Why does RAPD mapping work?

frist at ccu.umanitoba.ca frist at ccu.umanitoba.ca
Sat Mar 30 16:34:18 EST 1991

The latest 'hot' technique making the rounds is RAPD mapping. For those of 
you not familiar with the technique, it basically involves the use of short
PCR primers to amplify genomic DNA, generating genotype-specific patterns of
bands, which segregate just as traditional RFLP markers do. For a mapping
project, the investigator will typically screen a hundred or more primers of
random design, choosing primers that give a manageable number of bands 
(eg. 5 or 6), and of those primers, choosing as informative ones  those that
detect polymorphism (ie. presence vs. absence of a band) in the population. 
Linkage is established as with RFLPs. RAPD (pronounced 'rapid') mapping is
described in:

Williams, J.G.K et al. (1990) DNA polymorphisms amplified by arbitrary primers
are useful as genetic markers. NUCL. ACIDS RES. 18:6531-6535.

Here's my question. At no point in the paper do the authors even hint that
they have restricted their genomic DNA. Yet, they get reproducible, character-
istic band patterns for each different primer. Note that only 1 (count 'em) one
primer is used in each reaction. Theoretically, elongating chains for a given 
primer should stop within a range of distances from the starting point, rather
than at discrete sites, UNLESS the DNA has been restricted, in which case all
chain elongation should stop at the first RE site distal to the primer.

The authors don't even ATTEMPT to explain why RAPD mapping works. Is there
any experimental evidence that will shed some light on this? Or is it that
they simply forgot to mention that they restrict their DNA?

Another problem: If RAPD mapping is simply doing a linear amplification, 
rather than a true polymerase chain reaction, which is exponential, I am a 
little skeptical that you would see bands in EtBr staining, even after 45
cycles (ie. making 45 copies of the starting material)

The only way I can see this working is if the primers represented inverted
repeats that were close enough together to allow both strands to amplify. 
Since the average primer length is 10nt, a template site should occur every
4e10 nt (~10e6nt), which means perhaps a thousand or more sites in a typical
eukaryotic genome. But given one site, the probability of a second site 
occuring is also 4e10, so the average distance between the two sites should 
be 10e6nt. Since the bands seen by the authors were typically in the 0.5 to 
3.0 kb range, we can exclude the idea of a distal perfect 10-mer match 
in the opposite orientation occurring within this distance.  

Perhaps the best explanation is that the second site does not have to be 
a perfect match, but need only match several (eg. 6) nucleotides at the
3' end of the primer. Exactly such a phenomenon appears to occur in 'primer
dimer' formation, in which some primers can prime themselves by pairing at
their 3'ends, creating two 3'recessed ends. This is well documented.
If imperfect 2nd strand priming is the explanation for RAPD mapping, then
a suitable downstream site would occur within a few thousand bp. Of equal
importance, this mechanism would also predict that BOTH strands would
amplify, resulting in exponential, rather than linear growth.

Anyone care to comment?

Brian Fristensky                |  
Department of Plant Science     | Can you say 
University of Manitoba          |
frist at ccu.umanitoba.ca          |     CHICKEN UBIQUITIN, CHICKEN UBIQUITIN,
Office phone:   204-474-6085    |     CHICKEN UBIQUITIN, CHICKEN UBIQUITIN
FAX:            204-275-5128    |     CHICKEN UBIQUITIN, CHICKEN UBIQUITIN...?

More information about the Methods mailing list