Dr. Atchison
9.19.96
Lecture 14

knockout of FosB transcription factor gene causes a nurturing defect - 
mom won't care for neonates any more. So, losing proteins can affect 
behavior as well. Not that this is why human moms kill their kids or 
anything....

in Genetic Engineering handout is list of genes which can be looked at, 
screened for, etc, and associated diseases. Also, chart of genetic tests 
and diseases to screen for in apparently normal people.

We did talk before about the Southern Blot, used for DNA analysis / 
probing DNA with radiolabelled probe.

In 1975, the Northern Blot was developed. The material run on the gel in 
this test is RNA instead of DNA - otherwise identical.

Western Blot - protein is run on the gel and detected w/Ab probe.

Now, you can look at DNA from an individual w/o cloning it, but it has to 
be homogenous. if you are looking for a mutation, ALL the cells from 
which you isolate the DNA must have the mutation to detect it by southern 
blot. If only 1 cell per thousand or whatever have the mutation, eg in 
leukemia, the southern blot won't detect it. 

BUT there is a new method that will detect that one in a million cell! 
PCR- polymerase chain reaction, developed in CA in 1984 by an LSD using, 
surfing, nobel prize winner.

see diagram. You take a strand of DNA, and you can amplify it by taking 
one strand, adding an oligonucleotide of the complementary strand and 
adding polymerase so it copies the rest of the strand, etc etc etc - can 
amplify DNA greatly.

So, in the leukemia patient, you often have translocations...eg, between 
chromosomes one and 19. You can try to amplify this sequence in a PCR to 
test for this problem. 

You will only get amplification of a specific band, because your 
oligonucleotides are specific and directional, so a normal chromosome 
won't give you the amplification.

So, you do a PCR with the appropriate oligonucleotides and you can see if 
the translocation is present. You can monitor tx this way, as well as 
diagnosing the patient this way. In most cases you don't totally 
eliminate the disease causing cells. Even in patients appearing to be in 
complete remission, you can still usually detect some of this.

Diseases caused by trinucleotide repeats can also be detected this way. 
You can use PCR by putting the oligonucleotides on either side of the 
repeat sequence, so you amplify the sequence containing the trinucleotide 
repeat. You know what size of product to expect...if abnormally high 
repeats, you will see an abnormally large band on the gel.

This is also a good way to test for viruses, eg HIV, FIV, etc. You can 
see if a virus exists within a body - any virus you know the sequence of, 
you can try to isolate if from a patients DNA. 

Now, if you have a point mutation, that doesn't affect a restriction site 
(so can't use southern blot), no good RHLP sites flanking it, what do you 
do? Well, you can amplify the DNA using PCR, and then sequence the DNA. 
Most common sequencing method is the Sanger dideoxy method. This method 
works by using mixtures of various nucleotides, DNA polymerase. DNA 
polymerase needs a primer, remember. So, you set it up so that all the A 
bases, for example, have no 3' OH - eg, dideoxy A. So, whereever an A is, 
in the chain, synthesis will STOP, because it can't add on when there is 
no hydroxyl. So, you get chains of DNA all stopping whenever there is an 
A. Then, do it again, w/dideoxy G, then dideoxy C, then dideoxy T. Now in 
the A tube, chain stops at every A, in C tube, stops at every C, etc etc, 


separate these out by size on a gel (you get one nucleotide resolution) 
and look where the bands stop. So, you can sequence it in this manner. 
Very simple, really.

The above are all mainly diagnostic techniques, but biotechnology can 
also be used for treatment.

anti sense technology.

say you have an mRNA:

5'cap-------------[*]----------------------AAAAA

* = single mutation

for a protein called Ras, which is involved in cellular signal 
transduction, in the cell pathway controlling growth. We'll meet it again 
when we talk growth factors. IT is commonly mutated in a single base pair 
change in cancers. Very subtle mutation which directly contributes to 
cancer. How could you try to get rid of this RNA coding for this mutant 
protein?

Well, you make complementary oligonucleotide of complementary DNA, and it 
will PREVENT translation of the affected RNA, because it will base-pair 
with the affected portion of the RNA. In a dish, this works really really 
well - you put in the oligonucleotide and the cells change to normal . In 
vivo, well, you can give them cancer by inducing a Ras mutation- eg K-Ras 
contributes to lung cancer, and you can give mice cancer by inducing this 
mutation, and then inject those mice with a complementary oligonucleotide 
and cure 80% of the mice. But, the oligonucleotide might be degraded 
before it reaches the site. It needs to be in pretty high concentrations, 
which is hard, because the kidney tries to filter it out, and you have to 
continuously give it until the cure is effected. After cure, you don't 
have to give it  - it seems that affected cells are deleted via 
apoptosis.

Other approaches at trying to cure disease - gene therapy - the first 
gene therapy at penn was done here at the vet school. There are many 
disease being approached this way - basically, to replace affected gene 
with normal gene in sick animal. Hasn't worked in people yet. There are 
some animals that have gotten "better" from treatment, but this was 
transient. It seems that the inserted genes tend to "shut off", very 
disappointingly.

Drawbacks/issues with biotechnology:

ethical issues...these technologies are very powerful. you can learn a 
lot about an individual and a fetus in this way. you can take one cell 
from a blastocyst and use PCR to check for affected gene. Is that ok to 
do to screen IVF for a particular embryo free of disease vs disease 
carriers? Some find this to be a very thorny issue. No one is doing this 
in people though.

what diseases should be screened for in this way? Problems are enormous. 


For example, Huntingdon's disease - late onset (about age 40 or 50) 
progressive neurological disease. Causes dementia and loss of motor 
control. It's a dominant disease, too. If you are heterozygous for it, 
you have it. [these issues aren't test material, btw]. So, this is a late 
onset dz. What if one of your parents had it? You have a 50% shot at 
having it. Should you screen your fetus for it? IF it comes up 
positive......then you have to live with it....do you want to know if you 
have it? How old does a person have to be before it's ok to tell the 
person they have it? Also, insurance companies can charge more if you 
have a genetic disease.  Ethically, practicioners will not screen for 
this prenatally, since you get about 45 yrs of good life out of it.

Someone was researching albinism, and found that one woman was planning 
to abort her albino fetus! the researcher got very upset and didn't 
release the test result to her.

Prenatal screens are done for cystic fibrosis, muscular dystrophy, 
fragile x, downs syndrome.

back to preB leukemia translocation.
there is a crossover between chromosomes 1 and 19, which disrupts 2 genes 
- the E 2 a gene, which is involved in B cell developement, and a 
chimeric protein is made, which is half one factor, and half another - it 
has the activation domain of one factor, and the binding domain of 
another, and it aberrantly turns on genes related to growth control. Lots 
of translocations invovle transcription factors.