Exam 2019 questions Flashcards
typical resolutiokn values for LM
~200nm
plug in NA~1.4
and wavelength from 400-650 nm
What can influence the actual obtainable resolution in a practical experiment with
cells or tissue stained with fluorophores compared to the theoretical resolution
formula? How can you measure the actual resolution in a classical diffraction
limited imaging system?
The theoretical resolution is only obtainable with infinite SNR. The SNR can go down
due to e.g. out of focus light or non-specific binding of fluorophores. Next to this, the
sample can induce aberrations, due to refractive index mismatch between the
sample and the immersion or due to heterogeneity of the sample itself.
The actual resolution can be measured by FRC or by imaging sub-diffraction limited
sized beads (e.g. diameter ~50 nm) and then measuring the effective PSF of the
system, or by imaging single point emitters. Then measuring e.g. the FWHM (Full width at half maximum)
Precision of localizing single fluorophore
Abbe/sqrt(N)
N(#photons)~100-10,000. Where 1000 are most common.
thus 5-50 nm dx
FRC/FSC for measuring the resolution. As a poor man’s solution
you could image either known structures and check if you can resolve them, e.g. the
hollow structure of microtubili, DNA origami rulers. That is “poor man’s” because the
structure spatial structure has to be known, and not over all spatial ranges they are
available.
The fluorescent probes should/must have certain properties in SMLM. Describe
them.
ratio
off at least 1/100 (on/off).
the dyes should be small, and shortly connected to the
epitope (e.g. no secondary antibody labelling). The dye should emit as many photons
as possible during their on-state (high quantum yield). In addition the dyes should
satisfy similar conditions as for traditional fluorescent imaging: specific binding, large
Stokes shift.
How can you measure the amount of fluorophores in PALM?
In PALM the number of localizations is not easily related to the number of
fluorophores present. Even if all fluorophores are imaged (which can take a long
time to ensure proper imaging conditions), a single dye can be imaged many rounds
(for PALM this can be >1-10 times) as the photobleaching step might not work
properly (over/under counting problem). In addition you would like to relate the
number of visible fluorophores to the number of fluorophores present. This number
can be largely different due to immaturation of proteins or misfolding. If you want to
be really sure you need to perform titration curves as explained during the literature
discussion.
How can you measure the amount of fluorophores for STED?
In STED imaging the measured intensity is linear with the amount of visible
fluorophores if you do not have bleaching effects. Therefore at least relative
concentrations are easily measurable. Absolute counting the number of
fluorophores is in general not possible (or very, very hard practically as you need to
calibrate the intensity carefully)
3D SIM cannot give very good sectioning beyond …
10 micrometer
How do you get sheets of light?
The sheet of light is typically
obtained by scanning a laser line very fast up and down
method for conducting 3D EM for a whole embryo
block-face
approaches (Serial Block-Face EM or FIB-SEM) or serial sections (with TEM or SEM).
With block-face the upper layer of a resin block is imaged and then sliced off with a
microtome inside the SEM (SBF-SEM) or milled away with a focused ion beam (FIB).
This can give the highest z-resolution (even isotropic resolution with FIB-SEM), but is
slow as everything has to be scanned at high resolution because the sections are lost
during imaging. With serial sections, all sections can be re-scanned and thus one can
first inspect everyting at low resolution to identify the areas for higher resolution
inspection. An alternative advantage for serial sections may be that sections could
be immuno-labelled post-sectioning (post-embedding), aiding correlation to the light
microscopy (LM) and also giving equal z-resolution for LM and EM
Potential disadvantages for serial sections are: (1) all sections have to be collected in
the order of sectioning, loss of sections during the process leads to data loss; (2)
each section may be distorted slightly differently by the sectioning, which can make
the 3D reconstruction challenging
Potential disadvantages for block-face are: (1) sections are lost after imaging and can
not be re-imaged; (2) imaging the upper side of the resin block may be prone to
charging and thus artefacts in the images; (3) Fixation and staining may need to be
relatively strong leading to more black-white contrast while grayscale may be
prefered for data interpretation, also possibilities for fluorescence preservation and
correlation may be less: fluorescence on sections could match z-resolution for EM
and LM and thus give better correlation
Sequence of sample prep for EM
Fixation - Staining - Embedding – Sectioning. Staining could also be conducted after
sectioning. Labelling could also be mentioned either before or after embedding.
acquisition time for electron
microscopy data of a 500x100um embryo
General SEM thick sections for tissues are ~100nm
=> 500.000 / 100 = 5000 sections
Each sections covers 100 μm x 100 μm =10^10 nm^2
With 10 nm x 10 nm = 100 nm^2 pixels, we would 10^8 pixels per sections.
With pixel dwell time on the order of 10’s μs, say 50 μs, we would have
5x50x10 11 x10 -6 = 2.5x10 7 seconds of scan time.
This is about 7000 hours or ~290 days
excluding overhead like sample/stage movements, re-
focusing, loading new sections, etc.
Important experimental parameters are section thickness, pixel size, pixel dwell
time, and overhead
Beam current vs brightness
The (reduced) brightness is the current per area per solid angle (per voltage). While
the current in the beam may vary in the column due to the use of apertures, the
(reduced) brightness may be conserved throughout the column
SE vs BSE information
Secondary electrons are low-energy (under 50eV) electrons and originate only from the
upper few nanometers in the sample. They does give contrast depending on sample
topography. Back-scattered electrons are high-energy electrons (above 50 eV, but mostly
close to the primary beam energy. Higher atomic number materials generate more
back-scattered electrons.
Describe how the energy spectra of the electrons after the specimen
plane can be used to determine the elemental composition of the
sample. Also describe how these electron energies can be
determined.
As electrons pass through a specimen, they interact with atoms of the sample.
Many of the electrons pass through the thin sample without losing energy. A
fraction will undergo inelastic scattering and lose energy as they interact with the
specimen. The energy loss will be specific for particular elements. The energy
distribution of all the inelastically scattered electrons provides information about
the local environment of the atomic electrons which in turn relates to the physical
and chemical properties of the specimen. This is the basis of electron energy-loss
spectroscopy (EELS). One can detect these scattered incident electrons in
a spectrometer (energy filter) as it gives rise to the electron energy loss signal
negative stain vs cryoEM
Quick, cheap, simple & gives strong (amplitude) contrast. Stain does not
penetrate the particle and hence only provides low resolution information. Artifacts
due to adsorption and staining are frequent.
Sample is in a near-native state in (often) physiological buffer solution. No
adsorption is involved in the process and hence no artifacts from adsorption are
present. Difficult, expensive and time-consuming; achieving consistent ice
thickness is trial-and-error procedure and often not very reproducible. Due to the
absence of stain, amplitude contrast is poor.
