Apoptosis and cancer development

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Apoptosis and cancer development

Apoptosis is a highly regulated process of programmed cell death present in multicellular organisms. It plays an important role during different stages of development and normal physiology [1]. Cells undergoing apoptosis can be characterised by cell shrinkage, plasma membrane blebbing, DNA fragmentation and chromatin condensation [2]. Inactivation of this process is central to the development of cancer. Besides enabling malignant transformations defects in apoptosis, cancer also may result in resistance to chemotherapies [3]. Therefore much research has been done to find a way to get around this resistance in order to improve the anti cancer therapies. It has provided the basis for novel targeted therapies that can induce death in cancer cells that include those targeting extrinsic as well as intrinsic pathways.

Proteasome Inhibition and Cancer therapy

Apoptosis is controlled by multiple regulatory pathways and their proteins. Examples include p53, the nuclear factor kappa B, the phosphatidylinositol 3 kinase pathway, and the ubiquitin/proteasome pathway [4]. In ubiquitin/proteasome pathway, proteins are tagged by ubiquitin and presented to the proteasome, where the protein is digested and ubiquitin is recycled in the cell [5]. The proteasome system is important in the degradation of damaged or unneeded proteins and serves as an important regulator of cellular processes, therefore it plays a considerable role in tumorigenesis [6]. The proteasome is found in all eukaryotic cells, both normal and cancerous and is highly conserved from yeast to man. It is a multicatalytic enzyme with an important impact on many regulatory pathways.

The inhibition of protein degradation through the ubiquitin-proteasome pathway is a recently developed approach to cancer treatment that extends the range of cellular targets for chemotherapy [7]. Bortezomib was the first proteasome inhibitor that was approved by US FDA in 2003 for treatment of refractory multiple myeloma [8]. Other inhibitors include natural products such as lactacystin, peptide aldehydes such as MG132, ALLN, and MG115, which are in the preclinical stages.


Phase contrast microscopy

Phase contrast microscopy is an optical microscopy contrast-enhancing technique that is used to produce high-contrast images of transparent specimens, such as living cells [9]. It employs a mechanism that translates minute variations in phase into corresponding changes in amplitude, which can be visualized as differences in image contrast. When light travels from one medium to another, it undergoes a change in amplitude and phase depending on the properties of that medium. These changes give rise to familiar absorption of light which gives rise to colors which is wavelength dependent. The human eye measures only the energy of light arriving on the retina, so changes in phase are not easily observed, yet often these changes in phase carry a large amount of information.

A phase contrast microscope does not require staining to view the slide and living cells can be examined in their natural state. This provides an insight into the dynamics of ongoing cellular processes which can be analyzed in high contrast with sharp clarity of minute specimen detail.

This technique is widely applied in biological and medical research for example it is used in diagnosis of tumor cells and the growth, dynamics, and behavior of a wide variety of living cells in culture.

Fluorescence microscopy

A fluorescence microscope is used to study properties of organic or inorganic substances using the phenomena of fluorescence and phosphorescence [10]. The basic function of a fluorescence microscope is to illuminate the specimen that is labeled with a fluorescence dye with a specific band of wavelengths, and then to separate the much weaker emitted fluorescence from the excitation light. As a result part of specimen marked with fluorescent dye will light up against the dark background [11].

In recent years fluorescence microscopy has gained immense importance in fields of biology and medical research due to advancement in technology of microscopy and development of fluorescent molecular probes. By using these fluorescent probes we can characterize subcellular structures, location of signaling proteins and indicators of physiological states. Green fluorescence protein has been frequently used as a reporter of expression. The presences of such fluorescent probes have triggered the development of fluorescence microscopy to visualize cells over time expressing specific proteins that have been fluorescently tagged. These probes can also be used in expressing the protein in small sets of specific cells. It allows to optically detect specific types of cells in vitro, or even in vivo.

Automated fluorescence microscopy and high performance computing have allowed the emergence of high content screening as a useful tool in the early stages of drug dicscovery [12]. Use of fluorescent proteins has provided novel insights into compound-induced responses in drug discovery. Since they are non-invasive, non-destructive and can be genetically-encoded, fluorescent proteins are attractive candidates for labeling drug target of interest based on live-cell analysis.

Time-lapse microscopy

Analyzing microscopic images and videos to extract useful information is becoming an increasingly important activity in many scientific laboratories [12]. Time-lapse microscopy imaging is broadly applied to record living cells over an extended period ranging from days to weeks. Recent advances in this field have made it possible to study different cell processes including apoptosis, cell division and cell migration [13]. This technique provides an insight into the nature of cellular functions which can be helpful in research areas like drug discovery, stem cell research, genomics and proteomics [14-15]. However, with the rapid increase in amount of data generated, our ability to interpret this information remains limited. Manual analysis of these massive data files takes weeks of tedious work, with the possibility of losing vital information stored in these images. Therefore, we need an automated and quantitative cell population monitoring system which facilitates the analysis of massive biological data.

Automated Image Analysis System

Difficulty in handling and analyzing large amounts of image dataset generated has urged the need for a sophisticated image analysis system. Existing tools for image analysis, such as NIH ImageJ and MetaMorph, are limited in their functionality to analyze high-throughput image analysis data. Commercial software are also available, developed by companies like Cellomics, Molecular devices and GE healthcare [16]. These software's in addition to being limited in their scope are expensive and come along the hardware which makes it impractical to test several programs for a new project. We cannot get any information about the algorithms used and make any modification in them due to its proprietary nature. Therefore to process high throughput data we need software having ability to correctly identify objects and analyze their size, texture, shape and intensity quantitatively. In order to extract meaningful measures from image we need efficient algorithms, which are used for feature extraction, cell segmentation, pattern recognition and statistical modeling [17]. To bridge up this technological gaps programmers are developing new algorithms that are fast and accurate in identifying objects and extracting their features.


CellProfiler (http://www.cellprofiler.org/) is an automated open-source image analysis software that can analyze thousands of images obtained through image acquisition instruments. The software contains already developed methods that are applicable to diverse assays [18]. It produces rapid, quantitative, and accurate results. Since the software is open-source it allows researchers to design and contribute new methods and improve the existing ones. CellProfiler contains advanced algorithms for image analysis that can accurately identify cell clumps and non mammalian cell types. It has a user friendly graphical user interface (Figure 1) with a modular flexible design allowing analysis of new assays and phenotypes. A pipeline is constructed based on individual modules that are placed in a sequential order. Each module processes the image and sends it to the next module in line. CellProfiler contains published and tested algorithms for object identification [19-23]. For each identified cell it can measure a large number of features including size, shape, intensity, texture and location. These measurements can either be viewed by CellProfiler's built in viewing and plotting data tools or they can be exported directly to excel or database (MySql or Oracle). System workflow is explained in figure 2.

Processed image

Identified objects

Measurements for every cell in every

image (location, size, shape, intensity,

texture) can be viewed by:

· CellProfiler data tools

· Exporting to spreadsheet

· Exporting to database

· Exporting to MATLAB

Identification of Apoptotic Cells

CellProfiler is used to study the effects of drug induced cell death using time-lapse microscopic images. These Images studied were obtained from a live cell imaging system (IncuCyte from Essen Biosciences, http://essenbioscience.com/) offering time-lapse microscopy. Apoptotic cells are identified from phase contrast images using their special features that distinguish them from normal cells.

Features of an apoptotic cells

The proposed method identifies apoptotic cells from observed images based on following features [24].

  • The edges contain much higher grey level as compared to central area
  • Their shape is quasi-circular form.
  • Their edges do not overlap with neighbouring cells.

Using these features a pipeline is setup in cellprofiler for accurate identification of apoptotic cells.

Aim of the study

The overarching goal of the research presented in this thesis was to develop, implement and validate algorithms, based on CellProfiler, that are able to extract information rich features of growing cell populations that can be used to quantify drug effects related to apoptosis and proteasome inhibition.

Materials and Methods


Following cell profiler modules were used during analysis of microscopic images.

Color to gray

This module is used for the conversion of RGB (Red, Green, Blue) color images to grayscale. There are two options Combine or Split. If we select Combine All channels will be merged into one grayscale image and if we select Split each channel will be extracted into a separate grayscale image.

Enhance edges

This module takes greyscale image as an input and finds the edges of objects producing a binary image where the edges are white and the background is black.

Several algorithms have been used to enhance edges.

  • Sobel Method uses the Sobel approximation to the derivative. It derives a horizontal and vertical gradient measure and returns the square-root of the sum of the two squared signals.
  • Prewitt Method Applies the Prewitt approximation to the derivative. Points where gradient of the image is maximum are returned as edges.
  • Roberts Method uses the Roberts approximation to the derivative. It looks for gradients in the diagonal and anti-diagonal directions and returns the square-root of the sum of the two squared signals. This method is fast, but it creates diagonal artifacts that may need to be removed by smoothing.
  • LoG Method applies a Laplacian of Gaussian filter to the image and finds zero crossings.
  • Canny Method looks for local maxima of the gradient of the image that is calculated using the derivative of a Gaussian filter. The method uses two thresholds to detect strong and weak edges, and includes the weak edges in the output only if they are connected to strong edges.

Identify primary object

This module is used for identification of primary objects (e.g. nuclei) in grayscale images that show bright objects on a dark background. It contains a modular three-step strategy for object identification.

  • In step 1 it is determined whether an object is an individual nucleus or two or more clumped nuclei.
  • In step 2 edges of objects are identified, using thresholding if the objects do not appear to touch and using more advanced options if the object is actually two or more nuclei that touch each other.
  • In step 3 identified objects are either discarded or merged together based on user defined rules. For example, if the objects are at the border of the image and are incomplete they can be discarded, and objects that do not lie in specified size limits can either be discarded or merged with nearby larger ones.

Thresholding methods

To distinguish between foreground and background pixels we need to set an intensity threshold. CellProfiler contain several methods to find intensity threshold automatically.


This method is used when percentage of image that is foreground vary from image to image. It takes into account the maximum and minimum values in the image and log-transforms the image prior to calculating the threshold.

Mixture of Gaussian (MoG):

This method assumes that the pixels in the image belong to either a background class or a foreground class, using an initial guess of the fraction of the image that is covered by foreground.


This method is used for images in which most of the image is background. It finds the mode of the histogram of the image, which is assumed to be image background, and selects a threshold at twice that value. This can be very helpful for images that vary in overall brightness but the objects of interest are always twice as bright as the background of the image.

Robust background:

This method trims 5% of brightest and dimmest pixels and assumes that the remaining pixels represent a Gaussian of intensity values that are mostly background pixels. It then calculates the mean and standard deviation of the remaining pixels and calculates the threshold as the mean + 2 times the standard deviation.


It is a very simple method and its result resembles that of Otsu. It starts by choosing an initial threshold and then iteratively calculates the next one by taking the mean of the average intensities of the background and foreground pixels determined by the first threshold, repeating this until the threshold converges.


This method computes the threshold of an image by log-transforming its values, then searching for the threshold that maximizes the sum of entropies of the foreground and background pixel values, when treated as separate distributions.

Measure object size shape

This module extracts area and shape features of each identified object for example area, form factor, solidity and orientation.

Measure object intensity

This module extracts intensity features for each identified object based on one or more corresponding grayscale images.

Measure object neighbor

This module determines how many neighbors each identified object has. We can specify which objects should be considered neighbours by providing a distance. If objects fall within this distance they are considered neighbours.

Classify object

This module classifies objects into a number of different bins according to the value of a measurement (e.g., by size, intensity, shape). It reports how many objects fall into each class as well as the percentage of objects that fall into each class.

Filter by object measurement

This module removes selected objects based on measurements produced by another module in pipeline. All objects that do not satisfy the specified parameters will be discarded.

Export to Spread Sheet

Measurements are converted to character-delimited text formats and saved to the hard drive in one or several files.

Export to Database

Measurements can be exported directly to a database or to a SQL-compatible format through this module.

Identifying apoptotic cells through phase contrast image

Cell culture and image Acquisition

HCT-116 human cell lines were treated with the drug doxorubicin (apoptosis inducer). Plates with fixed cells were analyzed using the IncuCyte live cell imaging system from Essen Bioscience. Each image is obtained after a duration of 1 hour (Figure 3).

Image analysis

Images were analyzed using CellProfiler 2.0. A pipeline was setup based on modules shown in table 1.

Table 1: Pipeline for identification of apoptotic cells

Module used

Parameters used

Load images

image format (tif)

Color to gray


Enhance edges

Edge finding method (Sobel)

Identify primary automatic

Size of object (15-40), thresholding method (background global), threshold correction factor (1.2)

Measure object intensity

Identified objects

Filter object

Category of measurement (intensity), Feature (mean intensity edge),Minimum and Maximum value required (0.6 - 1)

Measure object area shape

Filtered objects

Filter object

Category of measurement (AreaShape), Feature (form factor),Minimum and Maximum value required (0.8 - 1)

Measure neighbour

Filtered objects

Filter Objects

Category of measurement (neighbours), Feature (Percent Touching), Minimum and Maximum value required (0- 0)

Export to spreadsheet or database

Identifying apoptotic cells through fluorescence images

Cell culture

Mv4-11(Nacute myeloid leukemia cell line) was used to study the effect of drugs in apoptosis induction.


Six experiments were set up each with different combinations of apoptosis inducing drugs and caspase3 inhibitor (Table 2).

Table 2: Different experiments conducted






Etoposide + Caspase inhibitor




AKN + Caspase inhibitor


No drugs


No drugs + Caspase Inhibitor

Image analysis

Images were analyzed through cell profiler 2.0. The modules used in identification of fluorescent objects are shown in table 3.

Table 3:



Load Images

Image format (.avi)

Color to gray

Split (orig green was selected)

Identify primary object

Size of object (3-30), thresholding method (Robust background), threshold correction factor (1.1)

Measure object Shape Size

No adjustment required

Measure object intensity

No adjustment required

Proteasome inhibition

Cell line and image acquisition

The human melanoma cell line MelJuSo was used. These cell lines were treated with different concentrations of MG132 (0.1 µM, 1µM, 10µM) and were incubated for three days. Microscopic images and videos were obtained through IncuCyte Flr.

Image analysis

Images obtained are analyzed through cell profiler 2.0.



Load Images

Image format (.avi)

Color to gray

Split (orig green was selected)

Identify primary object

Size of object (3-30), thresholding method (Robust background), threshold correction factor (1.1)

Measure object Shape Size

No adjustment required

Measure object intensity

No adjustment required