The Case of the Vanishing Intron

[Thierry Nouspikel]

Hello netters,

Can anybody explain the folowing mystery ?

Background
=========
- When studying my favorite gene in a patient, I discovered an aberrant 
  plicing phenomenon, due to a cryptic acceptor site, 55nt downstream of the 
  genuine one.
- When PCRing cDNA in that region of the gene (RT-PCR), I got the expected 
  290bp band  plus a non-stoichiometric amount of shorter (misspliced) product 
  235bp long.
- When PCRing genomic DNA , I got a single 605bp band.
  Sequencing revealled there was a 315bp intron in it, with a poor acceptor 
  site, explaining why a cryptic donor site is occasionnaly used.
 
No problems so far...

Puzzling observation
=================
I recently repeated this analysis with two new patients. In both cases RT-PCR 
yielded a 290bp band and a small amount of 235bp misspliced product (as 
expected).

BUT with genomic PCR I got:
- Two bands for patient #1: the expected 605bp band and ~290bp band !!
- A single short band (~290bp) for patient #2 !!!

Question
=======
Where did the intron go ??????

Tentative answers
===============
- The intron is absent in one allele of patient #1, and in both alleles of #2.
BUT, if there is no intron why do I get aberrant splicing in patient #2 ?

- My PCR reaction has been contaminated
  by RNA (but it wouldn't serve as a template without previous RT, would it ?)
  by cDNA from the parallele RT-PCR reaction.
BUT this does not explain why I don't see the regular 605bp genomic band in #2.

- One of the PCR primer can anneal at a wrong site and this just happens to 
yield a product with the same size than the cDNA product.
BUT, not very likely, and why does this happens a) always in #2, b) half the 
time in #1, c)  never in the original patient ?

I'm sure there is an obvious solution to this mystery, I just can't figure it 
out. If you have an explanation, please let me know. Any suggestion is 
welcomed.

Many thanks
                                            Thierry Nouspikel

Thierry Nouspikel, MD                 ¦ "Un technocrate c'est un mec,
Genetics & Microbiology Dpmt          ¦ tu lui poses une question,
University of Geneva, Switzerland     ¦ quand il a fini de répondre
nouspike at cmu.unige.ch                 ¦ tu comprends plus ta question."
fax:41/22/329.34.04                   ¦ Michel Colucci, dit "Coluche"