Two postings back, I promised a commenter called Sierkovitz that I would discuss epigenetics. This is an important subject with major implications for understanding natural genetic engineering in evolution. So here is the first of at least three related blogs.
"Epigenetics" literally means "over or above genetics." It refers to hereditary changes in genome expression that do not involve alteration of DNA sequences.
Contemporary ideas about epigenetics have two independent historical sources that have subsequently merged in a remarkably satisfying way. The first source was theorizing about cell differentiation and morphogenesis by Conrad "Hal" Waddington, one of the most imaginative and penetrating mid-20th-century geneticists. Waddington realized that a heritable control process was necessary for cells with the same genome to form tissues containing different kinds of cells. In 1942 he called this the "epigenotype," meaning a higher-level regime placed over the genome during development so that different sequences could be expressed in distinct cell types.
The second source of epigenetic ideas came from observations on DNA packaging in the cell. The DNA in our cells would be over 6 feet in length if stretched out, but the nucleus is only about 1 ten-thousandth of an inch across. Clearly, our genomes are densely compacted to fit in such a small volume. Moreover, the packing has to be highly organized so that replication, transcription, chromosome movements, and all other genome functions proceed smoothly.
The historical reality is that cytogeneticists (literally, cell geneticists) had been observing DNA compaction since the 19th century through their microscopes. They described various forms of "chromatin" (i.e., colored material) along the length of chromosomes. The prefix "chroma-" refers to the coloration of chromosomes by various stains used to make them visible. Normal staining was called "euchromatin" (i.e., "true" chromatin), and darker staining was called "heterochromatin" (i.e., "different" chromatin).
Using distinguishable chromatin regions in her maize stocks, the pioneer cytogeneticist Barbara McClintock and her student Harriet Creighton were the first to demonstrate that chromosome physical structure corresponds to a genetic linkage map. From studying what was initially considered a marginal phenomenon in genetics, "position effect variegation," geneticists came to understand that differences between eu- and heterochromatin had a profound impact on genome expression.
Today, we understand that the molecular basis of DNA compaction into chromatin provides the epigenetic control system that Waddington first postulated in the 1940s. The way the chromatin forms regulates how accessible the chromosomal DNA is to proteins and RNA molecules that carry out replication, transcription, repair, recombination, natural genetic engineering, and attachment of protein motors and filaments for moving the genome within the nucleus.
During cell differentiation and development, distinct cell types "index" different regions of the genome into expressed and unexpressed chromatin domains. Thus, the set of encoded functions can be "canalised" (Waddington's term, with British spelling) into those appropriate for each specialized cell type. There are special signals and processes that punctuate the genome for formation into chromatin domains that may span a significant number of separate coding regions.
DNA in chromatin is modified chemically and compacted in two ways:
Cells control chromatin structure exquisitely. They have a chromatin formatting and reformatting system that is a wonder of molecular signaling and control. There are arrays of specialized "chromatin-formatting" enzymes that add or remove methyl groups from the DNA and other enzymes that add or remove various chemical groups from specific amino acids in the "tails" of the histones that peak out from the nucleosomes. These covalent (stable) chemical modifications of the DNA and the histones constitute an intricate code that the cell can read to determine the accessibility status of the underlying DNA, independently of its sequence.
Chromatin structures can be inherited or changed during the life cycle and across generations. This is most easily appreciated from the phenomenon known as "imprinting." The expression of many genetic loci does not depend on the DNA sequence alone; it also depends on which parent contributed it to your genome. When sperm and egg cells form in male and female gonads, particular regions of the genome are epigenetically imprinted into special chromatin domains so that they will either be expressed or not post-fertilization, during embryonic development and after birth. Some loci are only expressed from the father, and some only from the mother. The imprintings are stably inherited through all the cell generations it takes to make an individual.
We are beginning to discover how many epigenetic configurations are inherited across generations. This is the kind of inheritance that can have a profound influence on evolution, because it can be selected and expand in a population. Some very stable multigeneration genome changes are, in fact, "epimutations" that do not involve any DNA sequence alteration, only chromatin reformatting.
Other transgenerational changes are reported to be induced by a wide range of environmental factors, such as environmental stress, endocrine disruptor chemicals, infection, and even maternal grooming. The ability of environmental factors to cause inherited epigenetic modifications makes this form of genome coding particularly attractive to neo-Lamarckists.
There is a great deal more to say about epigenetic inheritance. In the next couple of blogs, I will discuss epigenetic memory and targeting, overturning a six-decade mainstay of evolutionary dogma, and the connection between epigenetic changes and natural genetic engineering of DNA sequence organization. Stay tuned for some eye-opening research.
This article points out an interesting parallel between epigenetic factors controlling cellular development and memory/cognitive functions.
http://www.cell.com/neuron/retrieve/pii/S0896627311004338#MainText
An example:
"Investigation of the precise molecular mechanisms in both cellular development and memory has increased over the past two decades, and an interesting new understanding has emerged: developmental regulation of cell division and cell terminal differentiation involves many of the same molecular signaling cascades that are employed in learning and memory storage. Therefore, cellular development and cognitive memory processes are not just analogous but are homologous at the molecular level."
The authors are of course not talking about natural genetic engineering (NGE), but could some of their findings apply to NGE?
I'll discuss at least some of what we know about epigenetic control of NGE in the third blog of this series. Maybe you'll see a relationship there, but I think more work needs to be done first like the exploratory experiments outlined in the last two blogs. Once we have evidence for coordinated changes in real time, it becomes possible to investigate what brings the coordination about.
Berthajane Vandegrift
A Few Autistic Questions about Freud, Marx, and Darwin.
Meiosis reduces the number of chromosomes in each cell by half and so produces gametes that are called "haploid." The reduction occurs because meiosis involves only one round of DNA replication that doubles the genome but two separate cell divisions that reduce the number of chromosomes by a factor of four. The two divisions thus end up with a net halving the number of chromosomes in each cell.
Although I am not an expert and may be wrong, I suspect what you heard is that meiosis in females is halted after the first meiotic cell division early in life. There is then a long period, and the second meiotic division occurs just before the haploid eggs are released into the uterus at ovulation for fertilization.
Berthajane Vandegrift
New work from Howard Hughes Medical Institute (HHMI) scientists suggests those abundant molecules (piRNAs) may be part of the cell's search engine, capable of querying the entire history of a cell's genetic past.
“This is really remarkable. It implies that an organism has a memory of all the previous gene sequences it’s ever expressed before.”
Craig C. Mello
http://www.hhmi.org/news/mello20120625.html
wow!
harry
Welcome back and thanks for the link.
The next blog deals with just this kind of epigenetic memory. The emphasis is more on DNA sequences that have come in from outside, but the process is basically the same.
Your "wow!" validates my promise, "Stay tuned for some eye-opening research."
thanks. I'll definitively stay tuned.
If humans also have a piRNA system, the implications for cancer are ominous. As I recall, quite a few tumors are "reverted" cells, with properties which resemble embryonic stem cells. The piRNA system could let such cells sail right through, since the embryonic genes had been expressed once before in an ancestral cell.
As I replied to Harry, the piRNA system is a kind of immune memory of past invaders, not of expressed functions. In the few cases I know about, the piRNAs serve to direct silencing, not permit expression. They are also specific to the germline and would not be active in somatic tissue that becomes cancerous.
So, please don't lose any sleep over this part of epigenetic memory. We do know there are profound epigenetic changes in tumors, and there may be occasion to visit them in the future. But I need to learn more about the subject myself.
I look at the human body, where each cell has 750MB of code and each different type of tissue is built by epigenetically expressing different portions of that code. I see an amazing efficiency. The genome is 750MB and Microsoft Windows is 25 gigabytes; humans are certainly superior to Windows.
As a person who builds all sorts of systems, trying to make use of every resource I have, it makes me wonder: Trillions of cells, each with the same 750MB of code, seems excessively redundant.
However if each one of those cells could epigenetically apply a unique context to its own information in real time, thus storing more information, you have an incredibly clever mechanism for archiving not just terabytes but petabytes of additional data, epigenetically distributed across billions of cells.
And if those same cells can communicate with each other, it might better explain how an entire human body adapts to, say, training for a marathon. The stress affects every cell in the body, each in a slightly different way. Thus an entire organism would have more than enough space to store incredibly detailed information about its own unique context in the world.
Then perhaps the very most important adaptations would be passed to offspring, IF the signalling systems could coordinate appropriate information to the reproductive cells.
I suppose all this could sound far-fetched to some. But is there evidence that any of this could be true?
I was glad to see that you pointed out how the epigenetic encoding tremendously increases the capacity and flexibility of the genome as a RW storage system. With all the possible chemical modifications of the histones, the combinatorial possibilities are immense, and there are also specialized histones for certain tasks, like DNA repair.
As for passing on the effects of life history experiences to future generations, see the links provided by Rhynocstylus below. Amazing examples in a long-term study of an isolated Swedish population. Now that we have mechanisms for non-DNA inheritance, we will see many more well-documented cases emerging in the literature. Some of the laboratory studies are also posted at http://shapiro.bsd.uchicago.edu/ExtraRefs.GenomeCompactionChromatinFormattingEpigeneticRegultion.shtml under the heading "Epigenetic responses to stimuli, epigenetic memory, and transgenerational effects."
Re: "Cells control chromatin structure exquisitely. They have a chromatin formatting and reformatting system that is a wonder of molecular signaling and control."
Formatting, signaling and control would seem to be the result of intelligence and forethought. what is the evolutionary explanation for those attributes?
If you'll forgive a pun, I find it a gas when people want to learn about the nitty-gritty of what cells can do with their genomes. Gives a whole new meaning to the idea of what life is all about.
Does the good Professor thoroughly the fallout from this basic question? I am sure he does. Science advances by the willingness of investigators to proceed in the face of 'received dogma' which constrains the interpretation of new data to the strictures of old and failing paradigms.
We still lack a detailed explanation for the origins of this remarkable control system. We can see simpler versions of it in yeasts and other microbial eukaryotes and can observe how the system becomes more complex through events like duplications and domain shuffling of the proteins involved.
Here are some links to recent papers that may give you some leads if you can navigate the PubMed database. I don't have space to provide the complete references: http://www.ncbi.nlm.nih.gov/pubmed/21507350, http://www.ncbi.nlm.nih.gov/pubmed/20097869, http://www.ncbi.nlm.nih.gov/pubmed/20210320, http://www.ncbi.nlm.nih.gov/pubmed/21119628, http://www.ncbi.nlm.nih.gov/pubmed/21596317, http://www.ncbi.nlm.nih.gov/pubmed/21647299, http://www.ncbi.nlm.nih.gov/pubmed/22388813, http://www.ncbi.nlm.nih.gov/pubmed/21663790, http://www.ncbi.nlm.nih.gov/pubmed/20738881, http://www.ncbi.nlm.nih.gov/pubmed/21119629, http://www.ncbi.nlm.nih.gov/pubmed/21527910.
As you might imagine, the literature on this subject is vast, and the field has not yet been able to digest if all from an evolutionary perspective.
Thank you for your comment and for the hyperlinks. That should provide enough research material to keep me off the streets for some period of time.
But I think my original question might have been misunderstood to some extent. My thoughts were rather going along the lines - if the mechanisms responsible for the epigenetic mechanisms are themselves encoded - can they still account as if they "do not involve any DNA sequence alteration, only chromatin reformatting.".
Or is it also plausible that the genetic changes involved in conservation of epimutations are just not as straightforwardly linked to them as the regular mutations (eg. epimutations are caused by regular mutations elsewhere, epimutations are a result of some changes in promoter sequences of regulatory genes etc.). The topic is quite interesting and somewhat linked to my research, at least conceptually.I research evolution of bacteria and how acquisition of new "allele" or variants in a population can happen via alternative routes rather than "regular" mutations. And the thought that continuously bothers me is - if let's say natural transformation is still a mechanism that is coded, if its prevalence can be traced to particular conditions and particular mutations - is it still horizontal or rather a laterally transferred but horizontally active system. Sounds bit confusing, but its a concept that I still have not tackled completely, so please do feel free to ask for clarifications where needed:)
If you follow the link on "epimutations," you will find a paper on "paramutation" by Vicki Chandler. This is one case where we know that the stable changes are in chromatin formatting and are not the consequences of sequence changes elsewhere in the genome.
It remains, true, however, that mutations disrupting the epigenetic silencing system can reverse a wide variety of epimutations (where the epimutant state is silent) and induce yet others (where the epimutant state is expressed). In the next blog, I will explain a case where a stable mutant state can be induced by disrupting epigenetic silencing and then reversed by targeted epigenetic silencing.
There are no clear examples of similar stable epigenetic changes in bacteria, but we do know about many loci where DNA methylation plays a key role in regulation and switching expression of particular sequences on or off (e.g. http://www.ncbi.nlm.nih.gov/pubmed?term=17220888).
Please say something about your research and how it may be linked to epigenetic regulation.
http://www.sciencenews.org/view/generic/id/341931/title/Climate_adaptation_may_be_a_family_affair
Instant epigenetics!
Thanks for the link. This may be another case of transgenerational inheritance of epigenetic modifications. But that is not the only possible explanation. We are also beginning to find reports about transmission of small RNA molecules to offspring, and there may be yet other forms of non-DNA inheritance at work.
It is exciting that the molecular studies are beginning to give us ways to understand phenomena that we were taught for so long were unthinkable or had been disproven. My next blog will contain a different example of a previously "disproven" way that cells change their genomes in response to life history events.
Berthajane Vandegrift
A Few Autistic Questions about Freud, Marx and Darwin
I am not familiar with the work you are mentioning. Please give us a reference, or even better a link. That way we can all see what you are bringing to our attention. Thanks in advance.
http://en.wikipedia.org/wiki/%C3%96verkalix_study
I actually had to read the whole article and pay attention while doing it.
Can't wait to read the follow up articles.
You have warmed my heart. One of the reasons I started blogging was to learn how to communicate contemporary science in a way that is accessible to non-specialists. It's gratifying to hear that I may actually have succeeded. Thank you.