Replies

Oh my. When I first read 5'-3' problem I figured you were talking about DNA duplication and the way length is continually lost from a DNA strand during successive duplications unless telomerase is there to fix it. But no. . .

You have two major flaws in your reasoning. The first I think is probably explained best by a diagram. Closest I can figure you think that you can't chop a piece of DNA out of a sequence, reverse it, and reinsert it because the directionality of DNA won't allow it. It will, the duplex that is excised is just flipped and reinserted so that a gene that was previously on strand 1 of the duplex is now on strand 2 (this you also assume is a problem when it isn't, but I'll handle that later).

So here's a DNA duplex that we're going to attach another one to the end of, we'll let this stand in for chromosome 2p:

5' ----->---->------>- 3'
3' -----<----<------<- 5'

The arrows are just to show the coding direction of the strand. So now here is our second duplex, which will stand for chromosome 2q:

5' -->-a--b->---- 3'
3' --<-------<---- 5'

The a--b indicates the start and stop codons for a gene. Now we could splice this onto the end of 2p as it is, but we want to reverse it. So let's do that:

3' ----<-b--a-<-- 5'
5' ---->------->-- 3'

Note that although when you're reading left to right the gene looks like it's backwards, when the ribosomes actually come along the duplex they will travel 5'->3' and thus the gene is still forwards to them. Now we can splice this on, but because of the way it's written we need to flop it front to bottom, which again changes none of the information present on this duplex.

5' ---->------->-- 3'
3' ----<-b--a-<-- 5'

Now the DNA segment is reversed and oriented properly for splicing onto the other chromosome.

5' ----->---->------>------->------>-- 3'
3' -----<----<------<------<-b--a-<-- 5'

Voila! Human chromosome 2!

The reversal of the second duplex doesn't affect insertion or splicing at all, and the information on both chromosomes is still intact. This is because although you seem to think that one strand of this duplex is coding, that is, has all of the working genes on it and the other one is noncoding junk, in reality both DNA strands of a duplex have genes on them. The terms "coding strand" and "noncoding strand" only have meaning so far as you are talking about a particular gene. Here--let's put another gene in and label the strands:

1: 5' ----->---->-1--2->------>------>-- 3'
2: 3' -----<----<------<------<-b--a-<-- 5'

So far as gene 1-2 is concerned, strand 1 is the coding strand and strand 2 is noncoding. So far as gene a-b is concerned, strand 2 is the coding strand and strand 1 is noncoding.

So there is not 5'-3' problem, and the merging of ape chromosomes 2p and 2q to give human chromosome 2 is eminently reasonable.