Kalkvik, Project Assignment

April 8th
2014
Maria Helene Kalkvik
Cell Biology BI2012
N o r w e g i a n U n i v e r s i t y o f S c i e n c e a n d T e c h n o l o g y
Live cell imaging and advanced image analysis
of the Golgi apparatus in plants

2
Abstract
This experiment was conducted in order to obtain a better understanding of laser scanning
confocal microscopy, and also study the movement and stucture of the Golgi apparatus and
endoplasmatic reticulum in plant cells. The chosen organells were labled with fluorescent
proteins in order to observe them in the microscope. Arabidopsis thaliana was the chosen
organism for the experiment. A Leica SP5 microscope was used along with computer software.
Image processing softwares Amira and ImageJ were used to analyse the images of the specimen.
Overview images of both the ER and Golgi apparatus was captures, and measurements of the
Golgi apparatus’ size were contucted. Point spread function of fluorescent beads was determined
in order to perform deconvolution.
Abbreviations
GFP: Green fluorescent protein
PSF: Point spread function
FWHM: Full with half maximum
ROI: Region of interest
ER: Endoplasmatic reticulum
LUT: Look up table
NA: Numerical aperture

3
Table of contents
ABSTRACT....................................................................................................................................... 2
ABBREVIATIONS........................................................................................................................... 2
1 INTRODUCTION ........................................................................................................................ 4
1.1 CONFOCAL MICROSCOPY ................................................................................................................................ 4
1.2 FLUORESCENSE................................................................................................................................................. 5
1.2.1 Green Fluorescent Protein.............................................................................................................................. 5
1.3 POINT SPREAD FUNCTION .............................................................................................................................. 6
1.4 DECONVOLUTION............................................................................................................................................ 7
1.5 RESOLUTION ..................................................................................................................................................... 9
DYNAMIC RANGE..................................................................................................................................................11
1.6 BIOLOGICAL SAMPLES.......................................................................................................... 11
1.6.1 ARABIDOPSIS THALIANA ............................................................................................................................11
1.6.2 THE PLANT CELL .........................................................................................................................................12
1.6.3 ENDOPLASMATIC RETICULUM ..................................................................................................................12
1.6.4 THE GOLGI APPARATUS .............................................................................................................................12
1.6.5 ACTIN.............................................................................................................................................................12
2 MATERIALS AND METHODS..................................................................................................13
2.1 SPECIMEN PREPARATION..............................................................................................................................13
2.2 BEAD PREPARATION ......................................................................................................................................13
2.3 MICROSCOPE ...................................................................................................................................................13
2.4 IMAGE PROCESSING AND SOFTWARE .........................................................................................................13
2.5 OPTIMAL RESOLUTION..................................................................................................................................14
2.6 PSF ANALYSIS..................................................................................................................................................15
2.7 DECONVOLUTION..........................................................................................................................................15
3 RESULTS......................................................................................................................................15
3.1 IMAGE ANALYSIS OF THE GOLGI APPARATUS ...........................................................................................15
3.2 IMAGE ANALYSIS OF THE ENDOPLASMATIC RETICULUM........................................................................16
3.3 DECONVOLUTION OF THE GOLGI Z-STACK.............................................................................................17
3.4 SIZE ANALYSIS OF THE GOLGI Z-STACKS ..................................................................................................17
4 DISCUSSION ...............................................................................................................................18
4.1 IMAGE ANALYSIS OF THE GOLGI APPARATUS AND ER...........................................................................18
4.2 OPTIMAL RESOLUTION..................................................................................................................................18
4.3 PSF ANALYSIS..................................................................................................................................................18
4.4 DECONVOLUTION OF THE GOLGI Z-STACK..............................................................................................19
4.5 SIZE ANALYSIS OF THE GOLGI STACKS ......................................................................................................19
5 LITTERATURE ...........................................................................................................................21
5.1 ILLUSTRATIONS...............................................................................................................................................22
7 APPENDIX...................................................................................................................................23
A CALCULATIONS .................................................................................................................................................23
A.1 Resolution limit, pixel size and number of pixels needed.................................................................................23
B METROLO J ANALYSIS ......................................................................................................................................24
B.1 PSF analysis of fluorescent beads....................................................................................................................24
B.2 Size analysis of the Golgi stacks .....................................................................................................................24

4
1 Introduction
1.1 Confocal microscopy
Laser scanning confocal microscopy has become an important tool for studying
and observing biological cells. Through this technique one can obtain knowledge
on how the organells in the cells move through the cytosol, and gain insight in
cellular functions. It also gives the opportunity to perform optical cross sections of
transparent sampels, withouth physically slicing them into sections. The ability to
remove glare from out of focus layers has been an important improvent for studies
involving biological imaging. [1]
In a laser scanning confocal microscope (LSCM) a laser is used as a light source.
The light is then filtered by an acousto-optical tunable filter (AOTF), which allows
for regulation of both the wavelength of emitted light from the laser as well as
exitation intensity. These microscopes makes it possible for examination of
fluorescence emisson from 400 to 750 nanometers. [2]
After the light is filtered by an AOTF, the light is reflected by a dicromatic mirror,
which only reflects certain wavelengths and let others pass through. The Leica SP5
uses an acousto-optical beam splitter (AOBS) insted of a dicromatic mirror. Here,
the crystalline materials only reflect certain wavelengths of light by the interaction
of acoustic waves. This is done by manipulates the refractive index of the crystal.
[3]
The specimen is scanned by the laser, while two high speed oscillating mirrors
direct the beam in a certain pattern across a chosen area of the specimen. One
mirror controls the light along the x-axis, and the other along the y-axis. When a
region of interest is found (ROI), the light moves along the x-axis from a starting
point, and then returns to the starting point to scan in the y-dimension. [4]
Before the emission light reaches the detectors, it passes through a pinhole
aperture. The pinhole’s function is to reduce light disturbance, like blur, from
planes above and below the plane of focus. Only a small part of the light from
other planes in the specimen will pass through the pinhole (Figure 1). [4]
Figure 1: Illustration of the pin hole’s function in a confocal microscope 2

5
1.2 Fluorescense
Fluorescense is the ability certain compounds have to emit visible light. When the
compund is exited by a photon, electrons move to a higher state of energy. This
state is highly unstable which results in the electrons moving back to their original
state, and when they do, energy is released. Energy in this case is in the form of
light. Different compounds have different electron configuration and orientation of
chemical bonds. This leads to emission of different wavelengts as the exited state
for the electron may be in an higher or lower state of energy, depending on the
compund. Since light can be viewd as a form of energy, differences in energy
because of electron configuration, the electrons exited state and its original state,
leads to emission of different light.
The wavelenth of light required to exite a compound is designated as 𝜆!"#
, where
the unit is nanometers (nm). Some of the energy is lost as it forms heat and
vibration, not only photons (light). Therefore the photons emitted from the
compund has less energy, and thus a longer wavelength than the light used to exite
the compound. This difference between the exitation wavelength and emission
wavelength is called Stokes shift. The emission wavelength is given by 𝜆!"
, and is
also in nanometers.
1.2.1 Green Fluorescent Protein
Green fluorescense protein (GFP) was isolated from the jellyfish Aequorea victoria.
Research shoes that the molecule is able to fluoresce when it is expressed by itself
in any organism, thus making it a well suited marker for biological studies and cell
imaging [5]. GFP consists of eleven 𝛽-barrels and one 𝛼-helix, surrounding the
cromophore in the core (Figure 2). The cromophore is the part of the molecule
that makes fluorescense possible. It has been modified in order to create a more
stable and effective version of the fluorescent protein, compared to the wild type
found in A. victoria. One of the modified versions are called enhanced yellow
fluorescent protein (EYFP). It has a 𝜆!"#
of 514 nm and a 𝜆!"
of 527 nm. [6]

6
Figure 2: A model of Green Fluorescent Protein (GFP). The cromophore is not
depicted in the model. The model was created in the software Amira, by using the
protein data bank, 1EMA.
When studies involving fluorescent proteins are contucted, the protein is fused to
proteins who binds to, or target the chosen organelle. When studying ER, the
protein is fused to a signal peptide called Arabidopsis thaliana wall associated
kinase 2 (AtWAK2 ). The fusion happens at the N-terminus of the peptide, and the
ER recognises the signal sequence His-Asp-Glu-Leu at the C-terminus. When
studying the Golgi apparatus the fluorescent protein is fused to a cytoplasmatic and
transmembrane domain of a protein calles soybean 𝛼-1,2-mannosidase I.[6]
1.3 Point spread function
A point spread function is a result of diffraction of light in a specimen. When light
travels through the microscope it is defracted as a result of interactions with
materials it passes through, lenses. The consequence of light diffraction is that a
given point in the specimen will seem larger than it actually is. It can be said that
the PSF is a measurement of how many neighboring pixels are affected when one
pixel is fully enlightened. The given point will often be surrounded by alternating
dark and light rings, as a result of light diffraction. This patterns is refered to as an
Airy pattern, and was first discovered by George Biddel Airy. [7] The central disc is
called an Airy disc. This pattern in three dimentions is what creates what is called a
point spread function, since it describes how light spread out from a single point.
[2]

7
Figure 3: The corrolation between an airy pattern and a point spread function. The graph (red)
shows the PSF as light intensity, with the airy pattern in the background. 3
1.4 Deconvolution
Blur is created as a result of the diffraction of light when it interacts with lenses in
the microscope. As blur is an undesirable factor when working with image analysis,
it can be corrected for by using the PSF. Every point, or pixel, in the image is
essensially a point spread function with a corresponding airy pattern. The out of
focus blur is a result of the alternating dark and light rings of the airy pattern. The
goal is to remove these rings, and obtain a single point in focus. In order to do so,
one must apply the point spread function to every point of the object. [8]
Deconvolution is a mathematical process based on algoriths, that determines the
most likely estimation to reassigh out of focus blur back to its point source. To
determine the point spread function, beads with a known size can be used to
calculate it. A 3D image of the beads can be analysed with image processing
software (Figure 4).

8
Figure 4: A 3D visualization of beads in the image processing software Amira.
As the PSF shows spreading of light from the point source, it can be used to
reverse the blurring effect of convolution. The PSF can be applied to chosen
images by using image processing software, and thereby reducing the blur in the
image. This creates a more realistic depiction of the specimen.

9
Figure 5: The raw image (left) and the deconvolved image (right). The grana in the chloroplasts
was almost impossible to observe before deconvolution.
1.5 Resolution
Several factors affect resolution of an image taken with a confocal microscope. The
numerical aperture (NA) of the objective plays a big role in the resolution of an
image. NA is defined by [10]:
𝑁𝐴 = sin 𝛼 𝑥 𝜂 (1)
Where 𝛼 is defined as half the angle of the cone of ligth the objective captures
from the focal point, and 𝜂 is the refractive index of the immersion medium used
[10]. As a result, when the objective size increases or the refractive index increases,
so does the NA. A high NA value gives a higher resolution.
The Rayleigh criterion is an optical unit that describes the minimal distance
between two point sources at which they are distinguishable from each other. It is
given by the formula [9]:
𝐷!! =
!,!"!!"#
!"
(2)
Where 𝐷!" is the Rayleigh criterion, 𝜆!"#
is the wavelength of the excitation light,
and NA is the numerical aperture of the objective. In the axial dimension, the
Rayleigh criterian can be calculated using the following equation [9]:

10
𝐷! =
!!!"#!
(!")!
(3)
Where 𝜂 is the refractive index of the immersion medium. The most common
immersion mediums are water, air, oil and glass. Their refractive indices are
presented (Table 1).
Table 1: Refractive indices of the immersion mediums water, air and oil/glass [12]
Immersion medium Refractive index
Air 1.00
Water 1.33
Oil/Glass 1.52
To illustrate how the NA affects PSF, one can generate different PSFs with varying
NA values (Figure 6).
Figure 6: How NA affects PSF. The NA varies from 1.0 (left), 1.2 (middle) and 1.4
(left). The 𝜆!"#
was set to 514 for all PSFs. The model was created using the Amira
software.
The sampling rate is also an influencial factor to resolution. The Nyquist theorem
states that it should be two samples per resolvable element to obtain an accurate
image.[9] The size of the pixel determines the sampling rate, as one pixel only can
have one light intensity value. Therefore the the sampling rate directly correlates to

11
the pixel size. The following equation must be used to calculate pixel size in lateral
dimentions, in order to fulfill the Nyquist theorem:
𝑃𝑠!" =
!!"
!
(4)
Where 𝑃𝑠!" is the pixel size.
A voxel is relevant when the axial dimension is under study. While a pixel only has
two dimensions, a voxel has three, making it a volume element rather than just an
area. The voxel size is determined by the pixel size, in the x- and y-dimensions, and
also the distance between scans in the z-dimension. The Nyquist theorem can also
be applied here. The depth of a voxel should not exceed the value given by the
following equation:
𝑃𝑠! =
!!
!
(5)
Where 𝑃𝑠! is the voxel depth.
The number of pixels needed in an image can be calculated using the following
equation, by taking the Nyquist theorem in account:
#𝑝𝑥 =
!
!!"
(6)
Where the number of pixels needed in both the x and y dimensions are denoted
#𝑝𝑥, and 𝑠 represents the size of the ROI.
Dynamic range
To ensure good image quality, the microscope and the software needs to be set up
and adjusted correctly. The bit depth is an important aspect to consider. A high bit
depth will be more likely to produce an accurate depiction of the object under
study. A low bit depth results in a larger variance in light intensity [12]
The software needs to be adjusted so that the image is shown within the detectors
dynamix range. A LUT is usually used, as it gives different colours to different light
intensities.[12] One should make sure that no pixels are completely saturated, in
order to achieve a better understanding of the pixels intensity relative to each other.
1.6 Biological samples
1.6.1 Arabidopsis thaliana
Arabidopsis thaliana is one of the most commonly used species in plant studies. It
is a plant with simple growth requirements and a relativly short life cycle of eight
weeks, which makes concucting studies on the plant quite easy. Also, A. Thaliana’s

12
genome has been fully mapped, and it is therefore often used in genetic
experiments. The plant is also susceptible to genetic transformations, only by
spraying the plant with a bacterium which holds the gene of interest.
1.6.2 The plant cell
The plant cells contain an endomembrane system that incorporates all membrane
bound organelles in the cell. These organelles are the endoplasmatic reticulum and
golgi apparatus, which will be discussed in the report, and also the nuclear
envelope, the cell membrane, vacuoles, lysosomes and transport vesicles. The
vaculoles controlles turgor pressure in the cell, as well as it works as a storage space
for water and inorganic ions among other things. As a result, the plant vacuole is
quite large, which only makes it possible for the other organelles to occupy the
small cytoplasmic space between the central vacuole and the cell membrane.[13]
1.6.3 Endoplasmatic reticulum
The endoplasmati reticulum (ER) is an organelle consisting of both the rough and
smooth ER, which have specialized functions in the cell. It is a continuous network
of tubules and sacs. The rough ER plays a prominent role in protein synthesis, as
well as protein modification and marking. It is called the rough ER because of
ribosomes sitting on its membrane, giving it an uneven surface. The smooth ER is
active in lipid synthesis, detoxification and calcium storage, among other things.
Overal the ER has a wide range of functions including biosyntesis, metabolism and
storage. Like several other organelles the ER is connected to actin filaments which
makes movement of the ER possible. [13,14]
1.6.4 The Golgi apparatus
The Golgi apparatus consists of several sacks, each enclosed by a membrane. Its
functions include modification of protein marking, transport and secretion of
proteins and other molecules. Vesicles from the ER fuse together with the Golgi
apparatus, where the cargo’s final destination is decided by the Golgi. Either it is
sent to the area in the cell where it is required or secreted through exocytosis. The
reciving side of the Golgi is called the cis-golgi network, and is the side closest to
the ER. The opposite side is the trans-golgi network, where the proteins or
molecules are secreted in a vesicle formed by the Golgi membrane. [13,14]
1.6.5 Actin
Actin is a part of the cells cytoskeleton. It is necessary for the cell to have a
cytoskeleton to obtain its structure, and also for transporting organelles or vesicles
containing cargo molecules. Actin contributes to shaping the cell surface, as well as
cell movement. Actin makes movement of the Golgi and ER possible. The
transport process is mediated by the acto-myosin system, where actin filaments

13
represent the road or track the organelles move on, and myosin functions as a
motor protein, moving the organelles in a direction. [14]
Actin filaments are created by polymerization of actin monomers. Latrunculin B
forms complexes with actin monomers, resulting in inhibition of the
polymerization process. If the cell is exposed to Latrunculin B, movement of the
organelles will be inhibited. [15]
2 Materials and methods
2.1 Specimen preparation
An Arabidopsis thaliana plant was marked with fluorochrome YFP or GFP, in
order to visualize the golgi apparatus or the endoplasmatic reticulum.
A small leaf was removed from a young A. Thaliana plant. The leaf was placed on
an object slide with a drop of MQ-water. Any dirt og particles, as well as air
bubbles was removed from the object slide, and the specimen was sealed using
wax.
2.2 Bead preparation
A solution of TetraSpeckTM
microspheres was vortexted for two minutes. Three
different samples were created by diluting the solution to the factors 1:100, 1:1000
and 1:10 000. The purpose of diluting the solution was to avoid cluttering of the
beads, in order for visualization of a single bead to be possible.
2.3 Microscope
The type of microscope used was a Leica SP5, along with the Leica Application
Suite Advanced Fluorescence (LAS AF) software. An argon laser was used, with an
intensity of 30 %. The chosen wavelength of 514 nm light was set to 15 %. The bit
depth used was 12 bits, and the pinholde aperture was set to 1 airy unit (AU). One
photo-multiplier tube (PMT) detector was set to detec light with wavelengths in the
range between 520 and 570 nm.
The overview images and the RGB images of the ER and Golgi were taken with a
10x/0.40 air-immersion objective.
2.4 Image processing and software
The images of the specimens containing golgi and ER-marked cells were scanned.
The images were then analysed and applied a LUT of chosen colour in the
computer software ImageJ. The gain and offset values of the images was adjusted
to make the structures of interest clear. Scalebars and LUT colour bar, indicating
light intensity was added to the overview images.
Different ROIs were selected for the movies, resulting in the RGB images, of the
ER and Golgi apparatus. The acquisition mode of the microscope software was set
to xyt, with a frame rate of 1 frame per second.

14
Frame number 1, 6 and 11 were chosen in both movies to illustrate movement of
the organelles in the RGB images. The RGB images were processed in the software
ImageJ. The first frame is red, the second green and the third blue. Merging the
pictures of the three frames create an illustratin of organelle movement in the cells.
Since the colours red, green and blue create white when they overlap, white areas
represent those areas where no movement was detected. The organelles that moved
will have either a red, blue or green colour, or a mixture of two of the colours.
2.5 Optimal resolution
The z-stacks of the Golgi apparatus and the fluorescent beads was visualized using
a 63x/1.20 water immersion objective. A PMT detector was used for the z-stacks
of the golgi apparatuses. For the imaging of the beads a hybrid detector (HYD) in
photon counting mode was used. Acquisition mode was set to xyz.
Latrunculin B was used on the specimens containing the Golgi apparatuses to
inhibit movement. To be sure that there would be no movement in the specimen,
one should wait one hour to let the inhibitor work. After one hour the specimen
was scanned for ROI.
The Rayleigh criterion was applied. The Dxy and Dy was found, and the Nyquist
theorem was applied with oversamling by a factor of three. The optimal pixel size
was set by adjusting the zoom factor and the number of pixels in the image
(Equation 6). The LUT was adjusted so that the pixels of interest were neither fully
saturated or at zero intensity.
When performing optical cross sections in the z-direction, the start point was set to
a plane above the chosen Golgi apparatus, where there was zero light intensity. The
same was done with the end point; it was set to a plane well below the Golgi
apparatus where the structure was not visible and there was zero light intensity.
This is done to ensure that the whole golgi stack was depicted in the z-region. If
the regions above or below the Golgi was too large, one can crop the image using
an image processing software. The distance between the scans in the z-dimensions
was set so that it corresponded with the optimal voxel depth calculated. The stack
was visualized using image processing software.
The settings listed above was also used for the fluorescent beads, except for the
zoom factor, which was set to a higher value. This resulted in greater oversampling
in this stack, compared to the stack of the Golgi apparatus.
A ROI was chosen based on the requirement of the visualization of three beads to
be possible in the chosen aera. The beads also had to be distinguishable from one
another. Here, the start and end point was also chosen were there was zero
intensity above and below the chosen beads.

15
2.6 PSF analysis
The z-stacks of the fluorescent beads were analysed in the software Amira. The
three chosen beads were marked, and a BeadExtract module was attached to the z-
stack module. The BeadExtract module estimates the PSF size based on the
average of the three marked beads. The PSF was then visualized using a Projection
View module.
The PSF was opened in ImageJ, where the MetroloJ plugin was used to calculate
size at FWHM of the PSF.
2.7 Deconvolution
The computer software Amira was used for the analysis of the z-stacks of the golgi
apparatuses. The PSF module from the beads’ PSF analysis were resampled in
order to have an sampling rate equal to the z-stack of the Golgi apparatuses. This
was done by connecting the two modules to a resample module (Figure 7). A
Deconvolution module was connected to the Golgi z-stack module and the
resampled PSF module. The correct NA and 𝜆!"#
values, as well as refraction
index, was entered in the deconvolution module. Deconvolution by maximum
likelihood estimation was startet, and 20 iterations were run.
Figure 7: Module window in Amira. The modules containing data is marked green,
the operation modules are coloured red and the yellow boxes show visualisation
modules. The operation modules process the raw data in order to produce
resampled and deconvolved images.
3 Results
3.1 Image analysis of the golgi apparatus
An image of a cell marked with EYFP of the golgi apparatus is presented as well as
a RGB-image to illustrate movement of the organelles (Figure 8). The RGB-images
were created by filming the movement of the organelles and choosing three
different frames, numer 1, 6 and 11, where number one is red, 6 is green and 11 is
blue. The three frames were overlayed, creating the images shown below. Frame

16
number 1, 6 and 11 were used both for the RGB-image of the Golgi apparatuses
and the ER.
Figure 8. (To the left) An overview of the Golgi Apparatus marked with EYFP in
plant cells of A. Thaliana. The image was taken using a 10x/ 0,4 air immersion
objective. The colour scale of the LUT to the right indicates the light intensity
raging from zero, black, to fullt saturated pixels, shown in white. (To the right). A
RGB-image to illustrate movement of the golgi apparatuses in the cell.
3.2 Image analysis of the endoplasmatic reticulum
The images presented are an overview image of cells with an EYFP-marked ER,
next to a RGB-image illustrating the movement of the organelle within the cell
(Figure 6). The procedure for creating the RGB-image was the same as the one for
the Golgi apparatus.

17
Figure 9: (To the left) An overview of the ER marked with EYFP in plant cells of
A. Thaliana. The image was taken using a 10x/ 0,4 air immersion objective. The
colour scale of the LUT to the right indicates the light intensity raging from zero,
black, to fullt saturated pixels, shown in white. (To the right). A RGB-image to
illustrate movement of the ER in the cell.
3.3 Deconvolution of the Golgi z-stack
The software Amira was used to obtain a deconvolved image of the golgi stacks.
This was done to remove blur from the image, in order to observe the golgi stacks
in 3D with better resolution.
Figure 10. The golgi z-stacks before deconvolution (left) and after (right). The
images were created in the Amira image processing software.
3.4 Size analysis of the Golgi z-stacks
The deconvolved z-stacks of the golgi apparatus were analysed so that the size
measurements of a single stack could be contucted. The software ImageJ was used
with the MetroloJ plugin. This software interprets the raw data and produces a
gaussian line fit based on the data. The line fit makes it possible to measure
FWHM. Three different golgi stacks were analysed, and the average size was
calculated (Table 2)
Table 2: FWHM size of the three Golgi stacks, and average size based on these
data. The asterisk denotes a possibly inaccurate value, as the fittet curve did not
incorporate all the data points.
Dimension Golgi 1 [nm] Golgi 2 [nm] Golgi 3[nm] Average size [nm]
x 260 447 266 324,33
y 302 502 250 351,33
z 595 617 809* 673,67

18
4 Discussion
4.1 Image analysis of the Golgi apparatus and ER
The images clearly illustrated the cells structure and organization of the organelles
under study. The cell membrane is clearly visible, making it possible to distinguish
between to different cells. This is probably a consequence of the vacuole pushing
the organelles towards the membrane. Individual organelles can be observed in the
cell, and their shape and size can be used for further analysis.
In the overview image of the Golgi apparatuses, the stomata is clearly visible. This
shows that the cell is one of the cells at the top layers of the leaf. The Golgi
apparatuses are clustered together in the stomata guard cells. The overview image
also shows that the Golgi apparatus is quite abuntant in cells. Golgi is a very
important organelle in the cell because of its role in sorting of proteins, among
other things.
The overview image of the ER shows a clear network structure. Its long tubular
structures creates a large surface area, suggesting that a large surface area is an
important tool to secure effiency of the processes occuring in the organelle. The
ER seems to be tightly packed near the cell membrane, as the light intensity here is
quite high. The underlying network is shown in blue.
The RGB images is able illustrate movement of the organelles in the cells quite
well. In both the RGB images of the ER and the Golgi apparatuses it shows
significant movement. Also, it shows areas where there was not detected any
movement, coloured white. It is possible to observe certain areas that are
particularly active in both the ER- and the Golgi image. These active areas can be
seen as multicoloured strands in the cell. In other areas the movement of the
organelles seem more random than directional.
4.2 Optimal resolution
The calculated values for the resolution limit, pixel and voxel size were enterted in
the Leica softwar for optimal sampling rate. The software automaticly rounded of
the numbers, creating a possible source of error. Because of this, the calculations
were performed with oversampling.
4.3 PSF analysis
The PSF from the bead extract could be seen as a dot surrounded by alternating
light and dark rings. This fits well with the theoretical knowlegde of the PSF. This
suggests that the z-stacks of the beads were accurate.

19
A 3D model of the PSF was generated using computer software, making it possible
to view the intensity of the PSF in the z-dimension. Through this model the airy
pattern was clearly visible (Figure 8).
Figure 11: Three different cross sections in the z-dimension of the PSF, showing
different light intensities through as a pillar in the chosen plane. The airy patterns,
or rings, are clearly visible in the PSF. The PSF was created by an NA value of 1.4
and the 𝜆!"#
value was created by Amira. The values used to create this model was
not from the results.
4.4 Deconvolution of the golgi z-stack
The deconvolved image was clearly an improvement from the raw image, as the
organelles are easier to see and most of the blur was removed from the image.
Deconvolution has shown to be a useful tool when it comes to live cell imaging.
However, the process did not work perfectly. This could be due to a deviation
between the PSF of the golgi z-stacks and the PSF from the beads. Nontheless, the
abberation seems to not have been very significant, as the deconvolved image does
not show any obvious flaws and is not missing any crucial information. It seems as
though the utmost of blur has been removed, and thus creating a better depiction
of the specimen.
4.5 Size analysis of the Golgi stacks
The FWHM size for three different golgi stacks were calculated using the MetroloJ
plugin. Gaussian line curves where fitted to the light intensity values of the z-
stacks. In the analysis of Golgi 3 the gaussin fit did not include all the data points in
the z dimension, resulting in a possibly inaccurate R2
-value. The R2
value is a
measure on how well the fitted line corresponds with the data points. A perfect line
will have a R2
value of 1.

20
There were differences between the three chosen Golgi apparatuses in size
dimensions, suggesting that the organelle is not symetrical in shape. This may be
due to the fact that the Golgi apparatus consists of several stacks, or cisternae,
which can have different shapes and sizes. The data suggests that all three Golgi
apparatuses are largest in the z-dimension. The minimum resolution was larger in
the z-dimension, resulting in inaccurate values as the smallest unit will become
larger.
The data collected from the experiment regarding size of the Golgi apparatus is not
representable. For one thing, only one cell type was analysed, and the sample size
was also quite low. To achieve a more representable measurement of the Golgi
more cell types should be analysed, to remove any inaccuracies.

21
5 Litterature
1. Corle, Gordon. Confocal Scanning Optical Microscopy and Related Imaging
Systems [0-12-408750-7;9786611046699] (1996).
2. Nathan S. Claxton, Thomas J. Fellers, and Michael W. Davidson. Laser
Scanning Confocal Microscopy.
3. Leica TCS-SP5 System Specifications. URL:
http://www.ibmb.csic.es/filesusers/Leica%20SP5_specifications(2).pdf
[Access date: 25.03.2014]
4. N. S. Claxton, T. J. Fellers, and M. W. Davidson. Laser scanning confocal
microscopy. 2005.
5. Y. Wang, J. Y.-J. Shyy, and S. Chien. Fluorescence proteins, live-cell imaging,
and mechanobiology: Seeing is believing. Annual Review of Biomedical
Engineering, 10:1–38, 2008.
6. K. F. Sullivan and S. A. Kay, editors. Green Fluorescent Proteins. Academic
Press, 1999.
7. J. W. Lichtman and J.-A. Conchello. Fluorescence microscopy. Nature
Methods, 2(12):910–919, 2005
8. R. W. Cole, T. Jinadasa, and C. M. Brown. Measuring and interpreting point
spread functions to determine confocal microscope resolution and ensure
quality control. Nature Protocols, 6(12):1929–1941, 2011.
9. W. Wallace, L. H. Schaefer, and J. R. Swedlow. A workingperson’s guide to
deconvolution in light microscopy. BioTechniques, 31:1076–1097, 2001.
10. E. B. Online. microscope.
Retrieved from:
http://www.britannica.com/EBchecked/topic/380582/microscope
[access date: 29.03.2014]
11. C. Press. CRC Handbook of Chemistry and Physics. Taylor and Francis
Group, LLC, 2013.
12. D. L. Taylor and Y.-L. Wang, editors. Methods in Cell Biology. Academic
Press, 1989. p. 167.
13. Bruce Alberts, Alexander Johnson, Julian Lewis, martin Raff, Keith Roberts
and Peter Walter. Molecular Biology of the Cell . Garkand Science, 5.
Edition (2008).

22
14. J. Hardin et al. Becker´s world of the cell. (8. Utg, 2012). San Fransisco: Pearson.
15. A. Kurenda, P. M, Pieczywek, A. Admiak, A. Zdunek. Effect of Cytochalasin B,
Lantrunculin B, Colchicine, Cycloheximid, Dimethyl sulfoxide and Ion Channel
Inhibitors on Biospeckle Activity in Apple Tissue. Department of Microstructure
and Mechanichs of Biomaterials, Institute of Agrophysics, Polish academy of
Sciences. Published by: Springer. (2013-12-01).
5.1 Illustrations
1. Cover photo: Plant seed. URL: http://wodumedia.com/wp-content/uploads/Plant-seed-from-
freshwater-pond-near-Moscow-Russia.-Photographed-with-fluorescence-10x-objective.--
960x1531.jpg
2. Illustration of the pin hole’s function in a confocal microscope
URL: http://www.photonic-lattice.com/en/technology/polarization-longitudinal-
slit-technology/
3. The corrolation between an airy pattern and a point spread function. The graph (red) shows the
PSF as light intensity, with the airy pattern in the background. URL:
http://www.asinen.org/

23
7 Appendix
A Calculations
A.1 Resolution limit, pixel size and number of pixels needed
Calculations of the resolution limit was performed using equations (2) and (3). The objective used
was a 63x/1.2 water immersion objective, which gives an NA value of 1.2 and 𝜂 = 1.33. The
excitation wavelength used was 514 nm.
Lateral resolution limit:
𝐷!" =
0.61 𝑥 514 𝑛𝑚
1.2
= 261,3 𝑛𝑚
Axial resolution limit:
𝐷! =
2(514 𝑛𝑚 𝑥 1.33)
(1.2)!
= 951,6 𝑛𝑚
Optimal pixel- and voxel size was calculated using equations (4) and (5):
Lateral pixel size:
𝑃𝑠!" =
261,3 𝑛𝑚
3
= 87,1 𝑛𝑚
Axial voxel depth:
𝑃𝑠! =
951,6 𝑛𝑚
3
= 317,2 𝑛𝑚

24
B Metrolo J analysis
B.1 PSF analysis of fluorescent beads
The gaussian line fit graphs for the PSF analysis of the fluorescent beads are shown (Figure B1).
(a) (b) (c)
Figure B1: Gaussian line fit in the x (a), y (b) and z (c) dimensions.
The R2
value is presented (Table B1).
Table B1. R2
values
Gaussian curve R2
a 0.99
b 0.99
c 0.98
B.2 Size analysis of the Golgi stacks
Gaussian line fits for Golgi apparatus number 1,2 and 3 are presented (Figure B2, B3, B4)
(a) (b) (c)
Figure B2: Gaussian line fits in the x (a), y (b) and z (c) dimensions, of Golgi apparatus 1.
The R2
value is presented (Table B2).
Table B2. R2
values
Gaussian curve R2
a 0.99
b 0.97
c 0.99

25
(a) (b) (c)
Figure B3. Gaussian line fits in the x (a), y (b) and z (c) dimensions, of Golgi apparatus 2.
R2
values are presented (Tabel B3)
Table B3. R2
values
Gaussian curve R2
a 0.98
b 0.98
c 0.99
(a) (b) (c)
Figure B4. Gaussian line fits in the x (a), y (b) and z (c) dimensions, of Golgi apparatus 3.
R2
values are presented (Tabel B4)
Table B4. R2
values
Gaussian curve R2
a 0.99
b 0.99
c 0.99

Kalkvik, Project Assignment

More Related Content

What's hot (15)

Viewers also liked (13)

Similar to Kalkvik, Project Assignment (20)

Kalkvik, Project Assignment