Tuesday 30 October 2007

What I learned in school today Oct 30

Today's lecture was a guest lecture concerning viruses and their role in cancer, particularly how they can cause cancer and can be used to treat cancer.

What is cancer?

Cancer, in short, is an uncontrolled proliferation of cells. It is the 2nd leading cause of death in developed nations - 6 million people worldwide die from cancer every year. Three main therapies are used to treat it: radiation, chemotherapy and surgery. The first two are specific in that they target cells that are rapidly dividing. This may be problematic in that tumor cells are not the only cells that divide rapidly; certain bone cells, for example, quickly divide and may be targeted by these treatments.

The process whereby a normal cell becomes cancerous is called transformation. A transformation may be a change in the morphological, biochemical or growth patterns of a cell. Cells which are transformed become immortal, that is, they can grow and divide indefinately. Normal cells are subject to what is called the Hayflick Limit, at which a cell will stop dividing (this occurs when the telomeres are shortened to a length below which becomes deleterious to the cell). Cells which are transformed may have one or more altered phenotypes, such as a loss of anchorage dependance (normal cells will not grow unless attached to the extracellular matrix), loss of contact inhibition (normal cells will stop growing when they fill the available space and contact one another), a decreased requirement for growth factors (i.e. they will continue to grow even when not supplied with the signal to grow), or gross morphological changes.

Both RNA viruses and DNA viruses may transform cells into cancer cells. There are three ways in which viruses can do this: by expressing a modified version of a cellular proto-oncogene, by cis-activation of proto-oncogenes or trans-activation of proto-oncogenes.

[Proto-oncogenes are genes that, when mutated, result in the formation of tumors]


Expression of Modified Proto-oncogenes

An example of this method of cancer induction comes from the Rous Sarcoma Virus, the first oncogenic virus studied. This virus contains a gene called v-src. It is very homologous to a cellular gene called c-src (v=viral, c=cellular). The c-src gene product normally encodes for a tyrosine kinase. It contains two domains at one end of the protein (in the kinase domain), the SH3 and SH2 domains. The SH2 domain is of particular importance; it binds to phosphotyrosine residues. At the opposite end of the c-src protein lies a regulatory domain. The 527th amino acid in the protein lies in this region, and is a tyrosine. This tyrosine gets phosphorylated, and the SH2 region of the kinase domain binds to it. This results in an inactive protein. The usual fuction of c-src is to work in a signaling cascade to turn on transcription factors that will express certain genes involved in the cell cycle; when it is on, cells grow and divide. When inactive, these genes do not get expressed, and no cell growth.

The v-src gene, however, encodes for a truncated protein. It is cut off at the 526th amino acid residue. That is, it does not contain the Tyrosine residue required to keep the enzyme inactive. When expressed in a host cell, it is always turned on, resulting in contant expression of the downstream genes, and uncontrolled cellular growth and division - in other words, tumor formation.

Cis-activation of proto-oncogenes

This method involves the integration of a viral genome in to the host genome. Eukaryotic cells are often tightly regulated, with promotor sequences, and regulatory or enhancer elements controlling the expression of the gene. Insertion of the viral genome is often a random process, and in the event that the genome were to insert into the repressor element of a gene, then the repressor would become nonfunctional. This would lead to an inability to turn the downstream gene off when it is not needed. The gene would be constitutively expressed. If this gene is a proto-oncogene, then the result is likely to be the formation of a tumor.
[This is called CIS-activation because the viral agent that is causing the activation is in cis, that is, is on the same chromosome, as the gene it is infecting. "Cis" comes from "cistron", the old name for genes.]

Trans-activation of proto-oncogenes

This method involves a viral protein directly acting on a transcription factor, which causes proto-oncogenes to be upregulated. The resulting upregulation can lead to increased or unconrtolable cellular growth.

Examples

An example of how viruses can cause cancer can be seen in the Papillomavirus family of DNA viruses. There are over 106 types of Papillomavirus, and over 40 of these are able to infect the mucosal surfaces of the the genetail and aero-digestive tracts. HPV (Human Papillomavirus ) has been detected in some 99.7% of cervical cancers.

HPV has two primary viral oncoproteins: E6 and E7. They are involved in disregulating the cell cycle. During the cell cycle, cells go through a phase of DNA replication, known as S-phase. If there is damage to the DNA, or errors with improper cell division, replicating the DNA during S-phase can result in dire problems for the cell. Tumor supressor genes will stop the cell from completing S-phase if this is the case. The HPV E6 and E7 proteins block the action of these tumor supressor genes, allowing cells to go through S-phase, and thus DNA replication, growth and division, unregulated.

One tumor supressor that is affected by HPV is the Retinoblastoma protein (Rb). Rb binds a transcription factor called E2F, and keeps it from fuctioning. When E2F is needed, Rb is phosphorylated and E2F dissociates and is allowed to activate the transcription of genes involved in DNA synthesis, and consequently, progression through the cell cycle. The HPV E7 protein interferes with the E2F:Rb complex - E2F is constantly active, leading to unscheduled cellular DNA synthesis as well as viral DNA synthesis. Also, there is a "High Risk" form of E7 where Rb is not only bound but is targeted for proteosome-mediated degredation.

The cell has a safeguard against this, however. The p53 protein is translated under cellular stress or DNA damage, and it halts the proliferation of cells, and plays a role in apoptosis. This ensures that defective cells do not run the risk of becoming cancerous. This is where the viral E6 protein comes into play - it binds p53 and blocks it from fuctioning. Much like E7, E6 has a 'high risk' variant. This form of E6 recruits an E2 ubiquitin ligase which polyubiquinates the p53, marking it for degradation via the proteosome.

From the above, it can be seen that DNA and RNA viruses have deifferent roles in cancer. RNA viruses encode for viral concogenes, while DNA viruses have mechanisms evolved to counter cellular tumor-supressor genes.

Oncolytic viruses

Another area of cancer that viruses play a role is in cancer therapy. Oncolytic viruses are viruses that are used as a form of cancer therapy. In this method, live viruses are used to selectively lyse cancerous cells, while normal cells are left unaffected. The viruses infect tumor cells, replicate, lyse the cells and then spread to other uninfected tumor cells. For some viruses, this selectivity is not perfect and normal cells can be infected; however, the viruses will grow better in tumor cells than in normal cells and (hopefully) will pick tumor cells over normal ones. The ideal features of such a virus are that they cause only mild human diseases that are well characterized, have secondary inactivation mechanisms, do not damage normal cells, have low mutation and recombination frequencies, do not spread from host to host and cannot integrate into the host genome.

Once such oncolytic virus is Onyx 015, a modified version of Adenovirus. This variant lacks the adenovirus E1b55K and E1b19K proteins which block p53 and apoptosis. It is hypothesized that its E1a protein binds Rb and E2F leading to increased DNA replication of the viral genome, and lysis of the tumor cells. What keeps it from destroying normal cells is that this also activates p53, and p53 keeps the infected normal cells from spreading the virus. The infected tumor cells do not have p53 and thus are unable to stop lysis and spread of the virus to other tumor cells.

Another such virus is myxoma virus. This virus usually infects European rabbits and not humans. However, it has the ability to infect human tumor cells instead. The benefit of this virus over the Onyx 015 is that it is natural rather than engineered.

Monday 29 October 2007

What I learned in class today...

Biology 390



My professor was gone on a conference today, so this week's Biology 390 seminar was given by our TAs. We went over three concepts which I was already familiar with, but this did present some new facts that I did not previously know. These three concepts were DNA sequencing, Ampicilin resistance, and cDNA libraries.


DNA Sequencing


The method of DNA squencing we will be using in our lab was the method devised by Fred Sanger (which ultimately won him the nobel prize). It utilizes the principal of dideoxynucleotide chain termination. Dideoxynucleotides (ddNTPs) are much like regular dNTPs except for the fact that the are missing the 3' OH (thus making them DIdeoxynucleotides, see picture to the right). Since the 3' OH is necessary for the addition of the following nucleotide during DNA synthesis, if a ddNTP gets incorperated into a growing strand, the synthesis comes to a halt. These are added to a reaction along with template DNA (the DNA that you wish to find the sequence of), regular dNTPs, primers (specific to the regions flanking your template in whatever vector you chose) and DNA Polymerase.


The DNA polymerase used for this method is one that has no ddNTP discrimination, such as AmpliTaq FS - that is to say, it does not incorperate the ddNTPs with any preference. They are added to the growing strand at random. Thus, as one would expect, any reaction would contain multiple strands of varrying lengths - the ddNTPs are added at random, and the strand synthesis halts at random positions. If the reaction is left to run a sufficent amount of time, strands of every possible length are formed. For example, if your template read ACCTG, you would end up with strands that were:

A
AC
ACC
ACCT
ACCTG

The letters in bold are the incorperated ddNTPs.

Each ddNTP has attached to it two dyes, a donor dye and an acceptor dye. The donor dye, usually fluorescein, when excited by a laser, emits energy which then excites the accetor dye. The acceptor dye, usually rodamine, then fluoresces light of a particular wavelength. The acceptor dye on each specific ddNTP is different and fluoresces at a different wavelenghth - ddATP may have a dye which fluoresces at 565nm, while the dye attached to ddCTP may fluoresce at 615nm. Based on the wavelength emited, one can determine which ddNTP ended a specific fragment.

All the fragments produced by the sequencing reaction are passed through a capiliary in a sequencing machine. As in gel electrophoresis, the larger fragments run through faster than the smaller ones. The capiliary matrix must have a resolution that can discern between fragments differing by as little as one base pair. As the fragments pass through the capiliary, they pass by a laser, which excites the dyes attached to the ddNTPs. The largest fragment passes through first, and the wavelength it emits tells us what the last letter in the sequence is. The next largest is the second to pass through, and from it, we learn what the penultimate letter in the sequence is. This continues until the smallest fragment, representing the first letter in the sequence, is passed through the matrix.

A computer records the wavelengths that are emitted as they pass through, matches them up to their corresponding nucleotide, and arranges the results in the form of a chromatogram (aka electropherogram).

This is a readout of your exact sequence. Sometimes the computer is unable to determine what a particular base might be, and replaces it with an N on the readout. In this case it may be possible to look at the wavelength recorded and determine the identity of that particular base yourself.

This method of sequencing often works for fragments up to 800 bp in size. For fragments longer than this, sequencing in both directions from opposite ends of your fragment is best. For fragments less than 1.6kb, this may give overlapping fragments which can then be aligned to determine the complete sequence. For example, the pBluescript II SK+ plasmid which we are using in our lab contains two promoter sites, a T7 Promoter and a T3 Promoter. These face in opposite directions and flank the region where the template fragment has been inserted. Using these promotors, one can do two sequencing reactions, giving two sequences that can be overlapped at their ends to determine the complete sequence.


Ampicilin resistance

The topic of ampicilin resistance was brought up because it is used to select for bacterial colonies which contain a pBS II SK+ plasmid (since the plasmid confers Amp resistance to the bacteria). The gene that does this is refered to as the bla gene, which encodes the enzyme ß-lactamase. This enzyme breaks down the ß-lactam ring found in the structure of Ampicilin and other members of the Penicilin antibiotic family.

Ampicilin itself binds to a class of proteins known as Penicillin Binding Proteins (PBPs) which help in the biosynthesis of peptidoglycan, a component of the bacterial cell wall. By inhibiting the cell wall biosynthesis, the bacteria die. ß-lactamase thus ensures that the cell wall production is not inhibited by Ampicilin.

Unfortunately for molecular biologists, Ampicilin is broken down over time, so plates containing Ampicilin that have been unused for a long time cannot be used to select for Amp resistant strains.

Using Amp resistance is not without its problems. The ß-lactamase enzyme is secreted into the media surrounding resistant colonies. This means that the media immeadetly surrounding the restiant colonies is Amp free. Other non-resistant colonies can grow here. These are known as satellite colonies and will not contain any plasmids, and possibly messing up your experimental results. Another problem is the possible loss of plasmids. Since the media becomes Amp free, there is no selection for colonies that have a plasmid to grow there. Replicating the plasmid before cell division requires energy, and for cells gowing in the Amp free area, it is better for them to not have a plasmid at all. Without selection for keeping the plasmid, it is possible that it could be lost from the population alltogether.



cNDA Libraries

The final topic was that of the creation of cDNA, or Copy DNA, libraries. cDNA libraries are collections of DNA fragments that represent all of the actively transcribed genes in an organism. In other words, for every mRNA in an organism, there is made a DNA "copy" of that mRNA.

First, all the mRNA in a sample is collected. The mature mRNAs in eukaryotes have a poly(A) tail at the end of them after processing. This poly(A) tail acts as a binding site for a poly(T) primer. This primer has three sections: the poly(T) site, a XhoI restriction enzyme site and a run of "GA"s called the "GAGA" site:

TTTTTTTTTTTTTTTTGAGCTGAGAGAGAGAGA

The bold portion is the XhoI cutting site.

Next, the enzyme reverse transcriptase is used to transcribe backwards from the poly(T) primer to the 3' end of the mRNA template. This creates a double stranded duplex consisting of one strans mRNA and one strand DNA. This reaction uses a special form of cytosine, 5-methyl-dCTP. When added to the growing duplex, it prevents the cDNA from being chopped up by restriction enzymes.

Next, the RNA template is degraded by RNAse H, and the little pieces of RNA remaining are used as primers for DNA Polymerase I and PFU to replace the RNA strand with a DNA strand. This results in a double strand DNA duplex copied from the original mRNA template.

This cDNA needs to be able to be ligated into a vector, and an additional step is to transform the ends of the cDNA into different sticky ends. Adapters are added to each end which contain an EcoRI cut site. Next, XhoI is added, which cleaves at the XhoI site in the original primer (this is the reason for the 5-methyl-dCTP; the cDNA wont be cut up when XhoI is added). The end result is a cDNA fragment, with a EcoRI sticky end upstream and a XhoI sticky end downstream, allowing for it to be directionally cloned into a vector that has been cut with these enzymes.