Do we know the genome contains all the information necessary for embryonic development?

Question

Since high school biology I've assumed that all information necessary for creating an organism is found in its genome. As Nature's Scitable site says:

Deoxyribonucleic acid, more commonly known as DNA, is a complex molecule that contains all of the information necessary to build and maintain an organism.

However, I've been reading Sean Carroll's "Endless Forms Most Beautiful" and here is one passage that illustrates the question in the title.

Ultimately, the beginning of spatial information in the embryo often traces back to asymmetrically distributed molecules deposited in the egg during its production in the ovary that initiate the formation of the two main axes of the embryo (so the egg did come before the chicken).

Dr. Carroll seems to be saying that the information determining whether genetic switches are on or off ultimately does not come from the embryo's genome, but rather comes from other molecules provided by the mother. So, my question is whether we know these molecules responsible for initializing the switches that regulate gene expression can be traced back to something in the mother's genome? If we don't know these molecules trace back to the mother's genome, then is there some other reason we think the genome "contains all of the information necessary to build and maintain an organism"?

Answer 1

The second quote is about stochastically determining which side is the "head" and which is the "tail", effectively, which occurs by positive feedback of an existing gradient. This is not really a matter of information stored in the sense that we think of information stored in DNA bases or bits on magnetic tape, the passage seems to be referring to information in the sense of a reduction in uncertainty about the state of the universe. The way a developing embryo chooses where the head is going to be is like dumping a bag of M&Ms on the table and having the head develop wherever the highest concentration of yellow M&Ms are. If you expect them to be uniformly distributed, you'd be correct statistically, but for any one iteration there will be a non-uniformity that determines what happens next: that's information, reduction in statistical uncertainty. If you, via the power of a thought experiment cousin of Maxwell's demon, were to scramble the positions of molecules in the egg cell, that would change where the "head" develops.

Yet, all of this is guided by proteins that are created and function according to the "instructions" in DNA. DNA doesn't have information about where in the future there will randomly be a concentration of yellow M&Ms, but it does have the information to synthesize proteins that react to concentrations of yellow M&Ms, causing a positive feedback loop that further concentrates them on a particular axis.

For embryonic development, these yellow M&Ms are mRNA transcripts, and at least in drosophila, these initial transcripts are indeed made in the egg cell, not post-fertilization (see https://en.wikipedia.org/wiki/Drosophila_embryogenesis#Anterior-posterior_axis_patterning_in_Drosophila). It's possible to introduce a mutation to a fly that will only affect her offspring, by interfering with transcription of these mRNAs. More generally, maternal effects are observed when the egg cell or developing environment has some influence on the offspring.

However, all those influences are coded in the DNA of the next generation, too, so it also seems reasonable to say that the genome contains all the information necessary. Containing all needed information to do X is not the same as being able to do X. To turn DNA into RNA transcripts, and to turn transcripts into proteins, requires more than information, it requires machinery. We can say the information to build those machines is also in the DNA.

Honorable mention to mitochondrial DNA, which comes entirely from the egg cell.

Generally, I think learning about biology from these grand pronouncements like "all information is in DNA" is a bad way to learn biology. Biology is not mathematics, we do not operate in the certainty of proof. Learn how things actually work, first, rather than trying to prove or disprove "truths". These statements are meant to capture wonder and convey the basic order of things, they are not dogma (even if someone regretfully calls it that).

Answer 2

First of all, concerning the theory of epigenetics and the concept of epigenetic modification of 'gene expression' -- or of the data encoded in DNA in the form of its base sequence and structural conformation --, it is important to realise that all the enzymes implicated in such epigenetic processes must themselves be encoded in DNA and produced through the mechanisms of transcription and translation, all of the enzymes and proteins associated with which processes, transcription factors and so on, and including histones themselves, must also be thus encoded.

Putting aside such contention about what is inevitably the self-controlling function of DNA transcription however, the conception of a genome constituted primarily by the DNA molecules comprising cell nuclear chromatin is a theoretical construct permitting a broad understanding and its more-or-less consistent experimental verification of the phenomena of heredity and of the relation between the old-fashioned ideas of the phenotype and genotype.

The derivative notion that DNA contains all the information necessary for the growth of organisms is only the imputation of the prevailing paradigm of causality to that relation: thus that the form of DNA 'causes' the intricacies of growth to occur -- and this is no doubt useful in an empirical context, particularly one in which the genome embodied in or by DNA is to be manipulated towards some end.

However, it might be helpful at times to consider the scenario from a different philosophical perspective, to put the horse before the cart as it were, and to imagine that the dynamic and evolving structure of DNA is rather a reflection, at what we define as the cellular level, of the singular process of life, growth and attendant evolutionary transition -- correspondingly understood as an orderly process -- conceived in its more essential sense, than its cause; therefore to view DNA as a theoretical instrument for describing fairly precisely and in an analytically consistent way in physico-chemical terms this singular process, notwithstanding significant gaps in the capacity for that theory generally to account for the entirety of that instrument: for example, to account for the function of non-coding DNA (which must inevitably require a revision and expansion of the philosophical perspective itself).

Clearly there is far more to this process of life and evolutionary growth than can be accounted for by the view of DNA as it currently stands in orthodox cellular biology, even if it is also true that there is no conceivable aspect of that process which will not be reflected -- which will not have its manifestation -- at some level, however subtle, of that dynamic structure and conformation of DNA.