Transcription factors
Oct. 30th, 2014 11:33 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
livSo I posted a link to one of those silly internet quizzes, this one being run as a promotion by a fairly minor scientific journal. And wow, I had forgotten how good those daft "what colour is your aura" personality quizzes are for generating conversation! I posted the type of protein one mainly because I was amused by how ridiculously over-specialist it is, but in fact people with no interest at all in protein chemistry wanted to have a go and talk about what the results meant.
And since people are interested, I might have a go at explaining the background behind the quiz, and also why I think transcription factors are cool. Slightly disappointingly, there were only five outcomes for the quiz, which considering how many types of proteins they might have included is a shame. But if you take the quiz, the options are: receptor, kinase, transcription factor, chaperone, metabolic enzyme.
This is not unconnected to the fact that the silly quiz appeared in Cell Signal, a highly specialized journal for people who have a very specific paradigm of what's important for studying cell function. I mean, I mostly work within that paradigm too, for preference, but I try to be aware that there are other ways to describe cell biology. So the idea of people who have a cell signalling mindset is that cells are like little decision-making machines. A bunch of inputs come in from the environment, which includes the surrounding tissue and the needs of the whole body in the case of multicellular organisms, and these set off a complex but ultimately deterministic program of changes in the cell, allowing the cell to adapt to changing conditions. The information being passed from outside the cell to change cell biochemistry and therefore behaviour is referred to as signalling.
So, a receptor is a type of protein that receives the initial signal from outside the cell. For the most part these signals are chemicals; sometimes they're electric currents or physical forces, but mostly a molecule originating from outside the cell binds to a receptor and starts off the signalling cascade. When we talk of a chemical binding to a protein, that means that there is a very close match in shape, charge and other properties between the shape of the chemical and the shape of the protein, so they can stick together, sometimes by forming chemical bonds and sometimes by just fitting in, but either way, binding isn't just being in the vicinity, it's changing the chemical nature of the protein. So a receptor acts as a kind of sensor for a very specific chemical signal. Typically receptors sit at the surface of the cell, the better to interact with neighbouring cells, the bloodstream and the environment. This means that if cells don't have a particular receptor, they can't respond to the chemical that receptor detects; iodine, for example does nothing at all except in the thyroid gland where it has dramatic effects, because only the thyroid gland has receptors that recognize iodine.
When a molecule binds to a receptor, the receptor changes to a new form. This in turn alters how the receptor interacts with other proteins, which we call signal transduction proteins (unfortunately not an option in the quiz!) Sometimes the receptor becomes enzymatically active and directly chemically modifies its targets. Sometimes it physically moves inside the cell so that it comes into contact with proteins not found at the surface. Sometimes physical changes in its shape kind of "push" other neighbouring proteins about, changing their shape and chemical properties in turn. A single receptor usually interacts with a number of signal transduction proteins, meaning that it can have more than one effect simultaneously. And the signal can become amplified, because one molecule binding to one receptor may alter dozens of transducers, and each of those can pass the signal on to lots more in an exponential way. And the targets may be lots of the same thing or several different things. (This is part of how, for example, the human eye can detect single photons a whole lot more effectively than any manufactured light sensor.)
One important kind of signal transduction protein is a kinase. Kinases are enzymes involved in moving phosphate ions around the place. This seems like a slightly obscure thing to care about, but it turns out to be massively important for cell function. The thing about phosphate ions is that they are highly electronegative, with a charge of -3. This means that having a phosphate ion stuck on to a protein (or other biochemical) massively alters the shape and function of that phosphorylated molecule, because everything in the molecule that has any kind of negative charge flees away from the massive repulsion of the phosphate ion. Moving phosphates around therefore takes huge amounts of energy to counteract this electrical repulsion, and it turns out that a huge proportion of the energy available to the cell (ultimately from metabolizing food, or possibly absorbing light photons directly in the case of plants) is used precisely for this purpose. So, a receptor may bind its molecule, and set off a cascade where a kinase-kinase-kinase (no, really!) is activated and phosphorylates a kinase-kinase, which in turn gets activated by the phosphate getting stuck to it, and phosphorylates a kinase, which in turn phosphorylates something else.
Eventually the signal reaches a target that actually does something other than passing on the signal. That could be, for example, a metabolic enzyme as mentioned in the quiz. Metabolic enzymes are those which catalyse changes in non-protein biochemicals, such as breaking down food into simple sugars and then using those sugars to generate energy for moving phosphates around. Um, and other energy-requiring processes like muscles contracting and so on, but we're in a cell signalling paradigm here! And anyway, even with muscles, changing the chemical energy from food into kinetic energy of contraction involves a fair amount of moving phosphate ions around.
But if you want to have changes on a timescale of more than minutes, you often need to change the gene expression pattern of the cell, not just the individual metabolic enzymes that happen to be present. The gene expression pattern refers to which of several thousand genes are active at any given moment, and therefore which proteins are actually present in the cell doing stuff, as opposed to the cell just having instructions for how to make them. Enter the transcription factors. Transcription factors have the ability to bind to specific sequences of DNA and set up a situation where the transcription machinery can do its thing of expressing genes, or carrying out the instructions of the DNA sequence to make proteins. So, if an activating signal reaches a transcription factor, that TF will bind to its target genes and start a chain of events which caused the genes to be transcribed, ie copied into mRNA, and the mRNA then acts as the template for the cell's protein-making machinery to make a particular protein.
Right, so, the cell's protein making machinery? It's a total Heath Robinson thing, it really has a very large number of components which interact in very complex ways. One component of the protein machinery is the chaperone proteins mentioned in the silly quiz. What they do is stop partially formed proteins from collapsing into a big useless mess, and instead guide them to take up their proper shape so that they actually work. Some people who took the quiz were a bit sad at the name chaperone because it suggests mean older relatives stopping young people from having fun. But what chaperones really do is stop the kind of "inappropriate associations" that are not sexy fun times, but rather the cause of conditions like BSE ('mad cow disease'), caused by proteins forming sticky blobs that just gum up cells instead of behaving properly.
Anyway, to return to transcription factors. See my icon there? That's part of p53, which is the most interesting protein in the world *ahem*, but also happens to be a transcription factor. If you look at that little blue spiral to the right of the picture, that is exactly the right width to lie in the groove of the DNA double helix. When p53 is activated, four sections like this join together to form a semi-rigid structure with its reading fingers exactly at the right spacing to recognize certain patterns in DNA. Reading is a key concept here: transcription factors bind specifically to a particular sequence of DNA bases, because of the shape of the DNA, without needing to separate the two strands of the double helix to expose the complementary base pair bonds. In other words, they can read a dormant sequence of DNA, identify a region near the start of a gene, and only act if that gene is one that should be switched on at this time. This means they act as a kind of landing platform for the rest of the transcription machinery, most of which can't read the DNA when the strands are stuck together, or really at all, the transcription machinery once activated pretty much just copies blindly whatever sequence happens to be there.
Most TFs are like p53 in that they need several components to stick together to be able to hold their reading fingers in the right position. This means that they can be very flexible; some may work with two or four of the same kind of thing, but many can also form complexes with different proteins. This is one of the things that mean that two different cells may receive the same signals, and activate two different sets of genes, because one has transcription factors A and B available, but the second only has A and C, and the complex AB will recognize different bits of DNA from the complex AC (and A won't do anything on its own).
So it gets complex and combinatorial fast. Here's an interesting fact, partly discovered by the lab I worked in in Sweden: there is no correlation between gene number and organism complexity. I'm a little suspicious of the concept of "organism complexity", because I think it's a little anthropocentric to assume that human-like things, ie multicellular animals, are inherently "more complex" than single-celled amoebae, but in as far as you can measure complexity, it doesn't correlate with gene number. You've probably seen a variant of this cartoon, which states:
Clear? Confusing? Over-simplified? Anyway I hope this goes some way to help you interpret your silly quiz result, and also to tell you why transcription factors are cool!
And since people are interested, I might have a go at explaining the background behind the quiz, and also why I think transcription factors are cool. Slightly disappointingly, there were only five outcomes for the quiz, which considering how many types of proteins they might have included is a shame. But if you take the quiz, the options are: receptor, kinase, transcription factor, chaperone, metabolic enzyme.
This is not unconnected to the fact that the silly quiz appeared in Cell Signal, a highly specialized journal for people who have a very specific paradigm of what's important for studying cell function. I mean, I mostly work within that paradigm too, for preference, but I try to be aware that there are other ways to describe cell biology. So the idea of people who have a cell signalling mindset is that cells are like little decision-making machines. A bunch of inputs come in from the environment, which includes the surrounding tissue and the needs of the whole body in the case of multicellular organisms, and these set off a complex but ultimately deterministic program of changes in the cell, allowing the cell to adapt to changing conditions. The information being passed from outside the cell to change cell biochemistry and therefore behaviour is referred to as signalling.
So, a receptor is a type of protein that receives the initial signal from outside the cell. For the most part these signals are chemicals; sometimes they're electric currents or physical forces, but mostly a molecule originating from outside the cell binds to a receptor and starts off the signalling cascade. When we talk of a chemical binding to a protein, that means that there is a very close match in shape, charge and other properties between the shape of the chemical and the shape of the protein, so they can stick together, sometimes by forming chemical bonds and sometimes by just fitting in, but either way, binding isn't just being in the vicinity, it's changing the chemical nature of the protein. So a receptor acts as a kind of sensor for a very specific chemical signal. Typically receptors sit at the surface of the cell, the better to interact with neighbouring cells, the bloodstream and the environment. This means that if cells don't have a particular receptor, they can't respond to the chemical that receptor detects; iodine, for example does nothing at all except in the thyroid gland where it has dramatic effects, because only the thyroid gland has receptors that recognize iodine.
When a molecule binds to a receptor, the receptor changes to a new form. This in turn alters how the receptor interacts with other proteins, which we call signal transduction proteins (unfortunately not an option in the quiz!) Sometimes the receptor becomes enzymatically active and directly chemically modifies its targets. Sometimes it physically moves inside the cell so that it comes into contact with proteins not found at the surface. Sometimes physical changes in its shape kind of "push" other neighbouring proteins about, changing their shape and chemical properties in turn. A single receptor usually interacts with a number of signal transduction proteins, meaning that it can have more than one effect simultaneously. And the signal can become amplified, because one molecule binding to one receptor may alter dozens of transducers, and each of those can pass the signal on to lots more in an exponential way. And the targets may be lots of the same thing or several different things. (This is part of how, for example, the human eye can detect single photons a whole lot more effectively than any manufactured light sensor.)
One important kind of signal transduction protein is a kinase. Kinases are enzymes involved in moving phosphate ions around the place. This seems like a slightly obscure thing to care about, but it turns out to be massively important for cell function. The thing about phosphate ions is that they are highly electronegative, with a charge of -3. This means that having a phosphate ion stuck on to a protein (or other biochemical) massively alters the shape and function of that phosphorylated molecule, because everything in the molecule that has any kind of negative charge flees away from the massive repulsion of the phosphate ion. Moving phosphates around therefore takes huge amounts of energy to counteract this electrical repulsion, and it turns out that a huge proportion of the energy available to the cell (ultimately from metabolizing food, or possibly absorbing light photons directly in the case of plants) is used precisely for this purpose. So, a receptor may bind its molecule, and set off a cascade where a kinase-kinase-kinase (no, really!) is activated and phosphorylates a kinase-kinase, which in turn gets activated by the phosphate getting stuck to it, and phosphorylates a kinase, which in turn phosphorylates something else.
Eventually the signal reaches a target that actually does something other than passing on the signal. That could be, for example, a metabolic enzyme as mentioned in the quiz. Metabolic enzymes are those which catalyse changes in non-protein biochemicals, such as breaking down food into simple sugars and then using those sugars to generate energy for moving phosphates around. Um, and other energy-requiring processes like muscles contracting and so on, but we're in a cell signalling paradigm here! And anyway, even with muscles, changing the chemical energy from food into kinetic energy of contraction involves a fair amount of moving phosphate ions around.
But if you want to have changes on a timescale of more than minutes, you often need to change the gene expression pattern of the cell, not just the individual metabolic enzymes that happen to be present. The gene expression pattern refers to which of several thousand genes are active at any given moment, and therefore which proteins are actually present in the cell doing stuff, as opposed to the cell just having instructions for how to make them. Enter the transcription factors. Transcription factors have the ability to bind to specific sequences of DNA and set up a situation where the transcription machinery can do its thing of expressing genes, or carrying out the instructions of the DNA sequence to make proteins. So, if an activating signal reaches a transcription factor, that TF will bind to its target genes and start a chain of events which caused the genes to be transcribed, ie copied into mRNA, and the mRNA then acts as the template for the cell's protein-making machinery to make a particular protein.
Right, so, the cell's protein making machinery? It's a total Heath Robinson thing, it really has a very large number of components which interact in very complex ways. One component of the protein machinery is the chaperone proteins mentioned in the silly quiz. What they do is stop partially formed proteins from collapsing into a big useless mess, and instead guide them to take up their proper shape so that they actually work. Some people who took the quiz were a bit sad at the name chaperone because it suggests mean older relatives stopping young people from having fun. But what chaperones really do is stop the kind of "inappropriate associations" that are not sexy fun times, but rather the cause of conditions like BSE ('mad cow disease'), caused by proteins forming sticky blobs that just gum up cells instead of behaving properly.
Anyway, to return to transcription factors. See my icon there? That's part of p53, which is the most interesting protein in the world *ahem*, but also happens to be a transcription factor. If you look at that little blue spiral to the right of the picture, that is exactly the right width to lie in the groove of the DNA double helix. When p53 is activated, four sections like this join together to form a semi-rigid structure with its reading fingers exactly at the right spacing to recognize certain patterns in DNA. Reading is a key concept here: transcription factors bind specifically to a particular sequence of DNA bases, because of the shape of the DNA, without needing to separate the two strands of the double helix to expose the complementary base pair bonds. In other words, they can read a dormant sequence of DNA, identify a region near the start of a gene, and only act if that gene is one that should be switched on at this time. This means they act as a kind of landing platform for the rest of the transcription machinery, most of which can't read the DNA when the strands are stuck together, or really at all, the transcription machinery once activated pretty much just copies blindly whatever sequence happens to be there.
Most TFs are like p53 in that they need several components to stick together to be able to hold their reading fingers in the right position. This means that they can be very flexible; some may work with two or four of the same kind of thing, but many can also form complexes with different proteins. This is one of the things that mean that two different cells may receive the same signals, and activate two different sets of genes, because one has transcription factors A and B available, but the second only has A and C, and the complex AB will recognize different bits of DNA from the complex AC (and A won't do anything on its own).
So it gets complex and combinatorial fast. Here's an interesting fact, partly discovered by the lab I worked in in Sweden: there is no correlation between gene number and organism complexity. I'm a little suspicious of the concept of "organism complexity", because I think it's a little anthropocentric to assume that human-like things, ie multicellular animals, are inherently "more complex" than single-celled amoebae, but in as far as you can measure complexity, it doesn't correlate with gene number. You've probably seen a variant of this cartoon, which states:
The study found humans only have about twice the number of genes as a fruit fly. Of course, the data could be off a bit. The work was done by researchers who had only about twice as many genes as a fruit fly.Now D. melanogaster just happens to be a common genetic model organism which we think of as pretty simple. They could have picked another organism to compare, such as a single-celled trypanosome which has more genes than a human, never mind half as many! But here's the thing: there is a correlation between organism complexity and the number of transcription factors. Because having more TFs gives you combinatorially more flexibility in your gene expression programme. So you can do more different things with the same basic set of genes. TFs are highly evolvable, in other words, because you make a slight change in a TF, and you turn on different genes in response to the same signal. And that's a huge source of variation for selection to act on.
Clear? Confusing? Over-simplified? Anyway I hope this goes some way to help you interpret your silly quiz result, and also to tell you why transcription factors are cool!
