# Is triple-strands or multiple-strands DNA possible?

Before the double helix DNA was discovered by Watson and his colleague, they theorized the existence of a triple-strands DNA and its properties. Is multiple-strands DNA possible perhaps by manipulating the structure of a double helix DNA with science? Multi means more than 2

• Note than in the double helix, only half of the chain provides information (the other half is formed by the "complementary" nucleotide, so you can derive it from the original one. It seems that the main function of the double helix is protection and control of the generation of ARNm. – SJuan76 Mar 29 '15 at 11:39
• I feel the need to say that this might be more at home on the biology SE. While it is still on topic here you might consider asking it there for a more scientific answer. – JDSweetBeat Apr 6 '15 at 18:22
• @DustinJackson Please be careful to not encourage cross-posting, particularly if the user might not have much experience with how the network works. Better to suggest flagging a question for moderator attention and requesting migration if so desired. – a CVn Apr 7 '15 at 18:09
• @MichaelKjörling Not what I intended, but I now see that my choice of words was poor :) – JDSweetBeat Apr 7 '15 at 18:13
• @DustinJackson It's okay, no harm done in this case. Just something to keep in mind for future occasions. – a CVn Apr 7 '15 at 18:14

There has been a claim that 4-strand DNA structures exists in nature, in fact in human cells. University of Cambridge article. The link to the the original paper in Nature seems to be broken though. I am pretty think the Nature paper was more detailed, but it is not my field.

I should also note that there is no evidence of a complete organism based on using triple or higher strands of DNA. But the scientists claimed they had stablized the quad strand DNA, so it would appear to be viable at a molecular level.

Though the Cambridge article is limited (as were the others I checked), there were some interesting details:

The findings mark the culmination of over 10 years investigation by scientists to show these complex structures in vivo – in living human cells – working from the hypothetical, through computational modelling to synthetic lab experiments and finally the identification in human cancer cells using fluorescent biomarkers. The research, published today in Nature Chemistry and funded by Cancer Research UK, goes on to show clear links between concentrations of four-stranded quadruplexes and the process of DNA replication, which is pivotal to cell division and production.

...

“The research indicates that quadruplexes are more likely to occur in genes of cells that are rapidly dividing, such as cancer cells. For us, it strongly supports a new paradigm to be investigated – using these four-stranded structures as targets for personalised treatments in the future.” Physical studies over the last couple of decades had shown that quadruplex DNA can form in vitro – in the ‘test tube’, but the structure was considered to be a curiosity rather than a feature found in nature. The researchers now know for the first time that they actually form in the DNA of human cells.*

...

While quadruplex DNA is found fairly consistently throughout the genome of human cells and their division cycles, a marked increase was shown when the fluorescent staining grew more intense during the ‘s-phase’ – the point in a cell cycle where DNA replicates before the cell divides.

It is possible but impractical.

Double-helix DNA's structure is something like this:

$$\text{B — (A/C/T/G) : (T/G/A/C) — B}$$

where B is a backbone molecule, A/C/T/G is adenine/cytosine/thymine/guanine, — is a molecular bond, and : is a hydrogen intermolecular bond.

What's important here is the molecule connecting the strands: since it's only connecting 2 backbones, it's a relatively simple molecule.

3-way DNA I can't accurately represent with MathJax, so here's a picture:

In this diagram, a line is a molecular bond, a dotted line is a hydrogen intermolecular bond, B remains the same, H and N are hydrogen and nitrogen, and (AF/CF/TF/GF) is some variation of (A/C/T/G) where the molecule must incorporate a fluorine molecule on the outside of the molecule.

The part in the center is $\text{NH}_3$, which I have picked purely because it has 3 outer hydrogen molecules. These hydrogens form hydrogen intermolecular bonds with the fluorine atoms in (AF/CF/TF/GF).

There are relatively few 3-way molecules like $\text{NH}_3$, so 3-way DNA is more difficult and more fragile than 2-way.

4-way DNA is also possible: substitute $\text{NH}_3$ in the diagram for $\text{CH}_4$, methane, which has 4 outer hydrogens and can thus form intermolecular bonds with 4 (AF/CF/TF/GF) - B groups.

The major point to note here is that DNA is the way it is because it's simple: 2-way DNA does not require another molecule in the middle like my representations of 3- and 4-way DNA do. While my ideas are probably not optimal, 2-way DNA is both simpler and stronger than any other type.

(Also, DNA helicase (one of the enzymes that process DNA) would have a hard time adapting for 3- or 4-way DNA.)

It is possible to produce triple stranded genes using some variation of DNA, PNA, XNA, etc. I recall reading about PNA as a third strand as a possible means to bootstrap our modern DNA scheme.

I doubt three-stranded genes would be found in a life system of a mature planet ecosystem. It would evolve toward simplicity, and lose one.

But, that's a way of having an intermediate step offering the possibility of changing the encoding system. E.g. an asymmetric Peptide-Ribo two strand system is good for early days because it is easy to emerge from RNA-world and has high matching affinity. But symmetric DNA is better after various other mechanisms mature and the cell gets more complex. So it first gains a 3rd strand, changes the RNA to DNA, then loses the PNA.

The real problem with multi strand DNA isn't so much if it is possible (Dr Science can do some pretty freaky stuff in the lab), but how it would work in the natural environment.

To put it bluntly, if there were some evolutionary advantage to 3 stranded DNA, then there would need to be some means of combining it during reproduction: 3 parents would be needed. If this means 3 different sexes are needed is probably up to you as an author, a more likely scenario is two "male" parents and a "female" parent. (Star Trek gets it wrong with their "Species 8472" since they claim 5 genders but only 3 strand DNA. Unless their reproduction includes phases where they burrow into a host like parasitic wasps, this is hopelessly inefficient, and since two of the genders are not represented in the DNA, why would they evolve in the first place? A five gendered species would have 5 strand DNA.)

• That is not how dna works. You do not get one strand of the double-helix from each parent - you get full chromosomes from each parent (excluding abnormalities where the offspring get only one or 3 or even 4). Additionally, there are many species in the real world with more than 2 sexes (some fungi have thousands) - this does not mean that every sex is needed for reproduction, just that there are many compatible potential mates of different kinds. – pluckedkiwi Mar 31 '15 at 17:27