Just had this thought occur to me.

If one were to take a DNA sample(or is it RNA?) of a caterpillar before it became a chrysalis, and attempt to match the sample against one taken after the chrysalis matured to a butterfly, would the two samples come up identical?

  • 1
    $\begingroup$ By 'signature' do you mean sequence? Is the sequence of caterpillar DNA the same as the sequence of butterfly DNA? And if you do mean this you probably need to differentiate between somatic cells and germline cells in the butterfly. $\endgroup$ – Alan Boyd Dec 10 '12 at 10:28
  • $\begingroup$ @AlanBoyd: DNA Sequence is probably what I had in mind. That thing the Ministry of Law use to definitely identify a person... $\endgroup$ – Everyone Dec 10 '12 at 18:37
  • 1
    $\begingroup$ Sometimes a caterpillar turns into a wasp rather than a butterfly. But then its DNA Is very different! (And it doesn't go on to make more catterpillars) $\endgroup$ – Nick Dec 11 '12 at 15:14

The genome (entire DNA sequence) of the butterfly would be identical to that of the caterpillar in all somatic cells. A caterpillar has the genes to produce wings, for example, however at that stage in it's development they are not 'switched on' to make the necessary proteins.

If you were to sample the mRNA produced from transcription of the DNA in the caterpillar and butterfly stages, there are likely to be many differences. As in the example I previously used, the mRNAs that are translated into wing-forming proteins are more likely to be apparent in the butterfly than the caterpillar. When saying this, please be aware that I'm assuming that the regulation of these genes is pre-transcriptional - I'm not an expert of butterfly physiology!

  • $\begingroup$ Forgive me, but would it be possible for you to elaborate on 'pre-transcriptional' ? $\endgroup$ – Everyone Dec 14 '12 at 19:34
  • 1
    $\begingroup$ @Everyone there are several points in protein production where regulation can happen to stop it being produced - this happens naturally (for example there is no benefit to producing chlorophyll in root cells where there is no light). Some of these points are before the DNA is transcribed into mRNA (pre-transcriptional) - if the gene is being suppressed here then there should be no template mRNA produced that could be detected. If it was being regulated post transcriptionally then mRNA would be present (though likely in small amounts before being degraded). $\endgroup$ – Rory M Dec 15 '12 at 14:12
  • $\begingroup$ Alan Boyd's Answer suggests that there is more going on than just gene suppression/expression, however. $\endgroup$ – Rory M Dec 15 '12 at 14:13

@Rory M

Here is evidence for developmental genome rearrangements in Hymenoptera. I have emphasised the relevant section of the Abstract. It is probably premature to assume that similar changes might not happen in somatic cells of some Lepidoptera.

Bigot Y, Jegot G, Casteret S, Aupinel P, Tasei JN (2011) DNA modifications and genome rearrangements during the development and sex differentiation of the bumble bee Bombus terrestris. Insect Mol Biol. 20:165-75.

Bombus terrestris is a bumble bee that, like most hymenopteran species, exhibits ploidy-specific sex determination controlled by a single sex gene. Depending on their ploidy and the queen pheromone repression, the imagoes differentiate into three castes: males, workers and queens. Here, we focus on the differences of genome organization that occur during development and sex differentiation. We found that cytosine methylation is a significant epigenetic factor with profiles that can be correlated with both processes. We also showed that two kinds of genomic rearrangement occur. The first consists of important DNA amplifications that have sequence profiles that differ in the different developmental instars and sexes. In the second kind, DNA losses also occur, at least involving the mosaic transposable element B. terrestris mosaic repeat 1 (BTMR1).

  • $\begingroup$ True, but the genome sequence is still the same. $\endgroup$ – nico Dec 10 '12 at 17:50
  • 1
    $\begingroup$ @nico I don't understand what you mean. If two cells differ because in one of them a region of the genome is amplified (e.g. single copy > tandem repeats) then the genome sequences are different. There may be no appearance of DNA with a novel sequence, but I don't think that anyone would argue that the two sequences were identical. $\endgroup$ – Alan Boyd Dec 10 '12 at 18:07
  • $\begingroup$ sorry, forget my comment, I had read the quote too quickly and thought it was just speaking about epigenetics. $\endgroup$ – nico Dec 10 '12 at 22:19
  • $\begingroup$ @AlanBoyd I imagine the germline cells' DNS remains the same? $\endgroup$ – Nick Dec 11 '12 at 15:15

I would like to expand on Rory's answer.

You can think about the DNA like a very complex book of recipes (genes) for constructing cellular machinery (e.g. proteins). Every genome has a large repertoire of machines that it can construct. For a given organism, all its somatic cells will have the same DNA (more or less, but let's put that aside) since it starts out as a single cell. In spite of this, you can see that cells can have completely different shapes and functions. For example, look at how different a neuron is from a muscle cell or a blood cell in terms of shape and function. In addition, cells can change in reaction to their environment.

Then how is the repertoire the same but the cells completely different?

The solution is choosing which genes to activate. This is the called gene regulation and is an extremely complex system as you might guess. It is to some extent the "brain" of the cell, with which the cells can make decisions and gain its plasticity in biological function. This system operates on many levels, one of them being by controlling transcription, i.e. transcriptional regulation, which controls how much RNA to make from each gene, which will later translate into some level of activation (no RNA means the gene will not be active). This is why the RNA will probably vary widely according to the cell type and cellular state.

In general, each species have a similar DNA sequence give or take some minor variations. The sequence of family members will be even more similar, and the sequence of clones (e.g. identical twins) will be virtually identical. This type of variation is random and thus will not be used as a control system. From this you can already guess that the change from caterpillar to butterfly will be controlled by its gene regulation program, as it is a developmental change, in the same way as a human grows from a single embryonic cell.

That said, it is always possible that some random changes occur somewhere along the way and would cause differences, as biology is stochastic. However, for these to be consistent between all cells in the organism, they would have to happen when the organism is a single cell, which is not the case when transitioning from caterpillar to butterfly.


Here's an interesting story from npr.org, Are Butterflies Two Different Animals in One? The Death And Resurrection Theory.

Apparently some people are proposing a theory that says (heavily paraphrased) that a butterfly is actually two different ancestral species rolled into one. This is the relevant quote:

The old view was that over millions of years, animals evolved this habit of switching from one set of instructions to the other. The new view is that this is not one animal gradually changing shape, but rather instructions for two different animals sandwiched together and this change is so radical, says Bernd, "with no continuity from one to the next, that the adult forms of these insets are actually new organisms."

Far out.

  • $\begingroup$ And a chicken is a new organism, radically different than an ovoid, calcium-covered egg. $\endgroup$ – mgkrebbs Dec 13 '12 at 7:48
  • 1
    $\begingroup$ That's a bit different of a scenario ... or do you look at a carton of a dozen eggs and classify them as 12 living and breathing organisms sitting on a shelf? $\endgroup$ – Steve Lianoglou Dec 13 '12 at 16:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.