounting it on the PALM microscope. The
subsequent list of suppliers and references to chemicals and other materials should help
finding the right reagent, its necessary quality and other useful tools. However, the list does
not aim to be complete nor does it imply certain preferences. Other suppliers might offer
products of the same quality and price. So it is worth while to search for local dealers.
Since in PALM molecules are sparsely activated and imaged many rules governing single
molecule detection (SMD) apply for PALM as well especially in regard to the care one has to
take in sample preparation, the purity of reagents used and the cleaning of the material
employed. Background artifacts can be severe and should be excluded as best as possible. The
trouble shooting section lists common error sources and how they can be avoided. Last not list
the provided further reading section and the literature list will give references to specialized
and more advanced PALM technologies. We apologize for those not cited due to space and
content restrictions.
We hope that the kind reader finds this little booklet of interest and help.
1.2
Principles of PALM and related methods
Albeit fluorescent labeling technologies allow targeting of proteins with high specificity their
precise localization is limited in conventional far-field imaging by the diffraction nature of
light. The smallest feature that can be resolved in the lateral image plane will be
approximately half of the wavelength of the illumination light [Abbe, 1873 #153]. For visible
light this corresponds to approximately 200 – 300 nm [Born, 1999 #634]. In response, a
variety of far-field super-resolution techniques have evolved that overcome this restriction
[Schermelleh, 2010 #359]. There have been two general ideas to overcome the diffraction
barrier. One is to temporarily create a sub-diffraction excitation pattern be it by interference as
in Structured Illumination Microscopy (SIM) (literature) or by fluorescence depletion as in
Stimulated Emission Depletion (STED) [Hell, 1994 #189]; the other is by stochastic photoswitching of molecules as realized in Photoactivated Localization Microscopy (PALM)
[Betzig, 2006 #92] or super-resolution optical fluctuation imaging (SOFI) [Dertinger, 2009
#379]. If non-linear optical effects are employed all concepts potentially can approach
resolutions down to the molecular scale [Hell, 2007 #133]. There are many more other
technologies available nowadays with different acronyms, but that all draw as well on either
of the two principles.
Fig. 1 The point spread function (PSF) with its diffraction limited extension can be fitted to a Gaus function to yield a localization precision (LP) with profoundly less uncertainty if the photon distribution originates from only one emitter. PALM relies on the stochastic activation,
localization and deactivation of single
fluorescent molecules. Localization
precisions (LP) are obtained by fitting the
photon distribution of an individual
emitter to an ideal point spread function
(PSF) [Thompson, 2002 #553] (Fig. 1).
Since the fitted localization precision is
directly proportional to the photon number,
bright emitters and high laser powers are
preferentially used or even necessary for
PALM imaging. Also, for highest
precisions and to avoid mislocalizations it
is advantageous that on average only one
molecule is emitting per diffraction limited
spot, although fitting algorithms are
available that allow for fitting overlapping
PSFs, however, at the cost of precision
[Holden, #735].
The cycle of activation-readoutdeactivation is repeated by recording a
time series of images. The super-resolution
image is than reconstructed from all
recorded images (typically thousands to
tens of thousands) with all their
localizations (typically tens of thousands to
millions) (Fig. 2).
Fig. 2 In PALM imaging molecules are sparsely activated (one per PSF) and quickly deactivated per cycle. The PALM image is reconstructed from the sum of all localizations obtained per each cycle. If too many molecules are in their on-‐state in the wide-‐field image (WF) they have to be first brought to their off-‐state before starting the activation / deactivation cycles. PALM used initially photo-switchable
fluorescent proteins (PS-FPs), also called
optical highlighter proteins, at high
densities in a total internal reflection
fluorescence (TIRF) illumination scheme
with its outstanding signal-to-noise ratio
(SNR) to achieve resolutions more than 10
fold higher compared to conventional
imaging [Wiedenmann, 2006 #658]. Other
related methods relying on the same
principles used variations in the
illumination and / or switching regime. In
fluorescent photo-activated localization
microscopy (FPALM) epi-fluorescence
(EPI) illumination is used necessitating
lower labeling densities and hence
achieving lower resolutions, but being able
to assess cellular structures well above the lower cellular membrane [Hess, 2006 #633]. In
stochastical optical reconstruction microscopy (STORM) organic dyes rather than fluorescent
proteins have been employed and switching of an acceptor dye is achieved via Foerster
resonance energy transfer (FRET) provided by a suitable donor [Rust, 2006 #100]. On the
other hand, the photochemistry of organic dyes has been directly used for photo-switching in
an implementation called direct stochastical optical reconstruction microscopy (dSTORM)
[Heilemann, 2008 #107] later also described as ground state depletion microscopy followed
by individual molecule return (GSDIM) [Folling, 2008 #244].
Common to all single molecule based localization methods is the need of fluorophores that
can be switched between two different states, commonly an “on-state” and an “off-state”. The
on-state could be, for example, a bright or a certain spectral emission state, whereas the offstate would than represent the dark or another spectral emission state, respectively. As
mentioned above, the most prominent employed fluorophores for PALM imaging are photoswitchable fluorescent proteins (PS-FPs) [Lippincott-Schwartz, 2009 #414] and organic dyes
[Heilemann, 2009 #102]. These two classes of fluorescent molecules are quite different in
their chemistry and require different staining and imaging conditions.
1.3
Photo-switching mechanisms
In typical cases PS-FPs
can be switched from a
dark (or one spectral) offstate by illuminating with
low amounts of activation
light (normally violet light
at 405 nm) to a bright (or
another spectral) on-state
and are converted back to
the off- state, e.g. by
photo-bleaching or
reversible photo-switching,
achieved by high power of
the respective excitation
light. PS-FPs fall in the
majority in either of three
classes: photo-activatable,
photo-convertible &
photo-chromic [Patterson,
2011 #742]. The two
former cases are based on
an irreversible photoFig. 3 Switching mechanisms for fluorescent proteins (FPs). Photo-‐
switch by chemical
activation is an irreversible photo-‐switch from a “dark” to a “bright” state by chemical modification. Photo-‐conversion is an modification and
irreversible photo-‐switch from “one” to “another” spectral backbone cleavage from a
state. Photo-‐chromicity is a reversible photo-‐switch between a “dark” and “bright” state by cis-‐trans isomerization. One dark off- to a bright ondistinguishes a between a positive and a negative (shaded with state or one spectral to
a grey background) reversible photo-‐switch. another spectral state,
respectively, whereas the latter is a reversible photo-switch between a dark and a bright state
that draws on cis-trans isomerization reaction (Literaturzitat) (Fig. 3). Combinations of these
mechanisms are also possible [Brakemann, 2011 #745]. Photochromic proteins might be
switched to the bright state by violet light and be converted back to the dark state by their
excitation light (positive switch) or might be switched to the dark state by violet light and be
turned on by their excitation light (negative switch).
By respective modifications of amino acids in the chromophore group or close by regions
many PS-FPs with new properties have been successfully designed (Tab. 1). Since PS-FPs are
essentially non-fluorescent (or fluorescent at a different spectral emission band) at the start of
the experiment before conversion, the density of molecules in the on-state can easily be fine
tuned by balancing the powers between the activation and excitation lasers, respectively.
Fig. 4 Reversible photo-‐switching of dyes in the presence of thiols. A fluorophore can either cycle between the singlet ground state (S0) and the singlet excited state (S1) emitting photons or can undergo system crossing with rate kisc upon irradiation to the triplet state (T1), a process that is promoted by excitation saturation. The triplet state can either react with molecular oxygen with rate kpho thereby converting the fluorophore back to its singlet state, or it can react with thiolate with rate kred to form the radical anionic fluorophore (T*) and the corresponding thiyl radical, which readily reacts with molecular oxygen to superoxide radicals and hydrogen peroxide. In contrast, the radical anionic form is chemically largely inert and can last for more than several seconds. Eventually it reacts with molecular oxygen to recover the singlet state with rate constant kox. Irradiation with violet light (around 405 nm) will promote this recovery as most anionic derivatives show pronounced adsorption in this spectral range. Some fluorophores can be further reduced to the leuco-‐form (TH) that can also recover to the singlet state by the same mechanism. In contrast, organic dyes are mostly in their on-state and the majority of the molecules have to
be turned to a relatively stable and reversible non-fluorescent off-state by irradiation with
sufficiently high laser powers of the excitation light [van de Linde, 2010 #757]. Thereby,
photo-bleaching and damage to the sample must be avoided. Photo-switching rates can be
tuned by adding different concentrations of reducing thiol containing compounds in and
tuning the pH of the buffer to stabilize the off-state [van de Linde, 2011 #751]. This is
accomplished by the reduction of the triplet state to a non-fluorescent stable radical anion,
which can be further reduced, with rate koff (Fig. 4). Turning back to the on-state is
accomplished by oxidation at rate kon and is supported by illumination with a wavelength
shorter than the excitation wavelength by helping to return the reduced species back to the
ground state. Both rates follow a linear dependence to the light intensity. Therefore the koff
rate is be controlled by the thiol concentration, singlet state excitation intensity and
determining the intersystem crossing yield, whereas the kon rates can be tuned by the radical
excitation intensity and the molecular oxygen concentration [van de Linde, 2011 #754].
Glutathione concentrations within a living cell are fortunately compatible for dSTORM
experiments, allowing studying the distribution and dynamics of proteins [Wombacher, #737].
The koff/kon ratio will in the end determine how many molecules are in the on- and off-states,
respectively. High ratio values guarantee that only few molecules will be in their on-state,
which is favorable to localization microscopy. Hence, longer off-state lifetimes (in the
seconds range), equivalent to shorter koff rates, require lower laser powers of the imaging laser
than working with dyes of shorter off-state lifetimes (ms-range) [Vogelsang, 2010 #766]. It
should be kept in mind that high laser powers can induce photo-toxicity in life cell imaging
often limiting the use of dyes of short off-state. In practical terms a laser with a maximum
power of ~100 mW is sufficient to control koff rates by an order of magnitude. The
concentration of the reducing thiolate (RS-) can be controlled by both, the concentration of the
thiol compound added and the pH, which can be adjusted between pH 6 to 9 and allows
tuning koff by 2 orders of magnitude in addition [Heilemann, 2009 #102]. The on-rate, on the
other hand, can be controlled by an order of magnitude by tuning the laser excitation for the
radical anion, e.g. the 405 nm line, by several orders of magnitude by employing an oxygen
scavenging system to deplete the oxygen leading to a profound stabilization of the radical
anion’s lifetime [van de Linde, 2011 #754]. A large variety of organic dyes have been proved
already to be suitable for photo-switching (Tab. 2).
To maximize the obtained photon numbers, frame rates are adjusted to the average photobleaching or conversion times being typically in the range of 10 - 25 Hz (100 - 40 ms frame
rate) for PS-FPs [Gould, 2009 #232; Shroff, 2008 #86; Shroff, 2008 #87]. With organic dyes
frame rates of 10 – 1000 Hz (100 – 1 ms frame rate) at excitation intensities between 0.5 and
50 kW cm-2 can be achieved. The strengths of PS-FPs lie in their outstanding specificity and
their small size, which is around 2 nm (Fig. 5). The latter feature guarantees potentially high
labeling densities. In contrast, antibodies are in the size range of 10 nm and the situation
worsens, when secondary antibody systems will be used. The advent of nanobodies (cameloid
like antibodies from camels, llamas and sharks) with their sizes in the range of 2 nm, however,
is starting to improve this situation [Ries, 2012 #410]. One disadvantage of PS-FPs are their
lower photo-stability and brightness when compared to organic dyes [Shaner, 2008 #292;
Tang, 2010 #411]. The former can emit a few hundreds of photons per cycle, the latter more
than a thousand. Since the localization precision (LP) - which can be approximated for
negligible background and detector noise by LP ≈ σ
(N being the photon number and σ
N
the standard deviation of the PSF) [Thompson, 2002 #412] - is inversely proportional to the
number of photons using organic dyes can potentially yield higher precisions, but due to
lower labeling densities not necessarily better resolutions.
1.4
Achievable Resolutions
According to the Nyquist-Shannon theory, the distance between neighboring localized
emitters must be at least twice as fine as the desired resolution [Shannon, 1949 #409]. For
example, if the resolution should be 20 nm, the fluorophores should be apart by not more than
10 nm in a linear array. Extending to the two- and three-dimensional cases this corresponds to
a density of 900 fluorophores / µm2 or 8100 fluorophores / µm3 within the diffraction limited
area assuming an extension of the PSF of 300 nm and 900 nm in the lateral and axial
directions, respectively. This also would mean that only one of these 900 or 8100
fluorophores is in its fluorescent bright on-state implying that the dark off-state must be 900
or 8100 fold longer than the lifetime of the former. Under experimental conditions such
labeling densities are hard to approach, however, and must as a matter of fact be effectually
smaller in order for reliable peak selection and fitting. But in principle smaller labels are
preferred as they allow for higher labeling densities and hence higher resolutions. As such FPs
bear the potential of higher resolutions than antibody staining (Fig. 5).
With achievable
resolutions down to 10 nm
the label size will affect
the measured size of the
structure under
investigation that often
itself is in the size range of
an antibody or even
smaller. E.g. the diameter
of microtubules is
estimated from electron
microscopy to be in the
order of 25 nm, yet in
super-resolution using
antibody staining it is
measured as approximately
45 nm (10).
To obtain an estimate on
the obtained resolution one
has to take into account the
labeling density (LD) and
the obtained localization
precision (LP). LP is the
direct outcome of the fit.
Fig. 5 Sizes of common labels drawn to scale. Depicted are the organic The labeling density can
dye Cy5 (dye), FlAsH binding to self-‐labeling TC-‐tag (tag), a quantum dot with emission around 620 nm (Q-‐dot), be estimated by drawing a
streptavidin-‐biotin (S/B), the fluorescent protein GFP (FP), a region of interest (ROI)
nanobody (Nb) and an antibody (Ab) with its F and F fragments. and dividing the number of
molecules in that region by the area. The resolution equals the localization precision, if the
labeling density fulfills the Nyquist-Shannon criterion; otherwise it will be twice the labeling
density.
ab
c
2
2.1
Strategic planning
PALM Instrumentation
To minimize sample drift as much as possible the system should be mounted on an vibrationisolated system table, ideally with active dampening. The fluorescence microscope set up
should be optimized to obtain very high fluorescence signals, that is collecting as many
photons as possible from the fluorophore. On the other hand background arising from
background fluorescence and detector noise should be as low as possible (19). Such demands
are best addressed by using wide-field illumination with an objective-type total internal
reflection fluorescence (TIRF) regime on an inverted fluorescence microscope equipped with
an oil-immersion objective of high numerical aperture (NA ≥ 1.45) (Fig. 6). Such a set up
allows a parallelization of single-molecule fluorescence imaging with a high signal-to-noise
ratio (SNR). The evanescent field produced in TIRF imaging has a penetration depth of only a
few 100 nm minimizing out-of-focus light contributions. However, this will limit the
observation depth in z to regions close to the lower membrane of the cell. Extended
observation throughout the cell needs a wide-field configuration at the expense of a lower
SNR. Deeper penetration depth is also accomplished by an illumination scheme called high
inclined and laminated optical sheet (HILO) illumination that produces inclined sheet
illumination and hence has sectioning effects (Fig. 7) (57). For whole cell and 3D imaging
epi-fluorescence illumination (EPI) at the expense of a lower signal-to-noise-ratio (SNR) has
to be used (literature).
Fig. 6 Possible PALM microscopic set up. The laser module might contain several laser lines that are combined via mirrors (M) and appropriate dichroic mirrors (DC), pass through an AOTF for wavelength selection and power attenuation and are focused onto a fiber coupler (FC) of high performance fibers (fibers). It can be advantageous to couple the InVis laser (405 nm) into a separate optimized fiber and combine its laser light via a suitable dichroic within the illumination module with the VIS lines (488, 561 and 641 nm). The fibers deliver the light into the illumination module. The beams are collimated by Collimator lenses (Colli) and pass through a laser clean up filter (LCF) that removes all unwanted wavelengths. By field adjustment optics (FAO) the beam diameter and hence the field of view and power density can be adjusted. Via an imaging lens system (ILS) consisting of two lenses (L) the beams are focused via a polychromatic mirror (PCM) into the back aperture of the objective lens (OL) of the microscope. A moveable mirror (MM) is introduced before the polychromatic mirror to be able to move the focused beam to different position in the back aperture. This allows to set the illumination regime to EPI, HILO or TIRF. The PCF reflects the excitation light (light orange) through the objective lens (OL) into the sample. Fluorophores are excited preferentially in the focal plane (FP) and will emit photons. The fluorescence light is collected by the same objective lens and passes through the PCM that hereby separates the excitation from the emission light. The emission light is spectrally filtered through an emission filter (EF), which is represented in the majority by either band pass (BP) or long pass (LP) filters. PCM and EF can be conveniently housed in a filter cube. The emitted light passes further through the tube lens that focuses the light onto the detector, mostly a sensitive EMCCD camera. Additional lenses (L) are arranged to form a telescope (tele) which ensures a pixel size of approximately 100 nm. For multi-line
excitation single laser
lines are
superimposed by a
suitable beam
combiner with
respective dichroic
mirrors and passed
through an acousto
optical tunable filter
(AOTF) for
wavelength selection
and power
attenuation (Fig. 6).
The light can then be
coupled into a high
power single-mode
fiber, that spatially
Fig. 7 Illumination schemes in PALM / STORM imaging. The beam is focused via a moveable mirror in the center of the back focal plane of the filters the light. The
objective in order to obtain epi-‐fluorescence (EPI) illumination (left). light can also be
Moving the focus of the beam far enough towards the high numerical aperture region of the objective will result in high inclined and directly coupled;
laminated optical sheet (HILO) illumination (middle). By further however, due to laser
moving to the edge of the back focal plane total internal reflection fluorescence (TIRF) is achieved, when the critical angle is reached safety issues
(right) and the light is totally reflected by the cover glass. This often commercially
also results due to reflection to a halo effect (ring). The illumination available systems
pattern is shown in light orange and fluorescence emission distribution by dark orange shading in the front view back focal plane. The back will be offered
reflected beam in TIRF (left spot in the front view) can potentially be exclusively with fiber
suppressed by a block at the corresponding position in the back focal plane. coupling. A laser
clean up filter after the fiber can be used to specifically select the desired wavelength from the
laser. Alignment to the optical axis can be achieved with suitable steering mirrors and irises in
the beam path, which enable control of the position and angle of the beams. The beams are
collimated and focused on the back focal plane of a high numerical oil-immersion objective
via suitable optics, e. g. a pair of lenses, to ensure homogeneous excitation of a relative large
field of view (FOV) in the range of a 50 x 50 µm² area. Optionally appropriate optics can be
placed before the objective lens to change the diameter of the collimated beam and hence the
illuminated area. This allows to adjust the density of the laser power. A movable mirror in the
excitation path before the objective lens is needed to change between EPI, HILO and TIRF
illumination (Fig. 7).
The fluorescence light emitted from the fluorophore is collected through the same objective
and separated from the excitation light via the polychromatic mirror. It is than spectrally
filtered by passing through an appropriate emission filter, mostly a band pass or a long pass
and projected on an EMCCD camera of high quantum efficiency that can be directly mounted
on a suitable port (mostly side or base) of the inverted microscope stand. A tube lens is
introduced enabling the direct attachment of the camera to the port. To adjust the pixel size to
generally ~ 100 nm, additional lenses and a telescope can be inserted in the detection path.
2.2
Fluorescence Labeling
The most efficient and specific way for in vivo and in vitro fluorescent labeling of cells are
genetically encoded reporter constructs that are derived from fluorescent proteins (FPs)
(70,71,72). The protein of interest is thereby fused to a suitable FP dependent on the
experimental requirements. For PALM photo-switchable FPs (PS-FP) must be employed. All,
the cell line used, the transfection procedure to introduce the expression plasmid into the cell,
and the site of the fusion within the protein of interest can have a profound influence on the
success of super-resolution applications and have therefore to be optimized. Once the
expression plasmid was taken up by the cell the fusion will be endogenously expressed.
Usually, cells are transfected the day or two days before the experiment. Expression levels
can easily be followed by wide-field fluorescence microscopy after activation with 405 nm, if
required. In transient transfections labeling efficiencies in the range of 10 – 70% of cells can
be achieved dependent on the construct. Moreover, if stable cell lines can be selected levels of
expressing cells can approach 100%. Once a sufficient expression level has been reached the
cells can be used for super-resolution imaging. At physiological expression levels structure
staining will be highly specific; however, fusions have to compete with native expressed nonlabeled proteins decreasing labeling densities. In addition, overproduction of the fusion can
lead to its ectopical expression and its unspecific attachment to unrelated structures or its
accumulation in the cytoplasm leading to increased background levels. On the other hand,
knock-out / knock-in procedures combined with native promoters yield expression levels
comparable to physiological concentrations that can even be controlled in their timely
expression. Another possibility for controlled application of PS-FPs is microinjection of the
protein. Large amounts of protein can be obtained by its purification from suitable overexpression systems. However, one should keep in mind that microinjection can be stressful to
certain cell types. As medium and serum components contain fluorescent components like
phenol red it is advisable to use phenol red and serum free medium or buffer systems for
PALM imaging. In can also be advantageous to freshly reseed the cells into a new dish or
onto a new cover glass prior to imaging to reduce any background arising from glass adsorbed
material.
In contrast to FPs, the use of synthetic fluorophores like organic dyes and quantum dots (Qdots) require a chemical labeling procedure. Specificity of labeling proteins with synthetic
fluorophores is provided by either self-labeling tags introduced genetically into the protein or
coupling the organic dye or Q-dot to antibodies specific for an epitope of the protein [Jones,
2011 #768]. Alternatively, peptides and oligonucleotides that specifically bind to a protein or
nucleic acid can be employed by introducing a label to them. E.g the peptides Lifeact and
Phalloidin strongly bind to Actin and aptamers (short oligonucleotides) will be specific for
any protein they have been selected for (73, lit. for apatamers). Whereas peptides and
oligonucleotides are usually labeled in a 1:1 stoichiometry, commercially available antibodies
often possess 2-6 conjugated fluorophores, rendering appropriate photo-switching strategies
more of a challenge (lit.). To augment signal strength, secondary labeling strategies or socalled sandwich techniques have come in use. In this case it is not the primary antibody
specific for the antigen that is labeled, but a secondary one that is specific for the primary one.
In another realization, the primary antibody is modified with streptavidin and then detected by
Biotin coupled to an appropriate dye (lit.). Specificity in antibody staining can be increased by
employing the Fab fragment rather than the whole antibody as this fragment contains the
antigen binding site without the Fc part that can unspecifically bind to some cellular structures
(lit.). Another advantage of Fab fragments is their smaller size compared to the full length
antibody. If the dye or dye conjugate is not able to cross the cellular membrane, especially
true for antibodies, working under live conditions will be restricted to the membrane and
staining proteins inside the cell necessitates fixation and permeabilization steps before
incubation with the antibody.
Whereas therefore immunofluorescence staining with antibodies is mostly restricted to in
vitro experiments many chemical labels are cell membrane permeable, like fluorophores
binding to self-labeling tags, and can as such be used in live cell applications (lit. zitat).
Several tag systems are commercially available including tetracysteine (TC)-tag reacting with
biarsenic fluorophores like FlAsH (fluorescent arsenical helix binder) (74); Snap-tag, a
derivative of O6-alkylguanine-DNA alkyltransferase (AGT) that allows labeling of proteins
with O6-benzylguanine derivatives (75,76); Halo-tag, a derivative of the enzyme halogenase
that can interact with fluorescein and tetramethylrhodamine (TMR) derivatives (77); and
trimethoprim (TMP)-Tag, a derivative of E. coli dihydrofolate reductase (eDHFR) that forms
a high-affinity non-covalent interaction with TMP (78,79,80). The cells are usually labeled ½
- 1 hr prior to imaging with the synthetic fluorophore at concentrations in the µM range. The
optimal substrate concentration has to be determined by trial and error and can depend
strongly on the cell type used (38). Especially when concentrations are chosen too high,
unspecific binding of the dye to the glass surface, cell membrane or other cellular structures
my result raising the background to prohibitive levels. Also there might be a lot of dye freely
floating within or diffusing out of the cell rendering the detection of single molecules a
difficult if not impossible task. Especially for live cell staining using the endogenous reducing
agent gluthatione dyes outside the cell without any reducing agent will stay in their on-state,
whereas the specific bound ones within the cell will be predominantly in their off-state upon
irradiation. To avoid unspecific adsorption, the glass surface might be saturated with glycine
(38). Even more efficient will be the trypsination of the cells and their transfer to a new
chamber well after labeling. If a dynamic process can be followed with sufficient time
resolution is controlled by the labeling density of the structure under investigation and the
photo-switching rate of the irradiation. The temporal resolution decreases with higher labeling
densities and lower irradiation intensities (36). Whereas the former is hard to control as
resolving a structure unambiguously requires full labeling of the structure, irradiation
intensities cannot be arbitrarily increased due to the evoked photo-damage and photo-toxicity
with a living cell. The best compromise has to be found experimentally.
2.3
Data Acquisition
After mounting the sample on the microscope it has to be focused on. This can either be done
with the eyepieces in wide-field illumination using transmitted light, if the sample is of high
enough contrast. Alternatively, if the sample is fluorescent from the beginning, e.g. photochromic FPs and organic dyes, a thermal light source with an appropriate filter can be used. In
normal cases there is a certain amount of conversion of non-fluorescent samples that could be
used for focusing. If the sample is non-fluorescent at all it can be advantageous to express a
fluorescent indicator to facilitate finding the sample, otherwise it can help to activate some
protein in a certain area for localizing the focus plane and then move to another area for
doing the PALM experiment.
Once the sample is found the zoom most adequate for the experiment is chosen. The higher
the zoom the faster the read out time of the camera can be set. For FPs the imaging laser is
constantly increased to usually intensities in the range of 0.5-5 kW/cm2 until single blinking
emitters are observable. The activation laser is set to a low power (≤ 0.01 kW/cm2) and slowly
increased until a satisfactory blinking density is observed, that is relative dense blinking but
clear visible single emitters. During the experiment, the activation and imaging lasers can be
used to adjust the blinking rate as more and more fluorescent proteins will be bleached out
with progressing time. The camera frame rate is matched to the average photo-bleaching /
photo-conversion rates, which for fluorescent proteins is in the order of 10 – 25 Hz.
Samples stained with an organic dye are first illuminated with low intensities of the imaging
laser to select the appropriate area and a wide-field image is recorded for later comparison.
Then most of the fluorophores will be switched to their off-sate by increasing the irradiation
intensity to typically 5-50 kW/cm2. It might be necessary in this step to switch of the gain of
the EM-CCD camera to avoid damage to the chip as organic dye as are very bright and the
process can require a couple of seconds or longer. Once an appropriate blinking density is
obtained, the laser power is reduced to intensities in the range of 1 – 5 kW/cm2. Tuning the
intensity is used to match the lifetime of the molecule to the camera frame rate, which can be
in the order of 10 – 1000 Hz. Increasing the laser power will decrease the lifetime.
Furthermore, the lifetime can also be adjusted by the thiol concentration (1 – 200 mM) and
the pH value of the solution (approximately pH 6.0 – 9.0). On the other hand, the lifetime of
the off-state can be shortened by irradiation with 405 nm light and / or the incorporation of
oxidizing agents (83). A prolongation of the lifetime of the off-state can be achieved by
oxygen removal, e. g. by an enzymatic oxygen scavenging system or purging with nitrogen
(51, 52). It might be necessary to adjust the power of both, the imaging and the activation
lasers in the course of the experiment to keep blinking density on the same level.
Since simultaneous irradiation with 405 nm and high powers of the imaging laser can induce
photo-bleaching alternative excitation with both lasers can be beneficial. E.g., the activation
laser is on only on during the frame transfer of the camera, where no images are recorded, to
(re)activate a subset of fluorophores being in their off-sate, whereas the imaging or readout
laser is on during the camera’s integration time. The number of required frames will depend
in the end on the structure under investigation and how efficiently it was labeled. In the
majority of cases, acquisition times between 1 – 30 min., are sufficient to obtain a fair
representation of the structure in the super-resolution image.
2.4
Data Analysis
After a successful acquisition the data set consists of numerous frames, each containing
numerous diffraction limited fluorescent spots. The PALM / dSTORM image is constructed
in basically three steps: first a peak in intensity (spot) in the image is identified (peak finding);
second a small area around each peak is cut out from the image (peak selection) and the
intensity distribution within the pixels is fitted with a model of the PSF to localize the
fluorophore emission (localization); and third all localizations are combined into one single
super-resolution image (reconstruction).
For peak finding (1st step) the image is generally smoothed beforehand and de-trended to
remove long and short distant intensity gradients. Smoothing can be accomplished by e.g.
using a moving average mask with a diameter of approximately one full width half maximum
(FWHM) of the point spread function (PSF) (39). For de-trending Laplace and Fourier
transformation can be applied to the image (literature). In such pre-treated images the
intensity in a pixel is proportional to the number of photons detected in the PSF. The pixels
with the highest intensities are then extracted within a certain size, typically in the order of 2 x
FWHM of the PSF (2nd step). This can be done with a local maximum filter algorithm (63) or
by setting a filter that compares the peak intensity with the standard deviation of the image
(literature). If the peak intensity is by a certain factor (filter) bigger than the standard
deviation, than the peak is regarded as bona fide and the peak with the next highest intensity
is investigated. Until the criterion is fulfilled the next brightest peak is taken; otherwise the
frame is not analyzed further and the next frame is subjected to the same algorithm. The peak
area is than cut out and fitted to a model PSF function, most commonly a two-dimensional
Gaussian function (literature). Either a fixed covariance matrix plus a constant background
can be used (mask fit) or the intensity distribution of the peak center is used with the
background estimated from the periphery (Gaus fit). Thus, four parameters will be fitted:
amplitude representing the numbers of photons from the emitter, center in X, center in Y and
background. The centers in X and Y with their standard deviation will define a high-precision
estimate of the emitter’s position (localization precision) (31, 69). If two and more
fluorophores emit at the same time within the diffraction limited area (~ 250 – 300 nm) a
photon threshold and the PSF’s geometry can be applied to minimize mis-localizations, which
cannot completely be avoided, especially when inappropriate photo-switching rates had been
used (55). If one can sacrifice localization precisions, special multi-emitter algorithms can be
employed (literature). For reversible-switching molecules, like photo-chromic FP’s and
organic dyes, each fluorophore will emit many times. Hence if one discards multi-emitter
events one still has saved the localization if the same emitter was localized individually at
least once. On the other hand, for irreversible photo-switchers like photo-activatable and
photo-convertable proteins the localization information is irretrievable lost or a false
localization is made in case of two and more spot events.
All localizations with their errors are finally plotted in a single super-resolution image
(PALM-image) that therefore represents a vector map of localization precisions. For
comparison, the background corrected PSFs can also be plotted in an image termed sum widefield image (SWF-image) to estimate the resolution gain one has obtained. The PALM image
can be further processed. First, drift can be corrected using fiducial-markers, if available, or
employing self-aligning algorithms based on models (literature). Second, in the case of
reversible photo-switchers, single emitter events can be defined as belonging to the same
molecule, where one generally filters according to the on-times, off-times and distance
between the localizations of the different plotted LPs, a process called grouping. Third, events
can be filtered by any parameter (upper and lower limits) used in the determination of the
localization precision, which are PSF width, x-position, y-position, number of photons and
background.
2.5
General workflow
Glass bottom dishes / chambers / plates or coverslips on which the cells are to be grown
should be as clean as possible. Unclean coverslips can produce elevated background as debris
deposited on them is often excited with the activation and excitation lasers. Also some glass
sorts can be doted with fluorescent material. These contaminants tend to be more efficiently
excited with redder shifted laser light. There are different protocols for cleaning coverslips to
remove debris and other contaminants, which involve more or less aggressive reagents. More
aggressive protocols cannot be used for glass bottom cell culture dishes, as they attack the
plastic or the glue material. In this case milder methods like alcohol or detergent washes must
be used. Examples of cleaning methods are given in Box 3.1-1.
Once coverslips have been cleaned they can be used to seed cells on them. Either a glass
bottom dish is used, or the coverslip is placed into a plastic cell culture dish. To increase
adhesion of cells, coverslips can be pre-coated by fibronectin or poly-L-lysine (see protocols
of Box 3.1-2). Pre-coating is also recommended, when gold particles are intended to be
deposited that will serve as fiducial markers (see protocols of Box 3.1-3). Optionally, the
surface of the coverslips can be blocked against unspecific binding by bovine serum albumin
(BSA), normal goat serum (NGS) or glycine (Gly) prior to cell seeding (see protocols of Box
3.1.4).
Cells can be cultivated in an appropriate medium like DMEM F-12 or RPMI. Cells can be live
stained by transfection of plasmid DNA carrying the protein of interest fused to a fluorescent
protein or expressing a tag to which a molecule attached to an organic dye and which can
penetrate the cell membrane can bind, e.g. a Snap-tag (see protocols of Box 3.2-2). There are
many protocols available for DNA transfection and most kits will provide detailed step-bystep guidelines (see protocols of Box 3.2-1). Cultivated cells that have been live stained can
be imaged after extensive washing in the live state, or need only to be fixed prior to analysis.
On the other hand, fixed cells can be stained by antibodies that target specific epitopes of the
protein of interest and that can be conjugated to organic dyes directly or are detected by a
secondary antibody-dye complex specific for the primary one. To this end they need to be
permeabilized in addition. Protocols for fixation and permeabilization are available in Box
3.2-3). A commonly used fixative is formaldehyde (FA). For permeabilization the non-ionic
detergent Triton-X-100 is often employed. Alternatively cells can be fixed and permeabilized
with methanol and acetone. Care should be taken for structural preservation and the best
protocol must be determined empirically. After the permeabilization step, the cells can be
stained with an appropriate antibody. A protocol with a primary antibody and a dye
conjugated secondary antibody is given in Box 3.2-4. After fixation or immunostaining cells
can be preserved by drying after a wash in water or mild detergent to remove salt and buffer
ingredients for prolonged storage at room temperature (see protocols Box 3.2-4).
For PALM imaging, cells are overlaid with an appropriate PALM imaging buffer (see recipes
of Boxes 4.1-1 and 4.1-2). Alternatively, coverslips can be embedded in a suitable mounting
medium (see Boxes 4.1-3 and 4.1-4).
3
Protocols
3.1
Coverslip treatment
Box 3.1-1: Protocols for coverslip cleaning
Method 1: Acid cleaning
1.
2.
3.
4.
Wash coverslips for 5’ in 1% HF
Wash coverslips for 5’ in MilliQ water
Wash coverslips for 5’ in 100% EtOH
Flame briefly and use directly
Remark: 1% hydrofluoric acid (HF) is well suited to clean conventional glass coverslips.
However, it will attack high-index coverslips rendering them opaque and hence unsuitable
for imaging. Wear protective gloves and eyewear when handling HF.
Method 2: Etch cleaning
1. Heat coverslips sitting in a corrosive-free staining rack with metal tongs in a solution
of MilliQ H2O: NH4OH 15.4 M : H2O2 50% = 5 : 1 :1 to 80-85 oC in a beaker and
stir for more than 2-3 hr, preferentially O/N.
2. Rinse with a copious amount of MilliQ H2O in a fresh beaker.
3. Place in Acetone (spectroscopic grade) in a fresh beaker. Sonicate in a water bath for
10’.
4. Rinse with a copious amount of MilliQ H2O.
5. Rinse both sides of the coverslip with Methanol (spectroscopic grade).
6. Air dry fast (best done by blowing with filtered air).
7. Flame 3 x for not more than 1’’.
8. Store in a closed container until use.
Remark: Coverslips can be re-cleaned and re-used several times. Store the used cover-slips in
a 2% solution of RBS detergent and rinse with copious amounts of water before repeating
steps 1. – 8. Wear gloves and take care to avoid exposing the skin to the corrosive solvents
used in the cleaning solution.
Method 3: Alcohol cleaning.
1.
2.
3.
4.
5.
6.
7.
Wash coverslip 2 x 2 ‘ with MilliQ H2O
Wash coverslip 1x 2’ with Ethanol (spectroscopic grade)
Wash coverslip 2x 2’ with MilliQ H2O
Wash coverslip 1x 2’ with Methanol (spectroscopic grade)
Air dry in a fume hood
Optionally expose to UV light in a fume hood.
Store in a closed container until use
Remark: This method can be used for cell culture dishes with cover glasses or LabTekTM
chambers. Work in this case under sterile conditions to avoid contamination when culturing
cells.
Method 4: Base cleaning
1. Incubate coverslip, glass bottom dish or LabtekTM chamber for 30`in 1 M KOH.
Sonicate during that time period in a water bath sonicator.
2. Rinse with a copious amount of MilliQ H2O.
3. Sterilize by at least 1 hr exposure to UV light in a hood.
Method 5: Detergent cleaning
1.
2.
3.
4.
5.
6.
Incubate coverslips O/N in 2% RBS-35 liquid detergent.
Rinse with copious amounts of water
Sonicate coverslips for 5’ in acetone (spectroscopic grade)
Rinse with copious amounts of water.
Wash briefly for 2’ in methanol (spectroscopic grade)
Flame briefly and store dry in a closed container.
Box 3.1-2: Protocols for coating coverslips
Method 1: Fibronectin pre-coating
1. Place coverslip (#1.5) in a 35 x 10 mm plastic dish or glass bottom dish, which
should be placed plastic dish to avoid scratching of the bottom glass surface, in a
larger sterile environment, e. g. a hood. Place glass bottom dishes in.
2. Overlay coverslip or glass bottom with a solution of 10 µg/ml Fibronectin in PBS);
200 µl should be sufficient for a 25 mm coverslip.
3. Cover and store O/N at 4 oC.
4. Remove fibronectin solution and wash 3x 2’ in PBS.
5. Aspirate PBS and add cells in medium.
Method 2: Poly-L-Lysine pre-coating
1. Place coverslip (#1.5) in a 35 x 10 mm plastic dish or glass bottom dish, which
should be placed plastic dish to avoid scratching of the bottom glass surface, in a
sterile environment, e. g. a hood.
2. Overlay coverslip or glass bottom with a solution of 0.1% (v/v) Poly-L-lysine in PBS;
200 µl should suffice.
3. Cover and incubate for 30’
4. Remove Poly-L-Lysine solution and wash 3x 2’ in PBS.
5. Aspirate PBS and add cells in medium.
Box 3.1-3: Protocols for depositing fiducial markers on coverslips
Method 1: Depositing gold particles before cell growth
1. Clean coverslips using the Etch cleaning method (Method 2 of Box 3.1-1 Step 1)
2. Put coverslips onto cover-slip support blocks into individual small Petri dishes (so
that applied solutions do not dry quickly). Write numbers on the edge of the top
surface to distinguish the top surface in the future.
3. Coat with Poly-L-Lysine (Method 2 of Box 3.1-2 steps 2-3).
4. Rinse with H2O and blow-dry.
5. At the same time dilute 80 nm Au bead stock to 1-2% and start sonicating (for the
same 30 minutes). Take Au stock out of the cold room, shake well, then vortex for
30 seconds. Take 5-10 microliters of the solution out of the stock vile, mix it with 500
microliters of H2O, and sonicate the solution for 30 minutes.
6. Put 1-2% Au beads (diluted in H2O) for 5-10 minutes on coverslip. It should cover
most of the top surface. After the incubation period rinse with H2O and blow-dry.
7. For permanent storage deposit 30-50 nm of SiO2 in a Denton Explorer system. As
cells attach poorly onto SiO2 pre-coating with Fibronectin or Poly-L-Lysine is
mandatory (Box 3.1-2 steps 1-5).
Remark: Coverslips should be deposited before the pre-coating step with Fibronectin or
Poly-L-Lysine. Ready-to-use fiducial coverslips of different densities and different spectral
range are commercially available (see Box 5.1-1). Without the SiO2 layer cells tend to
incorporate the gold particles over time when getting into contact with them.
Method 2: Depositing gold particles after cell growth
1. Wash fixed and stained cells 3x 5’ in PBS
2. Wash 3 x 5’ in H20.
3. Dilute Au bead stock to 1-2% and start sonicating (for the same 30 minutes of the
wah period). To this end take Au paticles out of the cold room, shake well, then
vortex for 30 seconds. Take 5-10 microliters of the solution out of the stock vial, mix
it with 500 microliters of H2O, and sonicate the solution for 30 minutes.
4. Put 1-2% (diluted in H2O) Au beads for 5-10 minutes on coverslip. It should cover
most of the top surface. After the incubation period rinse with H2O and blow-dry.
Remark: Do not dilute Au particles in salt containing solutions as they pretend to precipitate
out or aggregate in high ionic strength buffer. The inclusion of capping agents can be
beneficial to prevent aggregation. They include anionic (citrate and tannic acid), neutral
(2,2,2-[mercaptoethoxy(ethoxy)]ethanol and polyvinylpyrrolidone), and cationic
mercaptopentyl(trimethylammonium) compounds.
Method 3: Depositing Tetraspec beads after cell growth
1. Dilute TetraSpecksTM 1000 x in a suitable volume of imaging buffer.
2. Deposit on cells for at last 1 hr.
3. Wash 3 x in imaging buffer or image directly
Remark: With the high laser powers used for PALM, Tetraspecs might often be too bright.
Look out for Tetraspecs that are deposited at the edge of the field of view, where laser
powers are lower.
Box 2-1.4: Protocol for blocking coverslips
Protocol 1: BSA blocking
1. Prepare 1% BSA in phenol free medium (e.g. DMEM.HG) and heat inactivate at 56
o
C for 30’. Filter through 0.22 µm filter and store at 4 oC for up to 1 month.
2. Cover coverslip with 1% BSA-DMEM.HG for 1 – 2 h at RT. For a 25 mm Ø
coverslip or a 35 mm dish approximately 200 µl will be required.
3. Suck of solution and add 2 ml growth medium. Equilibrate at 37oC and 5% CO2 prior
to plating cells.
Protocol 2: NGS blocking
1. Prepare 10% normal goat serum (NGS) in phenol free medium (e.g. DMEM.HG)
and heat inactivate at 56 oC for 30’. Filter through 0.22 µm filter and store at 4 oC for
up to 1 month.
2. Cover coverslip with 10% NGS-DMEM.HG for 1 – 2 h at RT. For a 25 mm Ø
coverslip or a 35 mm dish approximately 200 µl will be required.
3. Suck of solution and add 2 ml growth medium. Equilibrate at 37oC and 5% CO2 prior
to plating cells.
Protocol 3: Glycine blocking
1. Prepare 2 M glycine solution in PBS.
2. Cover coverslips with 2M gycine/PBS solution for 1-2 h at RT. For a 25 mm Ø
coverslip or a 35 mm dish approximately 200 µl will be required.
3. Suck of solution and add 2 ml growth medium. Equilibrate at 37oC and 5% CO2 prior
to plating cells.
3.2
Cell preparation
Box 3.2-1: Protocol for transfections
Method 1: Lipofectamine (Invitrogen protocol)
1. One day before transfection, plate 1-3 x 105 cells in 35 mm cell culture dish in 2 ml of
growth medium (with the usual amount of serum) without antibiotics so that cells will
be 50-80% confluent at the time of transfection.
2. For each transfection sample, prepare complexes as follows:
a. Dilute 1-2 µg DNA in 100 µl of Opti-MEM® I Reduced Serum Medium (or other
medium) without serum. Mix gently.
b. Mix Lipofectamine™ gently before use, then dilute 2-25 µl in 100 µl of OptiMEM® I Medium (or other medium) without serum. Mix gently.
c. Combine the diluted DNA with diluted Lipofectamine™ (total volume = 200 µl).
Mix gently and incubate for 15-45 minutes at room temperature (solution may appear
cloudy). Note: Complexes are stable for 6 hours at room temperature.
d. For each transfection, add 0.6 ml of Opti-MEM® I Medium to the tube containing
the complexes (total volume = 800 µl). Mix gently.
3. Remove the growth medium from cells and replace with 0.8 ml of growth medium
without serum. Add the 0.8 ml of diluted complexes (from Step 2d) to each cell
culture dish. Mix gently by rocking the plate.
4. Incubate cells at 37°C in a CO2 incubator for 2-24 hours. We recommend starting
with 5 hours.
5. Add 0.8 ml of growth medium containing 2X the normal concentration of serum
without removing the transfection mixture. Note: If toxicity is observed after
transfection, replace medium with fresh, complete medium (with normal amount of
serum).
6. For transient transfection: Test for transgene activity 24-72 hours posttransfection as
appropriate for your cell type and expression vector. For stable transfection: Passage
cells at a 1:10 dilution into selective medium 72 hours post-transfection.
Method 2: Effectene (Qiagen protocol)
1. The day before transfection, seed 0.9–4 x 105 cells (depending on the cell type) per 35
mm cell culture dish with 1.6 ml appropriate growth medium containing serum and
antibiotics.
2. Incubate the cells under their normal growth conditions (generally 37°C and 5%
CO2). The cells should be 40–80% confluent on the day of transfection.
3. On the day of transfection, dilute 0.4 µg DNA dissolved in TE buffer pH 7 to pH 8
(minimum DNA concentration of 0.1 µg/µl) with Buffer EC, to a total volume of 100
µl. Add 4.2 µl Enhancer and mix by vortexing for 1 s. IMPORTANT: Always keep
the ratio of DNA to Enhancer (1:8) constant.
4. Incubate at room temperature (15–25°C) for 2–5 min, and then centrifuge briefly to
remove drops from the top of the tube.
5. Add 10 µl Effectene Transfection Reagent to the DNA–Enhancer mixture. Mix by
pipetting up and down 5 times, or by vortexing for 10 s. Note: It is not necessary to
keep Effectene Transfection Reagent on ice at all times: 10–15 min at room
temperature will not alter its stability.
6. Incubate the samples for 5–10 min at room temperature to allow transfection complex
formation.
7. While complex formation takes place, gently aspirate the growth medium from the
plate, and wash cells once with 3 ml PBS. Add 1.6 ml fresh growth medium (can
contain serum and antibiotics) to the cells.
8. Add 600 µl growth medium (can contain serum and antibiotics) to the tube containing
the transfection complexes. Mix by pipetting up and down twice, and immediately
add the transfection complexes drop-wise onto the cells in the dish. Gently swirl the
dish to ensure uniform distribution of the transfection complexes.
9. Incubate the cells with the transfection complexes under their normal growth
conditions for an appropriate time for expression of the transfected gene. The
incubation time is determined by the assay and gene used. Optional: In many cases,
removal of transfection complexes is not necessary. However, if cytotoxicity is
observed, remove the Effectene–DNA complexes after 6–18 h, wash the cells once
with PBS, and add 3 ml fresh growth medium.
10. For transient transfections: Assay cells for expression of the transfected gene. Cells
transfected with β-gal or cat reporter constructs are typically incubated for 24–48 h
post-transfection to obtain maximal levels of gene expression.
For stable transfections: Passage cells 1:5 to 1:10 into the appropriate selective
medium 24–48 h after transfection. Maintain cells in selective medium until colonies
appear.
Method 3 FuGene (Roche protocol)
1. One day before transfection, plate 1-3 x 105 cells in 35 mm cell culture dish in 1.6 ml
of growth medium (with the usual amount of serum) without antibiotics so that cells
will be 50-80% confluent at the time of transfection.
2. Dilute plasmid DNA by adding 16 µg DNA to 800 µl OptiMEMTM I Reduced Serum
Medium or another serum-free medium to achieve a concentration of 20 µg/ml in an
1.5 ml reaction tube. Vortex briefly to mix.
3. Dispense 100 µl of the diluted DNA (2 µg plasmid/100 µl) in a fresh 1.5 ml rection
tube. For a 3:1 FuGENE® HD Transfection Reagent:DNA ratio, add 300 µl
FuGENE® HD Transfection Reagent and mix immediately and gently by pipettig up
and down 15x. Add FuGENE® HD Transfection Reagent directly to medium. Do not
allow undiluted FuGENE® HD Transfection Reagent to contact the sides of the tube.
4. Incubate the FuGENE® HD Transfection Reagent/DNA mixture for 5–15 minutes at
room temperature. Optional: Add mixture to cells without an incubation period.
5. Take cell culture dish from the incubator and add 20 – 100 µl of the FuGENE® HD
Transfection Reagent/DNA mixture to the cell culture dish. Mix gently by shaking.
Return cells to the incubator for 24-72 hours depending upon the expressed gene.
6. Measure transfection efficiency using an assay appropriate for the reporter gene. For
transient transfection, cells are typically assayed 24–48 hours after transfection. For
stable transfection cultivate cells in selective medium.
Method 4: Nucleofectin (Lonza protocol)
1. Cells should be passaged after reaching 70–90 % confluency. Do not use cells after
passage 14 for Nucleofector™. 2 days before transfection, passage the cells and plate
1-3 x 105 cells in 35 mm cell culture dish in 2 ml of growth medium (with the usual
amount of serum) without antibiotics so that cells will be optimally 80-90% confluent
at the time of transfection.
2. Remove media from the cultured cells and wash cells once with HBSS; use at least
same volume of HBSS as culture media.
3. For harvesting, incubate the cells ~5 minutes at 37°C with recommended volume of
trypsinization reagent. Alternatively, if cells hardly detach use Trypsin 0.5%-EDTA
0.2%.
4. Prepare 35 mm dishes by filling 1 ml of supplemented culture media and preincubate/equilibrate plates in a humidified 37°C/5% CO2 incubator.
5. Harvest the cells by trypsinization.
6. Count an aliquot of the trypsinized cells and determine cell density.
7. Centrifuge the required number of cells (0.5–1 x 106 cells per sample) at 90 x g for 10
minutes at room temperature.
8. Resuspend the cell pellet carefully in 100 µl room temperature Nucleofector™
solution per sample.
9. Combine 100 µl of cell suspension with 1–5 µg DNA, and (recommended for initial
optimization) 2 µg pmaxGFP™ vector.
10. Transfer cell/DNA suspension into certified cuvette; sample must cover the bottom of
the cuvette without air bubbles. Close the cuvette with the cap.
11. Select the appropriate Nucleofector™ Program. Please try all Nucleofector™
Programs (A-024, T-016, U-012, U-023, V-013 for NucleofactorTM II device; A-24,
T-16, U-12, U-23 and V-13 for Nucleofector™ I Device) initially to determine the
most appropriate one for primary fibroblats for all subsequent experiments.. For other
cell types proceed as describes with all recommended programs.
12. Insert the cuvette with cell/DNA suspension into the Nucleofector™ Cuvette Holder
and apply the selected program.
13. Take the cuvette out of the holder once the program is finished.
14. Add ~500 µl of the pre-equilibrated culture media to the cuvette and gently transfer
the sample to the dish (final volume 1.6 ml).
15. Incubate the cells in a humidified 37°C/5% CO2 incubator until analysis. Gene
expression is often detectable after only 4–8 hours but ideally, cells should be left
undisturbed for al least 24 hours.
Remark. This protocol is optimized for primary fibtoblasts. Please refer to the manual for
other cell types that might require different handling and programs.
Box 3.2-2 Protocol for live cell labeling with fluorescent protein fusions or chemical protein
tags
Method 1: Labeling with Fluorescent Protein (FP) tags
1. Transfect approximately 2 x 106 cells with the respective FP tag-vector coding for the
protein of interest according to Box 3.2-1.
2. Let cells grow at 37°C and 5% CO2 in medium 24 – 72 h and check cell conditions
prior to imaging and protein expression.
3. Optional: For background reduction labeled cells can be detached from the glass
surface by trypsination (0.25%) at 37°C and transfer into cleaned new glass bottom
dishes or chambers. Allow cells to reattach 1 -3 h prior to imaging.
4. Image cells in phenol red free medium or PBS. Alternatively fix cells according to
Box 3.2-4. A permeabilization step is not needed.
Method 2: Labeling with Snap-tags
1. Transfect approximately 2 x 106 cells with the respective tag vector coding the
protein of interest according to Box 3.2-1. Incubate cells O/N to allow for the
expression of protein.
2. Wash cells 1 x in PBS or cell medium w/o FCS.
3. Prepare 20 µM stock solutions of the tag substrates (e.g. TMP-ATTO 655, SNAPCell TMR-Star, or SNAP-Cell 505 (benzyl guanine Rhodamine Green))v in DMSO.
Dilute the stock solutions in RPMI 1640 medium supplemented with 10% FCS
without phenol red to final concentrations of 0.1-2 µM. Add at various
concentrations.
4. Incubate for typically 20-30’.
5. Wash 5 x 5’ in phenol red free medium or PBS to remove unreacted substrate.
6. Optional: For background reduction labeled cells can be detached from the glass
surface by trypsination (0.25%) at 37°C and transfer into cleaned new glass bottom
dishes or chambers. Allow cells to reattach 1 -3 h prior to imaging.
7. Let cells grow at 37°C and 5% CO2 in medium over night and check cell conditions
prior to imaging.
8. Image cells in phenol red free medium or PBS. Alternatively fix cells according to
Box 3.2-4. A permeabilization step is not needed.
Box 3.2-3: Protocols for fixation and permeabilization of cells
Method 1: FA fixation and Triton permeabilization
1. Aspirate medium from cultured cells.
2. Wash cells 2 x for 5’at RT in PBS pre-warmed to 37 oC.
3. Fix cells for 1’ at RT in 4.7% formaldehyde (FA) in PBS, freshly made, filtered and
pre-warmed to 37 oC.
4. Wash 2x for 5’ at RT with PBS / 100 mM glycine, pre-warmed to 37 oC with gentle
shaking in the dark.
5. Permeabilize cells by incubation for 15’ with 0.2% Triton-X-100 (sonicated for 1 hr)
in PBS with gentle shaking in the dark.
6. Wash cells 3 x for 5’ at RT with 0.05% Triton-X-100 in PBS with gentle shaking.
Remark. For Cytoskeleton staining do not use PBS containing divalent cations Mg2+ and
Ca2+. It is recommended to use PHEM buffer.
Preparation of 4.7% Formaldehyde solution.
Preparation 1
Add 2 g paraformaldehyde powder to 100 ml of 1 X PBS. Heat to 70°C (do not exceed this
temperature) in a fume hood until the paraformaldehyde goes into solution (note that this
happens quickly as soon as the suspension reaches 70°C). Allow the solution to cool to room
temperature. Adjust to pH 7.4 using 0.1 M NaOH or 0.1 M HCl, if needed. Filter through e.g.
Fisherbrand P8 filters and store at 4°C protected from light.
Preparation 2:
Dilute 10% formaldehyde solution (depolymerized paraformaldehyde, EM grade, methanolfree) in PBS to ield final concentration of 4.7%.
Method 2: FA/GA fixation and Triton permeabilization
1. Aspirate medium from cultured cells.
2. Wash cells 2 x for 5’ at RT in PBS pre-warmed to 37 oC.
3. Fix cells for 15 ‘ at RT in 4.7% formaldehyde (FA) / 0.2 % glutardialdehyde (GA) in
PBS, freshly made, filtered and pre-warmed to 37 oC.
4. Quench GA by incubation with 0,1% (w/v) NaBH4 (sodium borohydride), freshly
made –use only when bubbling – and pre-warmed to 37 oC for 2 x 10’ at RT in the
dark with gentle shaking. Alternatively wash 2x for 5’ at RT with PBS / 100 mM
glycine, pre-warmed to 37 oC with gentle shaking in the dark.
5. Permeabilize cells by incubation for 15’ with 0.2% Triton-X-100 (sonicated for 1 hr)
in PBS with gentle shaking in the dark.
6. Wash cells 3 x for 5’ at RT with 0.05% Triton-X-100 in PBS with gentle shaking.
Remarks: Treatment with the reducing agent NaBH4 will reduce autofluorescence of GA.
Note that NaBH4 is hygroscopic and hardens. For Cytoskeleton staining do not use PBS
containing divalent cations Mg2+ and Ca2+. It is recommended to use PHEM buffer.
Method 3: Methanol fixation and permeabilization
1.
2.
3.
4.
Aspirate medium from cultured cells.
Wash cells 2 x for 5’ at RT in PBS.
Fix cells for 5’ with methanol pre-chilled to -20 oC.
Wash 2x for 5’ at RT with PBS / 1% BSA or PBS / 10% normal goat-serum (NGS),
pre-warmed to 37 oC with gentle shaking in the dark.
Remark: Methanol also will permeabilize cells, so there is no permeabilization step needed
for immunofluorescence staining.
Method 4: Methanol/Acetone fixation and permeabilization
1.
2.
3.
4.
5.
6.
7.
Aspirate medium from cultured cells.
Wash cells 2 x for 5’ at RT in PBS.
Fix cells for 5’ methanol pre-chilled to -20 oC.
Aspirate fixative.
Permeabilize cells for 1’ with acetone pre-chilled to -20 oC.
Aspirate acetone.
Wash 2x for 5’ at RT with PBS.
Box 3.2-4: Protocol for Immunostaining
Method 1: Immunostaining unlabeled cells with antibodies
1. Fix and permeabilize cells as described in Box 3.2-4.
2. Wash 2 x for 5’ at RT.
3. Block cells for 60’ at RT with 1% BSA or 10% normal goat-serum in PBS /0.05%
Triton-X-100 pre-warmed to 37 oC w/o shaking.
4. Dilute primary antibody in PBS / 1% BSA; as a guideline 1:1000 dilutions or higher
might apply. Spin down mix for 10’ at 10 000 g in a 1.5 ml Eppendorf tube. For actin
staining you might use 5 µl Phalloidin-Alexa dye per dish or coverslip. From the 1.2
ml mix take out 1 ml supernatant and add to cells.
5. Incubate at RT for 2 hrs w/o shaking.
6. Wash 2 x 10’ in PBS / 0.05% Triton-X-100 w/ gentle shaking at RT.
7. Dilute secondary antibody with conjugated organic dye in PBS / 1% BSA; as a
guideline 1:5000 dilutions or higher might apply. Spin down mix for 10’ at 10 000 g
in a 1.5 ml Eppendorf tube. From the 1.2 ml mix take out 1 ml supernatant and add to
cells.
8. Incubate at RT for 1 ½ hr w/o shaking.
9. Wash 2 x 10’ in PBS / 0.05% Triton-X-100 w/ gentle shaking at RT.
10. Wash 2 x 10’ in PBS w/ gentle shaking at RT.
11. Store at 4 oC under PBS for a maximum of 2 days or preserve by mild detergent wash
(see Box 3.2-4)
Remark: All incubation and wash steps should be carried out in the dark. Cover the samples
with a dark box or silver foil. For cytoskeleton staining do not use PBS containing divalent
cations. Steps 7.-9. are omitted for Phalloidin staining. A post-fixation protocol can be
inserted after step 9, which includes steps 2-4 of Box 3.2-2 Methods 1 or 3. The post-fixation
can help to stabilize antibody binding to its epitope under the high power light conditions in
PALM imaging
Method 2: Treatment of live labeled cells
1. Fix cells as described in Box 3.2-3. Permeabilization is not needed.
2. Store at 4 oC under PBS for a maximum of 2 days or preserve by water wash (see
Box 3.2-5)
Box 3.2-5: Preservation of fixed and / or permeabilized cells
Method 1: Water wash
1. Fix and permeabilize cells as described in Box 3.2-3.
2. Wash 2 x for 5’ in MilliQ H2O with gentle shaking at RT in the dark.
3. Flip dish over and place on lid so one side is propped up or lay coverslip with cell
side up and let dry O/N at RT in the dark.
4. Store at RT in the dark.
Method 2: Mild detergent wash
1. Wash 1 x 20” in MilliQ H2O.
2. Wash 1 x 20” in 0.1% Tween 20 (made up by adding 18 µl Tween to 18 ml MilliQ
H2O in a 50 ml Falcon tube).
3. Aspirate.
4. Flip dish over and place on lid so one side is propped up or lay coverslip with cell
side up and let dry O/N at RT.
5. Store at RT in the dark.
Method 3: Mounting
1. Fix and permeabilize cells as described in Box 3.2-3.
2. Wash 2 x for 5’ in MilliQ H2O with gentle shaking at RT in the dark. Flip dish over
and place on lid so one side is propped up or lay coverslip with cell side up and let
dry O/N at RT in the dark.
5. For Coverslips: Put a drop of imaging mounting media onto a slide. Place coverslip
upside down onto drop and let spread. Dry O/N in the dark.
For cell culture dish: Put a drop of imaging mounting media onto the cover glass.
Place a coverslip of appropriate diameter onto the drop and let spread. Dry O/N in the
dark.
6. Optionally seal with an appropriate sealant (Sealants 1-3 of Box 4.2-3).
7. Store at RT in the dark.
3.3
Protocol for PALM / dSTORM imaging
Step 1: Fluorescent labeling of cells (2-3 days)
1. Prepare coverslips or glass bottom cell culture dishes according to Boxes 3.1-1 to 3.14 as requested.
2. Plate cell. As appropriate transfect with the appropriate expression vector according
to Box 3.2-1 and 3.2-2 and / or imunostain cells according to Boxes 3.2-3 and 3.2-5.
Step 2: Preparation of Microscope set up (1-2 h)
1. Turn on system and switch on lasers. Set laser powers to low values (1 – 2 mW).
Most systems need to warm up for 1 – 2 h in order to calibrate with the environment.
2. Set up appropriate beam path with appropriate dichroic mirrors and bandpass filters
Measure laser power with the respective dichroics before objective. Determine
homogeneously irradiated area with a fluorophore solution (~10-8 M) and calculate
laser power in kW x cm-3.
Remark. Commercial systems are normally equipped with laser safety mechanism to
avoid damaging the eye. Nevertheless it is advisable to check before looking through
the eyepieces if the filter combinations are compatible with the selected lasers.
3. Use high numerical objectives (oil or water). Determine illumination profile of
quantum dots (Q-dots, QD), which should result in diffraction limited spots. The
emission profile should be independent of the laser irradiation.
Remark: The employment of QD655 is advantageous as it is excited by many
different wavelkenth (405, 488, 532, 561 and 641 nm).
4. Move TIRF mirror to achieve TIRF (Total Internal Reflection Fluorescence) or HILO
(High Inclined and Laminated Optical Sheet) illumination. You can use QD,
TetraSpeckTM beads or gold fiducial markers adsorbed to the coverslip to determine
the correct angle. In TIRF or HILO the signal-to-background ratio increases
substantially. As the irradiation intensity increases the gain of the camera should be
lowered when going from epifluorescence (EPI) illumination to TIRF or HILO to
avoid damage to the camera chip. For TIRF the beam is moved out completely until
the image gets black and than the angle is stepwise moved back until the fluorescence
intensity and signal-to-background ratio show their maximum values. When the
sample is not completely planar, re-focusing might be required when adjusting the
TIRF angle. For HILO the angle is moved from TIRF geometry towards EPI. The
point before a drop of in intensity and signal-to-background defines the start of HILO
illumination coming from EPI.
Step 3: Data Acquisition (1’ – 30’)
1. Screen for suitable cells and focus in transmitted light or Epifluorescence using a
thermal light source using the eye pieces.
2. Switch to laser illumination. Start imaging the cells with low laser power (< 0.1 kW
cm-2) using for example the AOTF (Acousto Optical Tunable Filter) control if
available. You might reduce the area of the camera chip to accelerate data acquisition
and reduce data volume.
3. Set the system to record a time series of 5000 – 20000 frames dependent on labeling
density and complexity of the structure. Before starting image cells with high laser
powers (5 – 30 kW cm-2) as along as it takes to transfer most of the molecules into the
off state and only a sparse subset of the molecules exist in the on state. The PSF of
the latter molecules should not overlap. The integration time should be set that a
molecule is not on in more than 1 to 5 consecutive frames. Readjust the focus so that
the PSF (Point Spread Function) become symmetrical. Set the camera gain for
optimal signal-to-noise ration (SNR).
4. Reduce irradiation intensity (1-5 kW cm-2) for live cell imaging. Start time series with
a frame rate of (10 – 40 Hz corresponding to 100 – 25 ms integration time). For fixed
cells higher irradiation intensities (up to 50 kW cm-2) can be used to increase
photoswitching rates and higher frame rates up to 100 Hz (corresponding to 10 ms
integration time).
5. Optimal fluorophore density should stay between 0.05 – 1 fluorophore / µm2 and the
PSFs should not overlap. Increase activation laser power if fluorophore density is too
low (typically 0.01- 0.1 kW cm-2 of a wavelength beneath the excitation line).
Temark: For most Fluorescent Proteins (FPs) the activation line is 405 nm. For
organic dyes excited with 488 & 561 nm it will also be 405 nm. For Cyan dyes the
488, 514 and 532 nm lines can be used in addition. For organic dyes the activation
laser directly excites intermediate radical ions dye states to activate the fluorophore.
For FPs the 405 nm line induces reversible or irreversible conformational changes
from the “off” to the “on” state of the fluorophore. There are three classes of
switchable FPs: photoactivatable (irreversible photoswitch from “off” to “on” state
via chemical modification), photoconvertable (irreversible photoswitch from one
spectral state to another spectral state via bond cleavage) and photochromic
(reversible photoswitch from “off” to “on” state via cis-trans isomerisation). Ideally,
the activation laser should be used in short pulses in length (few µs – ms) to reduce
autofluorescence, cell damage and photobleaching. E. g. the activation laser can be
switched on during the frame transfer time of the camera. If molecules in the “on”
state decrease over time, the excitation laser has to be increased to activate a constant
number of molecules.
Step 4: Data Analysis (10’-60’)
1. Peak selection. Set amplitude threshold to an appropriate level over background. Start
with a value of 6 and increase, and visually screen for specifically selected spots and
spots selected within background. Increase the threshold to reduce number of
background spots to a minimum without losing to many of specific spots. Adjust the
mask size so that it exceeds slightly the spot size. The periphery of the mask will set
2.
3.
4.
5.
6.
the background photon levels; its size determines what area is selected to be plotted
in the PALM image.
Fitting: Apply a Gaussian fitting to the detected PSFs, which will give you the
localization precision (LP).
Drift Correction: Apply drift correction using fiducial markers or image selfalignment algorithms, if necessary.
Grouping: If needed, group events to a molecule based on on-time, off-time and
distance.
Filtering: If needed, filter data according to photon numbers, localization precision,
PSF width, background etc.
Display: Select pixel size according to localization precision. According to Nyquist,
pixel size should be twice as small as the LP average.
Remark: Most commercial software products provide the various filter functions and
will perform them online and automatically.
4
4.1
Solutions
Stock solutions
Box 4.1-1: Imaging buffer stock solutions
Buffer 1: 10 x PBS
26.7 mM KCl (2 g/l)
15.7 mM KH2PO4 monobasic (2 g/l)
1379.3 mM NaCl (80 g/l)
80.6 mM Na2HPO4•7H2O dibasic (21.6 g/l)
Adjust to pH 7.4 w/ 10 N NaOH
Dilute 1:10 before use.
Remark: 10 x Dulbeccos modified PBS (10 x DPBS) contains in addition 9.01 mM CaCl2 (1
g/l) & 5.93 mM MgCl2•6 H2O (1 g/l)
Buffer 2: 10 x PHEM
600 mM PIPES free base (181.4 g/l) or sodium salt (195.6 g/l)
250 mM HEPES (65 g/l)
100 mM EGTA (38 g/l)
20 mM MgCl2 (9.9 g/l)
Adjust to pH 7 w/ 10 N NaOH
Dilute 1:10 before use.
Buffer 3: 10 x TN
500 mM Tris-HCl, pH 7.5
100 mM NaCl
Dilute 1:10 before use
Remark: 1 M Tris-HCl pH7.5 or 500 mM Tris-HCl pH 8.0 can be used for stronger buffer
strength.
Buffer 4: 10 x HN
500 mM Hepes-NaOH, pH 7.8
100 mM NaCl
Dilute 1:10 before use
Box 4.1-2: Oxygen scavenging stock solutions
System 1: Glucose-Oxidase (GLOX) -Catalase system
10 x stock:
100 mM Tris-HCl pH 7.5
25 mM KCl
1 mM TRIS(3-carboxyethyl)phosphine hydrochloride pH 7 (TCEP)
5 mg/ml Glucose Oxidase
0.4 mg/ml Catalase
Mix well than add
50% (v/v) Glycerol
Centrifuge for 10’ at 10000 x g.
Store supernatant in Aliquots at -20 oC for up to 2 years.
Prior to use dilute 1:10 in a suitable imaging buffer
Remark: TCEP is optional and serves in the GLOX stock to reduce S-S-bridge formation and
hence is intended to increase the shelf life of the solution. The pH value drops drastically
after 20 – 30 min dependent on the buffering effect of the imaging buffer.
System 2: Glucose-Oxidase (GLOX) -Peroxidase system
10 x stock:
250 mM Hepes-KOH pH 8
5 mg/ml Glucose Oxidase (stock 10 mg/ml in 50 mM sodium acetate pH 5.1)
0.25 mg/ml peroxidase [stock 10 mg/ml in 0.1 M potassium phosphate buffer pH 6].
Mix well than add
50% (v/v) Glycerol
Store supernatant in aliquots at -20 oC for up to 2 years.
Prior to use dilute 1:10 in a suitable imaging buffer
System 3: Protocatechuic acid/Protocatechuate-3,4-dioxygenase (PCA/PCD) system
40 x stock:
100 mM Tris-HCl pH 8.0
50 mM KCl
1 mM EDTA
Dissolve PCD powder to at least 1 µM end concentration.
Store at -20 oC.
40 x stock:
100 mM PCA in MilliQ H2O and adjust to pH9 with 10 N NaOH.
Store at 4 oC.
Prior to use mix the two solutions with an appropriate imaging buffer to obtain end
concentrations of 3.5 mM PCA und 50 nM PCD, respectively, i.e a 1:40 dilution of both
stocks.
Box 4.1-3: Glucose stock solution
Solution 1: Glucose
5 x stock:
50% (w/v) Glucose [e.g. D-(+)-Glucose anhydrous from Sigma Aldrich: cat. No. 49139] (0.5
g/ml) dissolved in water
Dilute to 10% in imaging buffer.
Remark: Glucose needs extended shaking in order to dissolve.
Box 4.1-4: Reducing agents stock solutions
Agent 1: β-mercaptoethylamine (cysteamine) (MEA)
100 x stock:
1 M MEA (0.077 g/ml) dissolved in 360 mM HCl - 1:33 dilution of 37% HCl in water, has
pH 8.5 or
1M MEA-Hydrochloride (0.114 g/ml dissolved in water; pH 8.5)
Remark: Dissolved MEA is unstable. Prepare either fresh or keep aliquots at -20 oC for
prolonged storage. Thaw only once. MEA is used in conentrations between 10 mM and 100
mM. Be aware that the pH can change with MEA, so one should control the pH of the final
buffer and adjust with NaOH or HCl according to one needs.
Agent 2: Glutathione (GSH)
100 x stock:
1 M L-glutathione reduced (0.307 g/ml) dissolved in water
Dilute to 10 mM in the imaging buffer
Agent 3: Dithiothreitol (DTT)
100 x stock:
1M DTT (0.154 g/ml) dissolved in water
Dilute to 10 mM in the imagig buffer
Agent 4: β-mercaptoethanol (β-ME, 2-ME, β-MSH)
100 x stock:
15.3 M β-ME (provided as 100% solution)
Dilute to 1% (v/v) in imaging buffer
Box 4.1-5: Reducing and oxidizing systems (ROXS)
Cocktail 1: Ascorbic Acid / N,N methyl viologen system (AA / MV) system
10 x stock:
5 mM Ascorbic acid (stock in water)
5 mM Methyl viologen
in PBS
Dilute 10 fold in imaging buffer.
Remark: Ascorbic acid is soluble to 0.5 M in water at room temperature. Make a 100x stock
solution. Ascorbic acid (AA) serves as a reducing agent, whereas N,N-methylviologen (MV)
and ambient oxygen serves as oxidizing agents.
Cocktail 2: Trolox / Trolox-Quinone (TX/TQ) system
10 x stock:
10 mM Trolox (100 mM stock in methanol)
in PBS
Remark: Trolox (X) serves as a reducing agent, whereas Trolox-Quinone (TQ) that is built in
the buffer through (photo-) reactions with molecular oxygen, serves as oxidizing agents.
Trolox has antiblinking and antibleaching proerties.
Cocktail 3: Ferrocene-based compound / Nitrobenzoic acid (FC/NBA) system
10 x stock:
1mM Ferrocenne-based compound
1 mM Nitrobenzoic acid (stock in water)
in PBS
4.2
Mountants
Box 4.2-1: Mounting media
Medium 1: Polyvinylalcohol (PVA) – glycerol based
10 % PVA (1 g/ml)
1.
2.
3.
4.
5.
6.
7.
Add 5.8 g of polyvinyl alcohol (PVA) to 12 g of glycerol and mix well.
Add 12 ml of distilled water and leave it on a rotator at room temperature over night.
Add 24 ml of 0.2M Tris-HCl at pH 8-8.5.
Heat in a water bath to 50°C while mixing for about 30 minutes.
(Optionally when an antifade should be added: Add 1.25g of DABCO and mix well.)
Centrifuge at about 2000 rpm for 5 minutes.
Aliquot the supernatant (1 ml aliquots) and store at -20°C.
Remark: 5% solution of polyvinyl alcohol exhibits a pH in the range of 5.0 to 6.5. The
refractive index (RI) of 1% PVA is 1.52 – 1.55. Do not refreeze as PVA polymerizes upon
contact with air. 2-3% PVA in an PALM imaging solution has been shown to support
blinking.
Medium 2: Fluoromount-GTM – aqueous
Remark: Ready-to-use solution with refractive index (RI) of 1.40.
Medium 3: FluoroGel – aqueous
Remark: Ready-to-use solution with refractive index (RI) of 1.358.
Medium 4: FluoroshieldTM – aqueous
Remark: Ready-to-use solution with refractive index (RI) of 1.365.
Medium 5: Crystal Mount – aqueous
Remark: Ready-to-use solution with refractive index (RI) of 1.354
Medium 6: CC/Mount – aqueous
Remark: Ready-to-use solution with refractive index (RI) of 1.364
Medium 7: 2,2'-thiodiethanol (TDE) – aqueous
Remark: Ready-to-use solution. Refractive index can be varied ranging from being that of
water (1.33) to that of immersion oil (1.52) by appropriately diluting with water.
Medium 8: ProLong® Gold Antifade Reagent – aequeous
Remark: Ready-to-use solution containing antifade reagent and with refractive index (RI) of
1.47
Medium 9: VECTASHIELD® Hard-Set™ Mounting Medium – aqueous
Remark: Ready-to-use solution containing antifade reagent and with refractive index (RI) of
1.36. When mounting thick sections, the mounting medium may not harden properly. Use of
non-hardeningVECTASHIELD® Mounting Medium (RI of 1.44) may be preferred.
Box 4.2-2 Antifading reagents
Antifade reagent 1: p-phenylenediamine (1,4-Benzenediamine hydrochloride) (PPD)
0.1% (w/v) PPD stock in water; store at -20 oC
Add 1 part of 0.1% PPD to 9 parts mounting media (Medium 1-7 of Box 4.2-1)
Remark: PPD is the most effective antifade compound. However, it can react with and cleave
cyanine dyes (especially Cy2) .
Antifade reagent 2: (1,4-diazabicyclo[3.3.2]octane) (DABCO)
3.5% (w/v) DABCO in water stock; store at -20 oC
Add 1 part of 3.5% DABCO to 9 parts mounting media (Medium 1-7 of Box 4.2-1)
Remark: Less effective as PPD, but also less toxic. Can be used in live cell work but exhibits
anti-apoptotic properties.
Antifade reagent 3: N-propyl gallate (NPG)
2% (w/v) NPG stock in water; store at -20 oC
Add 1 part of 2% NPG to 9 parts mounting media (Medium 1-7 of Box 4.2-1)
Remark: Not very soluble, needs heating and/or overnight to dissolve. Non-toxic and can
therefore be used with live cells, but exhibits anti-apoptotic properties.
Box 4.2-3 Sealants
Sealant 1: Entellan – Xylene based
Remark. Reday-to-use solution. Although sold as a mountant, Entellan has been
recommended as a sealant as it sets quickly within 20’’.
Nail polish as an alternative should be used with care as it has shown to influence the
fluorescence of certain FPs in some circumstances.
Sealant 2: CoverGrip Coverslip Sealant
Remark. Ready-to-use reagent. Suitable for sealing the edges of wet-mounted coverslips.
Unlike nail polish, CoverGrip Coverslip Sealant contains no ingredients that can leach into
aqueous mounting medium and affect specimen fluorescence.
Sealant 3: Cytoseal 60 – Xylene based
Remark. Ready-to-use reagent. Although sold as a mountant, Cytoseal can be used as a
sealant as it dries quickly.
4.3
PALM / dSTORM embedding solutions
Box 4.3-1: PALM imaging solutions
Dilute either 10 x imaging buffer stock solution (Either of Buffers 1-4 of Box 4.1-1) 1:10 in
MilliQ water.
Remark: These imaging buffers can be used for florescent protein fusions expressed in cells.
Box 4.3-2: dSTORM imaging solutions
Solution 1: Reducing
Combine and dilute
1. 10 x imaging buffer stock (Either of Buffers 1-4 of Box 4.1-1) 1:10
2. 100 x Reducing agent stock (Either of Agents 1-4 of Box 4.1-4) 1:100 in MilliQ
water.
3. Optionally 10 x ROXS cocktail (Either of Cocktails 1-3 of Box 4.1-5) 1:10
Remark: This cocktail is suited for organic dyes including Rhodamines (e.g. Ax488, Ax532,
Ax561, Ax594, Atto 488, Atto565) and Oxazines (e.g. Atto 655, MR121). The optimal
concentration of the reducing agent can vary and should be determined experimentally. If
molecules stay too long in the on-state, increase the concentration. If molecules bleach too
fast, decrease the concentration. Test between 1 mM – 100 mM (MEA, GSH, DTT) for
optimal results. Dilute accordingly.
Solution 2: Reducing and Oxygen scavenging
Combine and dilute
1. 10 x imaging buffer stock (either of Buffers 1-4 of Box 4.1-1) 1:10
2. 100 x (10 x) Reducing agent stock (either of Agents 1-4 of Box 4.1-4) 1:100 (1:10)
3. 5 x Glucose stock (Box 4.1-3) 1:5
4. 10 x (40 x) Oxygen removal stocks (Either of Systems 1 -2 of Box 4.1-2) 1:10 (1:40)
5. Optionally 10 x ROXS cocktail (Either of Cocktails 1-3 of Box 4.1-5) 1:10
in MilliQ water.
Remark: This cocktail is suited for organic dyes including carbocyanines (Ax647, Ax680,
Cy5, C7). Note that oxygen removal increases the reliability of photoswitching, but is not
mandatory. The buffer lasts for about 15 - 30 min in an open environment due to pH decrease
from the production of gluconic acid by glucose oxidase. Therefore, it is not suitable for
imaging in open chambers and the chambers should preferentially be seald with a lid or thin
silicon shet (press-to-seal). Higher Tris concentration or pH will make it last longer. For
example, 100 mM Tris at pH 8.0 can prolong the lifetime to 45-60 min. The optimal
concentration of the reducing agent can vary and should be determined experimentally. If
molecules stay too long in the on-state, increase the concentration. If molecules bleach too
fast, decrease the concentration. Test between 1 mM – 100 mM (MEA, GSH, DTT) for
optimal results. Dilute accordingly. The off-switching of Cy5 is about twice faster with MEA
than with β−ME. Glucose might be tested in concentrations between 1 – 10%. Dilute
accordingly.
Box 4.3-3: PALM mountant imaging media
Combine per 1 ml
800 µl Mountant (either of Medium 1-7 of Box 4.2-1)
100 µl 10 x Imaging buffer stock (either of Buffer 1-4 Box 3.1-1)
100 µl MilliQ H2O
Box 4.3-4: dSTORM mountant imaging media
Media 1: Reducing
Combine per 1 ml
790 µl Mountant (either of Medium 1-7 of Box 4.2-1)
100 µl 10 x Imaging buffer stock (either of Buffer 1-4 Box 3.1-1)
10 µl 100 x Reducing Agent stock (either of Agents 1-4 of Box 4.1-4)
Optional 100 µl 10 x ROXS cocktail stock (either of Cocktails 1-3 of Box 4.1-5)
Optional 1 µl Tetraspecs beads (to serve as fiducial markers)
Up to 1000 µl with MilliQ H2O
Remark: ROXS cocktail has to be prepared as 100 x as otherwise the mountant gets diluted
out too much. β-mercaptoethanol is the preferred reducing agent as it is the most stable. As
RO
Media 2: Reducing & Oxygen Scavenging
Combine per 1 ml
790 µl Mountant (either of Medium 1-7 of Box 4.2-1)
100 µl 10 x Imaging buffer stock (either of Buffer 1-4 Box 4.1-1)
10 µl 100 x Reducing Agent stock (either of Agents 1-4 of Box 4.1-4)
100 µl 100 x Oxygen Scavenging stock (ether system 1-3 of Box 4.1-2)
0.1 g Glucose
Optional 1 µl Tetraspecs beads (to serve as fiducial markers)
Up to 1000 µl with MilliQ H2O
Remark: Since ROXS cocktails need ambient oxygen supply, they will not work in
combination with Oxygen Scavenging systems. Glucose has to be added as a powder as
otherwise the mountant gets diluted out too much. β-mercaptoethanol is the preferred
reducing agent as it is the most stable.
5
5.1
Suppliers of lab ware & reagents
Suppliers of coverslips & coverslip accessories
Box 5.1-1: Suppliers of coverslips and fiducial covered coverslips
Warner Instruments (a Harvard Apparatus Company) [http://www.warneronline.com/]
coverslips #1.5 (0.17 mm) Model No. CS-25R15 25 mm Ø (Cat. No.
64-0715)
Marienfeld Superior (Laboratory Glassware) [http://www.marienfeld-superior.com/]
Precision cover glasses thickness #1.5H (0.17 mm) for high performance microscopes 25
mm Ø (Cat. No. 0117650) and 22 x 22 mm (Cat. No. 0107052)
ThermoScientific Nunc [http://www.nuncbrand.com]
Thermanox™ coverslips #1.5H (0.17 mm) 22 mm Ø (Cat. No. 174977) and 22 x 22mm (Cat.
No.
Bioscience Tools [http://www.biosciencetools.com]
High Precision Glass Coverslip #1.5 with 18 mm Ø (Cat. No. CSHP-No1.5-18) and 25 mm
Ø (Cat. No. CSHP-No1.5-18)
Glass Coverslips, 22x22mm (Cat No. CS-No1.5-22x22)
Hestzig LLC [http://www.hestzig.com/]
Fiducial coverslips 500 ± 50 nm spectral range at two densities (Cat. No. 550-30AuF & 550100AuF); Fiducial coverslips 600 ± 100 nm spectral range at two densities (Cat. No. 60030AuF & 600-100AuF)
Box 5.1-2: Suppliers of coverslip racks
Thomas Scientific [http://www.thomassci.com/]
Corrosion-resistant staining rack for holding coverslips (Cat. No. 8542E40)
Invitrogen – Molecular Probes [www.invitrogen.com/]
Coverslip Mini-Rack, for 8 coverslips (Cat. No. C-14784)
Electron Microscopy Sciences [http://www.emsdiasum.com]
Wash N'Dry Cover Slip Rack (Cat. No. 70366-16)
Box 5.1-3: Suppliers of coverslip chambers
Invitrogen – Molecular Probes [www.invitrogen.com/]
Attofluar® Cell Chamber for 25 mm coverslips (Cat. No.: A-7816); Spare O-rings (Cat. No.
O-14804)
Life Cell Instrument (LCI) [http://www.chamlide.com/]
ChamlideTM magnetic chambers (CMB) standard for 25 mm Ø (Cat. No. CM-B25-1), 22 mm
Ø (Cat. No. CM-B22-1) and 18 mm Ø (Cat. No. CM-B18-1)
1 well Chamlide CMS for 22 x 22mm coverslips (Cat. No. CM-S22-1)
Bioscience Tools [http://www.biosciencetools.com]
Chamber for replaceable 18 mm Ø (Cat. No. CSC-18) and 25mm Ø (Cat. No. CSC-25)
coverslips; replacement top glass cover for CSC holders 30 mm (Cat. No. CS-30)
Low profile chamber for replaceable 18 mm Ø (Cat. No. CSC-18L) and 25mm Ø (Cat. No.
CSC-25L) coverslips
Chamber for replaceable square 22 x 22mm coverslips (Cat. No. CSC-22x22)
Pecon [http://www.pecon.biz/]
POC Chamber System mini-2 for 17-22 mm Ø coverslips (Cat. No. POC mini-2)
5.2 Suppliers of glass bottom cell culture dishes, chambers and
plates
Box 5.2-1: Suppliers of glass bottom dishes, chambers and plates
Greiner bio-one [http://www.greinerbioone.com]
CELLview™ - Glass Bottom Dish one compartment, non-treated (Cat. No. 627861), tissue
culture treated (Cat. No. 627860) and advanced tissue culture treated (Cat. No. 627965)
Willco Wells [http://www.willcowells.com/]
WillCo-dish® Glass bottom dish #1.5 (0.16 - 0.19mm), 35 mm dish Ø, 12 mm ( Cat. No.
GWSt-3512) and 22 mm (Cat. No. GWSt-3522) glass Ø
MatTek Corporation [http://www.mattek.com/] & [glass-bottom-dishes.com/pages/]
Glass Bottom dishes #1.5, 35 mm dish Ø, 10 mm (P35G-1.5-10-C Case), 14 mm (Cat. No.
P35G-1.5-14-C Case ) und 20 mm (Cat. No. P35G-1.5-20-C Case) glass Ø
World Precision Instruments, Inc. [http://www.wpiinc.com/]
FluoroDish cell culture dish #1.5, 35.5 mm dish Ø, 10 mm (Cat. No. FD3510-100) and 24.5
mm (Cat. No. FD35-100) glass Ø
Asahi Technoglass Scitech Division IWAKI brand [http://www.atg.ushop.jp/]
Iwaki Glass Bottom dish #1.5, 35 mm dish Ø, 12mm (Cat No. 3911-035) and 27 mm (Cat.
No. 3910-035) glass Ø
ThermoScientific Nunc [http://www.nuncbrand.com/]
96 Well Optical Bottom Plates with #1.5 cover glass (Cat. No. 265300)
LabTekTM II Chambered Coverglass #1.5 Borosilicate, 8 well (Cat. No. 155409)
Sigma-Aldrich [http://www.sigmaaldrich.com]
Nunc® MicroWell 96 well optical bottom plates (Cat. No. P8866)
5.3
Suppliers of fiducial markers & fluorescent standards
Box 5.3-1: Suppliers of gold nanoparticles
BB International [www.british-biocell.co.uk] & [http://www.bbigold.com]
Gold Colloid 80 nm (Cat. No. GM.GC 80) and 100 nm (Cat. No. GM.GC 100)
Microspheres-Nanospheres (http://www.microspheres-nanospheres.com/)
40 nm (Cat. No. 790122-010) and 80 nm (Cat. No. 790120-010) nanospheres Au particles
Nanopartz [www.nanopartz.com]
Nanorods 550 (Cat. No. 30-25-550) and 600 (Cat. No. 30-25-600)
Box 5.3-2: Suppliers of fluorescent beads
Invitrogen [www.invitrogen.com/]
Molecular ProbesTM TetraSpeck™ Microspheres, 0.1 µm, Fluorescent
Blue/Green/Orange/Dark Red (Cat. No. T-7279)
Box 5.3-2; Suppliers of quantum dots
Invitrogen [www.invitrogen.com/]
Molecular ProbesTM Qdot® 655 Streptavidin Conjugate (Cat. No. Q10121MP)
Remark: Due to their blinking behavior, Qdots are not as well suited as fiducial markers than
gold or TetraSpeckTM.
5.4
Suppliers of mounting media, antifading reagents & sealants
Box 5.4-1: Suppliers of mountants
SouthernBiotech [http://www. southernbiotech.com]
Fluoromount-GTM (Cat. No.: 0100-01)
Sigma-Aldrich [http://www.sigmaaldrich.com]
FluoromountTM (Cat. No. F4680)
FluoroshieldTM (Cat. No. F6182)
PVA Type II (Cat. No. P8136)
TDE (Cat. No. 88559)
Crystal Mount (Cat. No. C0612)
CC/Mount (Cat. No. C9368)
Electron Microscopy Sciences [http://www.emsdiasum.com]
FluoroGel (Cat. No. 17985-10)
Fluoromount-GTM (Cat. No. 17984)
Interchim [http://www.interchim.com/]
Fluoro-Gel (Cat. No. FP-AL2551), w/ Tris buffer (Cat. No. FP-AL256), TES buffer (Cat.
No. FP-AL255A) or PIPES buffer (Cat. No. FP-AL255C)
Fluoromount-GTM (Cat. No. FP-483331)
Invitrogen [www.invitrogen.com/]
ProLong® Gold Antifade Reagent (Cat. No. P36930)
Vector Labs [https://www.vectorlabs.com/]
Vectashield HardSet Mounting Medium (Cat. No. H-1400)
Vectashield Mounting Medium (Cat. No. H-1000)
Box 5.4-2: Suppliers of antifade reagents
Sigma-Aldrich [http://www.sigmaaldrich.com]
p-phenylenediamine (PPD), ≥99.0% (GC/NT) (Cat. No. 78429)
n-Propyl gallate (NPG) for microscopy, ≥98.0% (HPLC) (Cat. No 02370)
1,4-Diazabicyclo [3.3.2]-octane (DABCO) D2522
Box 5.4-3: Suppliers of coverslip sealants
Merck Millipore [http://www.merckmillipore.com]
Entellan® (Cat. No. 107960)
ThermoScientific [http://www.thermoscientific.com]
Richard-Allan Scientific Cytoseal™ 60 (Cat. No. 8310-4)
Biotium [http://www.biotium.com]
CoverGrip Coverslip Sealant (Cat. No. 23005)
Box 5.4-3 Suppliers of chamber sealants
Invitrogen [http://www.invitrogen.com]
Press-to-Seal™ Silicone Sheet, 13 cm x 18 cm, 0.5 mm thick (Cat. No. P-18178)
Press-to-Seal™ Silicone Sheet, 13 cm x 18 cm, 1.0 mm thick (Cat. No. P-18179)
5.5
Suppliers of cell culture reagents
Box 5.5-1 Suppliers of transfection reagents
Invitrogen [http://www.invitrogen.com]
Lipofectamine® Transfection Reagent (Cat. No. 18324-012)
Qiagen [http://www.qiagen.com]
Effectene® Transfection Reagent (Cat. No. 301425)
Roche Applied Science [http://www.roche-applied-science.com]
FuGene® HD Transfection reagent (Cat. No. 04 709 691 001)
Lonza [www.lonza.com/]
AmaxaTM Basic NucleofectorTM Kit (Cat. No. VPI-1002)
Box 5.5-2: Suppliers of cell culture media
Sigma-Aldrich [http://www.sigmaaldrich.com]
DMEM F-12 (Cat.No. D8900)
Trypsin-EDTA Solution 10X 0.5% trypsin, 0.2% EDTA, trypsin gamma irrdiated without
phenol red, in saline (Cat. No. 59418C)
Invitrogen [http://www.invitrogen.com]
Gibco BRL DMEM High Glucose (DMEM-HG) (Cat. No. 31053-044)
Gibco BRL DMEM, High Glucose, HEPES, no Phenol Red (DMEM-HG) (Cat. No. 21063029)
Biochrom [http://www.biochrom.de/en/home/]
Hanks‘ buffered saline (HBSS, w/o Ca2+, w/o Mg2+, w/o Phenol red) (Cat. No. L 2045)
Lonza [www.lonza.com/]
Reagent Pack™ Subculture Reagent Kit containing Trypsin/EDTA, HEPES Buffered
Saline Solution (HBSS) and Trypsin Neutralizing Solution (TNS) (Cat. No. CC-5034)
Box 5.5-3: Suppliers of coverglass coating reagents
Sigma-Aldrich [http://www.sigmaaldrich.com]
Poly-L-lysine (Cat. No. P4832)
Fibronectin (Ct. No. F0895)
Box 5.5-4: Suppliers of coverglass blocking reagents
Sigma-Aldrich [http://www.sigmaaldrich.com]
Normal Goat Serum (NGS) (Cat. No. G6767)
Albumin from Bovine Serum (BSA) lyophilized powder, (RIA Grade) (Cat. No A7888)
Glycine ReagentPlus®, ≥99% (TLC) (Cat. No. G7126)
Box 5.5-4: Suppliers of coverglass cleaning solutions
Thermo Scientific Pierce [www.piercenet.com/]
RBS-35 liquid detergent concentrate (Cat. No. 27950)
Sigma-Aldrich [http://www.sigmaaldrich.com]
RBS-35TM solution concentrates (Cat. No. 83461)
Ethanol (EtOH) reagent, denatured, spectrophotometric grade (Cat. No. 245119)
Hydrofluoric acid (HF), 48 wt. % in H2O (Cat. No. 339261)
Hydrogen Peroxide (H2O2), 50% in H2O, stabilized (Cat. No. 516813)
Fisher Scientific [http://www.fishersci.com/]
Hydrogen Peroxide, 50% (Stabilized/Certified), Fisher Chemical (Cat. No. H341-500)
Ammonium Hydroxide, about 15.8 N (Certified ACS Plus), Fisher Chemical (Cat. No.
A669-500)
Box 5.5-5: Suppliers of fixatives
Electron Microscopy Sciences [http://www.emsdiasum.com]
Paraformaldehyde (Cat. No. 19208)
Formaldehyde EM grade, 32% solution (Cat. No. 15714-S)
Glutaraldehyde EM grade, 25% solution (Cat. No. 16200)
Sigma-Aldrich [http://www.sigmaaldrich.com]
Paraformaldehyde (PFA) (Cat. No. P6148)
Glycine (Cat. No. G-7126)
Sodium borohydride NaBH4 (Cat. No. 452882)
Methanol ACS spectrophotometric grade (Cat. No. 154903)
Acetone ACS spectrophotometric grade (Cat. No. 154598)
5.6
Suppliers of chemicals
Box 5.6-1 Suppliers of detergents
Sigma-Aldrich [http://www.sigmaaldrich.com]
Triton XTM-100 (Cat. No. T9284)
Tween 20 (Cat. No. P7949)
Box 5.6-2 Suppliers of Redox reagents
Sigma-Aldrich [http://www.sigmaaldrich.com]
Trolox® = (±)-6-Hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid (Cat. No. 238813)
Ferrocene-based compound, e.g. Ferrocenecarboxylic acid (Cat. No. 106887)
3-Nitrobenzoic acid (Cat. No. 85329)
L-Ascorbic acid (Cat. No. A5960)
Methyl viologen dichloride hydrate (Cat. No. 856177
Glucose Oxidase from Aspergillus niger (Cat. No. G0543)
Catalase from Bovine liver (Cat. No. C3155)
Horseradish Peroxidase Type VI-A (Cat. No. P6782)
L-Gluthatione reduced (Cat. No. G4251)
DL-Dithiothreitol (DTT) (Cat. No. 43815)
β-mercaptoethanol (β-ME) (Cat. No. 63689)
Cysteamine Hydrochlorid ((β-Mercaptoethylamine, MEA) (Cat.No. 6500)
Cysteamine (β-Mercaptoethylamine, MEA) (Cat.No. 30070)
Merck Millipore [http://www.merckmillipore.com]
Calbiochem® Trolox® (Cat. No. 648 471)
Box 5.6-3 Suppliers of buffers and buffer ingredients
Sigma-Aldrich [http://www.sigmaaldrich.com]
Phosphate buffered saline 10 x concentrate (PBS 10x) – BioReagent (Cat. No. P5493)
Dulbecco’s Phosphate buffered saline 10× concentrate (DPBS 10x) (Cat. No. D1238)
Potassium chloride (KCl) – ACS reagent, 99.0-100.5% (Cat. No. P3911)
Potasium dihydrogen phosphate (KH2PO4) – ACS reagent, ≥99.0% (Cat. No. P0662)
Sodium chloride (NaCl) – ACS reagent, ≥99.0% (Cat. No. S9888)
Disodium hydrogen phosphate heptahydrate (Na2HPO4•7H2O) – ACS reagent, 98.0-103.0%
(Cat. No. S9390)
Calcium chloride dihydrate (CaCl2•2H2O) – ACS reagent, ≥99.0% (Cat. No. 223506)
Magnesium chloride hexahydrate (MgCl2•6H2O) – ACS reagent, 99.0-103.0% (Cat. No.
M9272)
PIPES free base, ≥99% (titration) (Cat. No. P6757)
PIPES sodium salt, ≥99% (titration) (Cat. No. P2949)
HEPES, ≥99.5% (titration) (Cat. No. H3375)
Tris (hydroxymethyl) aminomethane, ≥99.9% (titration) (Cat. No. 154563)
Ethylene glycol-bis(2-aminoethylether)-N,N,N,N-tetraacetic acid (EGTA), ≥97.0% EGTA
(Cat. No. E4378)
Sodium hydroxide (NaOH) – BioXtra, ≥98%, pellets (anhydrous) (Cat. No. S8045)
Sodium azide (NaN3) – BioUltra, ≥99.5% (T) (Cat. No. 1289)
Hydrochloric acid (HCl) – 36.5-38.0%, BioReagent, for molecular biology (Cat. No. H1758)
5.7
Suppliers of filter devices
5.7-1 Supplier of filters
Fisher Scientific [http://www.fishersci.com]
Fisherbrand Filter P8 (Cat. No. 09-7958)