How does DNA really work

Ok, so DNA basically contains the entire code for building a creature from a single cell, right? But how does it actually do work? How does a double helix of various protein molecules tell a bunch of cells to keep dividing until they produce a heart, liver and all the other organs, all in the right shape and in the right place? I mean does each cell have some sort of identifier that tells in “you are heart cell 00124-01354-44412 and you belong at these coordinates.”?

Is there an example of something similar on a smaller scale I can wrap by brain around?

It’s complicated.

The connection between the very small scale of protein molecules and the very large scale of organs and organisms is still an active area of research, which is science-speak for “we haven’t got it fully figured out yet.” Basically, though, we know that there are immensely complicated networks of signals between a cell’s proteins and DNA and the external environment. That is, there are probably systems in place that let a cell essentially look around itself and say, “ah ha! I’m supposed to be a heart muscle cell!” and then turn on the appropriate genes at the appropriate times.

We understand the whole process much better in the model organism C. elegans, which is made up of a thousand or so cells that always split in exactly the same way. A lot of development work has been done on it, and we now know the life history of every single cell in the body, and have made huge progress in understanding the aforementioned networks. Fruit flies have yielded a great deal of additional information, though I think it’s safe to say we don’t understand it at the same level of detail. Yet.

For a very small scale example, here’s the kind of thing that happens. Some signal, like a hormone molecule, binds to a receptor protein on the surface of your cell. Let’s say it’s a growth factor that is trying to tell your cell to divide. The protein is activated by this binding (and in many cases, this mysterious-sounding “activation” is understood on an atomic scale), which causes it to directly or indirectly create a second messenger molecule inside the cell. Typically, this process is repeated several times - each messenger activates a protein, which creates another messenger. Eventually, a messenger binds to a protein called a transcription factor, which activates, binds to a particular stretch of DNA, and causes a particular gene to “turn on”. This gene might code for, say, a protein that begins the process of DNA replication, which has to happen before the cell can divide.

In short, every cell has identical DNA. What determines the developmental fate of a cell isn’t something internal, but a set of external signals (for the most part), as well as the history of that cell.

I was going to take a stab at giving an example for the aforementioned C. elegans, but it’s been done much more eloquently here.

By the way, it should also be mentioned that these “external signals” are generated by other nearby cells as part of their pattern of normal development.

I’m kind of glad I looked at the URL in the status bar before clicking that. It’s not really NSFW, but I wouldn’t want a supervisor looking over my shoulder when the page title came up.

Just about the best visualization I know off for how DNA leads to proteins (and some other things regarding mainly DNA replication) is this one; I believe I found the link to this some time ago on this board, so credit to whoever put it up back then.

Morphology is still largely an unexplored area when it comes down to the nitty-gritty of exactly how genetics determine the overall form of large-scale organs and structures like bones. That is, your femur and your skull are both made of the same stuff, more or less, and yet they initially form and then maintain completely different shapes. How?

Answers are going to be necessarily vague at this point in our understanding, but the short answer is that it probably works somewhat like a colony of bees. Worker bees aren’t clones, but they have very similar genetic makeup because of the way that bees reproduce. So while they don’t share truly identical DNA like the individual cells in your body (presumably) do, for practical intents and purposes, a worker bee is a worker bee is a worker bee. And yet, individual worker bees can take on a wide variety of distinct jobs within the colony – guards, caretakers, pollen gatherers, carpenters, etc. The same worker bee may even change roles in response to the hive’s current needs and environmental conditions. But this isn’t because the worker bee is smart enough to analyze something as complex as a bee hive in order to chose what it needs to do. Rather, its behavior is determined by chemical interactions and signals. Worker bees performing particular tasks release specific pheromones associated with that task. Worker bees organize simply by smell. Exposure to different levels of the various pheromones prompts behavioral changes via chemical action (or lack thereof) on neural pathways. So if, say, a large number of “nurse” bees die, the drop in that pheromone level may cause other worker bees to switch tasks and become nurse bees. As they do, they start releasing the associated pheromone(s), which brings the system back into equilibrium once enough workers have switched over.

Ok, so that’s a bit of a digression, but the situation with your own cells is similar. You cells obviously lack the facility for “thought” as we usually mean it, but they are in some sense aware of their surroundings. Cells exchange various chemical compounds with adjacent cells, and in some cases with distant cells via chemicals carried in the blood supply. These chemicals affect the organelles within the cell, which in turn affects RNA transcription, protein formation, and other cellular activities purely through chemical reactions. Alterations to a cell’s behavior can affect the chemicals that it produces and exchanges with neighboring cells and with the circulatory system, thus possibly altering their behavior, too. It’s similar to how one worker bee dying can cause another worker bee somewhere else in the hive to change jobs. It’s not that the other bee actually “knows” the first bee died, it’s just reacting to chemical changes in its own local environment.

Because there are so many chemical interactions to study, we just don’t know enough about them to give a really solid answer to your question. We do understand some of these interactions, but the ones we understand are only a tiny fraction of the whole. What’s more, the affect of two chemical signals in conjunction may different than the affect of either chemical signal alone, and the affect may be different depending on the current chemical state of the cell, or on externally influenced factors like temperature, or the amount of oxygen in the blood. As you can imagine, the situation is enormously complex. Hopefully the hive analogy gives you some idea of how this works, though.

Fair enough. That subject is what I do at work these days, so it didn’t occur to me. Mods, could you add a “Slightly Awkward for Work” tag, and break the link if you see fit? The link is to a discussion of vulva development in the worm.

That is so hawt. :stuck_out_tongue:

Well, better that than worm development in the vulva.

Yes, allow me to introduce the concept of emergence, where a few simple rules create complex behaviour.

An example of this is termite building a mound. Each individual termite just follows a few simple rules:
Find some earth and chew it with saliva to create a pellet.
Wander around at random.
When you find a raised area, drop the pellet and go back to Step 1.
When piles of pellets reach a certain height, start building arches.

The individuals don’t know they are building a mound, but a mound is the result, and the patttern of tunnels and arches is extremely intricate.

see:
http://www.forteantimes.com/strangedays/science/382/hive_minds.html
http://www.beart.org.uk/Emergent/

see also the Game Of Life, a simulation in which cells are born, die, or survive according to the status of cells next to them.

Various behaviour can emerge from the simple rules.
Some initial patterns move accross the grid, while keeping their original shape.
Some form “guns” that fire “bullets” which move accross the screen, keeping their shape. And various other interesting behaviour.

So, cells in a growing embryo can follow a number of simple rules, depending on stimuli they get from surrounding cells, and the result is a complex organism. If they receive a certain stimulus, they create a tiny bit of bone. Lots of cells following that rule create a skeleton. The genes do not contain any design for the skeleton, just the rules to obey. The bone structure is just something that emerges from the rules.

Also look up “Langton’s Ant”, a simple computer simulation in which an “ant” traverses a large square-grid and follows just two simple rules:

  1. If the cell just entered is black turn left, otherwise turn right
  2. Flip the current cell (black/white) before leaving

This seems really simple, but all kinds of interesting things happen, even on a blank sheet.

I wasn’t shocked or anything. :slight_smile: I just figured a web page with “vulva” on it would be tricky at work.

I would strongly question that basic assertion. Ultimately, I agree with most of what others have posted, but I’d approach it from a different direction in which we stop over-emphasizing the importance of DNA.

DNA contains code for two things:

  1. building protein
  2. regulating protein production

You can find dozens of examples of vital hereditary or functional components that aren’t in DNA. Mitochondria have their own DNA and are passed in the ovum, but are still essential to metabolism. While DNA has code for ribosomes, it can’t build them without pre-existing ribosomes (again, in the ovum for higher animals). DNA in real organisms forms a complex 3-D structure thanks to protein histones that are partly responsible for regulating/activating genes; like ribosomes, you inherit your first set of histones. Other key regulators are also not DNA - genes can be turned on or off by the presence of hormones, nutrients, metabolic by-products, methylation, etc.

So, to my mind, what’s surprising is that there are any genes (Hox genes, for example) that clearly relate to macrocellular structures. As a general rule, I think it’s better to envision DNA as being totally ignorant about position and function, yet responsive to a host of chemical signals that keep it doing the right thing in the right place.

If I were to take an ovum from an arbitrary creature, empty out the DNA from the nucleus with microscopic tools, and replace it with a full set of human DNA, and then let that ovum develop, what would I get?

Probably a dead cell, depending on how closely related the arbitrary creature was. dracoi has a point, though I wouldn’t put it as strongly as he does. The basic idea that each cell has the full set of DNA required to create the organism is true. However, there are a lot of other things that need to be in place as well, from the complement of protein required to start reading and expressing that DNA to, often, caches of maternal mRNAs to patterns of histone modifications and other epigenetic factors.

I do agree that we’ve spent so much time and energy deciphering the software side of things (DNA) that the hardware has been deemphasized, and now that genome sequencing is a matter of routine, we’re beginning to redress the balance somewhat.