I love that animation. I saw it back when Ben Stein et al were getting started with hyping "Expelled" and someone noticed their little cell animation happened to look a lot like that one.

I do have a question about the animation. That bit with the kinesin protein: Are they really that steady with their molecular locomotion? It just seems to me to be way too straightforward for "life", I'd expect something more like two steps forward, a step back, a little wiggle that's not a step really, then a step forward, etc. etc., but all in all headed mostly forward.

OK, see, that's just too freaking cool.

I really wasn't expecting it to be so ordered on that level, but now that you've pointed out the mechanism... just... wow.

You are right, of course, people who aren't fascinated by biology are way strange.

Well, the description itself was quite simplistic. Biology is highly ordered at the molecular level. The whole business of translating a tiny conformation change in a small binding pocket of a protein (ATP binding and hydrolysis) into a large scale movement requires a great deal of mechanical precision. Motor proteins like myosin, dynein and kinesin have various mechanisms for translating these tiny changes in a small pocket of the protein into a powerful push that moves the protein along the filament (this is responsible for, among other things, muscle contraction, cilia beating and the stiffening of hair cells in response to sound). For example, in a similar motor protein called myosin (moves along actin instead of microtubules, otherwise highly similar) has to perform a motion called a power stroke where the head ratchets forward in response to a very small change in protein structure due to ATP hydrolysis. This works because a small helix structure in the protein which is sometimes referred to as the piston of the myosin head, is swung out when ATP hydrolyzes inside the protein cleft, which amplifies the tiny conformation change into a motion analogous to a rowing stroke.

I've never understood how people couldn't be at least a little fascinated by those processes (even the ones that don't look like cartoon feet dragging a giant balloon). The very fact that we understand so much about things that happen on a cellular level ... you'd think that would be a point of pride, but people seem to find it boring, or they want to watch something on television ... I don't get that, I'm with you guys.

I'll raise my hand as someone who isn't fascinated by biology.  In fact at high school we had three options when it came to science: biology, chemistry & physics.  I chose the latter two precisely for that reason.

I think the one thing that is important to stress is that if you study molecular biology for long enough, you learn that the principles underlying absolutely everything in biological life are pretty much the same. Any interesting occurrance in the body can be characterized in terms of interactions between proteins, proteins and DNA, proteins and RNA. That's it. Molecular biology is little more than the study of those three molecules and their interaction, and anything at all in life can be characterized in terms of them. Cells operate by means of a complex set of feedback loops, switches, control mechanisms, signal integration and computation. Each cell is simply a microprocessor. Every cell in molecular biology can be characterized as a complex set of computational commands. In this way, molecular biology is highly similar to computer science.

Want to know, for example, how certain types of fishes can change their skin color? In fish cells called melanocytes a set of membrane-bound intracellular vescicles hold the protein pigments which, when expressed on the surface, induce a particular absorption spectra. These vescilces are regulated by two motor proteins called dynein and kinesin, already discussed. Inside the cell there is a body called a centrosome which is in the approximate center of the cell with microtobules protroding outwards pointing towards the periphery of the cell. Microtubules are highly polarized. The minus ends of the tubules point towards the centrosome, the plus ends point outward to the cell membrane. The protein pigments are by default carried to the periphery and are part of the constitutive exocytic pathway, which means they are secreted to the surface outside the cell by default. But they can be carried by both dynein and kinesin along the microtubule, and these two proteins walk in opposing directions, as such they exist in a tug-of-war. Dynein will take the pigment vescicles to the centrosome, inside the cell, whereas kinesin will take them to the periphery and deliver them to the exterior. Kinesin will take the vesicles to the surface by default, but when an induced hormome signal activates a cell-surface receptor which binds and activates an enzyme called phospholipase which hydrolyzes a molecule called phosphoatidylinositol 4,5 biphosphate releasing a soluble molecule called IP3 which binds to ion channels in the endoplasmic recticulum which opens them which lets Ca2+ ions into the cell which binds to a protein kinase which activates it which phosphorylates kinesin which deactivates it which allows dyein to win the tug-of-war and aggregate the vescicles near the centrosome, and as a result, the fish rapidly changes color.

The important thing to notice is that the whole process is described as a complicated, but very neat causal chain. "A signal...which induces...which induces...which induces..." etc. These causal chains characterize every single thing occurring in the cell and the body, and they are the object of our study. Depending on certain steps in the pathway, "This induces...which induces...etc" can mean something different. For example, when I say, "A release of Ca2+ ions binds to a protein called calmodulin and activates it" what I mean is "calmodulin has a binding site which recognizes Ca2+. It's normal function is to activate another protein called calmodin dependant kinase, and it is normally inactive. The binding site which recognizes CaM is normally not exposed in calmodulin. When Ca2+ binds, it changes the conformation of calmodulin which allows it to bind to CaM". These sorts of causal chains where proteins activate each other, control DNA, RNA, etc and vice-versa can be used to explain anything we want to in molecular biology. Everything you saw on the video can be characterized in terms of complex causal processes, each step depending on a set of conditions, which makes the cell a very powerful microprocessor. In this way, every process in the cell can be characterized essentially as a set of logic gates. And that is pretty much all biology is. As an example, I just drew an example on my computer which shows how cone cells respond to light:

Although not shown here, AND gates are very common, for example, since many causal chains require multiple inputs and turn on if and only if all input conditions are met.   IFF is probably the most common logical response of something in the cell. Cells can also display non-bivalent logical responses. For example, transcription of a certain gene can be influenced by many different complexes of gene regulatory proteins which either serve to turn on or off the gene, and they all directly influence the probability of transcription, which makes it a continuosly graded variable. As a result, gene transcription is, in many cases, more analogous to a variable resistor than an on-off switch.

I like to think that even though many people are never logical, their cells sure are.

Funny, so did I.

NOTE: Please read this:

Biology and Thermodynamics

Before continuing.

We (PZ and I) are in agreement. Most kinesins move at around 1ums^-1, and for, say 500 ATPase cycles, depending on the rate of ATP consumption, the kinesin will fall off every few seconds. When I said kinesin was processive, I meant it in a molecular context. Myosin, for example, is not processive. It takes a stroke and falls off. This is part of its function since the heads must get out of the way for the next set for a muscle to contract properly. Nonetheless, PZ’s excellent piece has prompted me to outline precisely what is wrong with the Inner Life of the Cell. The descriptions that I, or any other biologist, gives of molecular processes inside the cell can generate false impressions. So, to clear things up:

1.       The cell is ordered but not orderly. It is much denser, faster and more confusing than shown in ILOTC. Proteins and ligands are constantly violently colliding in the crowded cytoplasm. Proteins tumble on their main axis almost a million times per second. They form transient bonds with ligands are rapidly ripped apart. Most mRNA are destroyed and replaced every half hour. The cell is highly dynamic, much, much more active than shown in the video.

This artists conception gives a better impression of the inside of the cell (I’d give you a real picture but we haven’t invented the nanoscope yet)

2.       Our immediate question of concern before we proceed is what actually occurs in the cytoplasm? Our answer is that the cytoplasm, the occurrences in it, and so forth, are defined in terms of the proteins which occupy it. Proteins form signal pathways from the cytoplasm to the nucleus, back and so forth, gene regulatory proteins regulate other proteins, etc. etc. Enzymes operate on their substrates, complex multi-step pathways reach from the membrane to the nucleus, a maze of vesicles are being transported to and from the endoplasmic recticulum and Golgi apparatus, endosomes and lysomess and  plasma membrane, and to (but not from) the mitochondria, peroxisomes etc. Proteins are being shuttled to and from the nucleus, mRNA are being ejected and translated, protein machines are continually destroying old proteins, actin and microtubule and intermediate filaments are constantly assembling, reassembling, disassembling, at sites around the cell to give it shape, mobility, and dynamism. Signals from the outside are reacted to by the inside by means of computing a logical response in terms of the activity of proteins (for example, some extracellular signals switch certain proteins on, which have immediate effects, others turn specific proteins off, still others have more complex effects like inducing gene regulatory proteins to move to the nucleus where they induce the transcription of certain genes. Complex feedback loops ensure that  proteins and hence the activities they result in are regulated. For example, a molecule of tryptophan is a ligand for a gene regulatory protein which binds and stops transcriptions of a set of genes which create the proteins necessary to create tryptophan from substrates. The cell is very very busy maintaining itself in the environment

3.       Given that the cell has to maintain all of these complicated ordered, stepwise pathways in the context of the fact that the cell is a confusing hodgepodge of molecules bashing into each other, how is this ordered existence achieved in a non-orderly cytoplasm? To understand this in any meaningful way requires some understanding of protein kinetics. The function of any protein is determined by how it binds to other molecules. Some molecules (including other proteins) serve to regulate other proteins by up-regulating or down-regulating their function with respect to other molecules. Some molecules act as substrates to enzymes, where they bind and are acted upon by the catalytic active site of a protein. But in all cases where any molecule binds to a protein, it does so by means of a large number of non-covalent interactions with whatever molecule it binds to. This holds true for all interactions. Proteins interact with DNA via non-covalent interactions, with other proteins via non-covalent interactions, with small molecular ligands via non-covalent interactions. Non-covalent interactions are the basis of supramolecular structures such as ribosomes (held together with non-covalent interactions), proteasomes (held together with…well, I’m not even going to bother to tell you this time. Have a guess.) etc. The strength of the non-covalent fit between a protein and a ligand is determined by the number of non-covalent interactions, and the precise orientation of the protein for the ligand. If the protein forms a large number of non-covalent interactions, the protein-ligand complex will be more stable than if it was formed by fewer interactions. This gives rise to a concept called the binding equilibria. Consider a group of proteins and ligands diffusing in a cell. They will encounter each other and form complexes, then disassociate and so forth, similar to a chemical reaction. Eventually they will reach an equilibrium where the rate of association is equal to disassociation. The point being that the ratio of concentration of complexes over ligand/protein is proportional to the strength of the non-covalent interaction. Thus, protein-ligand interaction is a probabilistic consideration, based on thermodynamics. More specifically, for the following binding:

P+nL=>C

The free energy change for disassociation of a complex is related to the equilibrium constant in the following manner:

∆G=∆G^Θ+RTln[C]/[P][L]ⁿ

This is the primary way in which pathways in the cell are restricted. Proteins and potential ligands are violently colliding all the time in the cell, but only if they actually fit (hence form strong noncovalent contacts) will there be a high probability of the complex being maintained. If not, thermal buffeting in the cell will simply rip it apart. If the molecule doesn’t fit, it is not energetically favorable for a complex to form. As a result, proteins only respond to their specific ligands, be they other proteins, small molecules, or DNA, or (sometimes) all three. Many proteins have active sites which have been so precisely tuned for their ligands by evolution that there is little more that can be done to make the binding stronger.

This is good, but not sufficient to explain the ordered nature of all the processes in the cell (vesicular transport, GRP-based regulation, nuclear localization, translocation, translation and transcription etc. etc.) despite the violent and stochastic nature of the cell and the molecules that interact. The cell uses other strategies for this. For example, typically, a substrate must be passed from one enzyme to another to another to form the useful end product. For this reason, there are exceedingly few soluble free enzymes in the cytosol. Most of the time, enzymes are arranged in multi-enzyme complexes. Additionally, with all the protein-based signal pathways being relayed in the cell, cross-talk and confusion can be avoided by means of strategies such as the formation of transiently induced scaffolding complexes or microdomains in the bilayer, or the need for a set of proteins to assemble like lego bricks on a certain induced complex to relay a signal. The cytoskeleton can be regulated by means of confining accessory proteins (like those which nucleate the filaments) to certain regions of the cell. Protein complexes in the membrane tend to diffuse rapidly in the bilayer plane. They can be kept in certain areas by means of junctions or large aggregates which form by protein-protein complexes.

Thus we consider the two central concepts of processes in the cell: concentration and affinity. The nature of chemistry means that the processes are based highly on probability and large number laws. In the same way that you wouldn’t know anything about the probability of getting heads if you flipped a coin three times, but would know a lot if you did it 500,000,000 times, similar mechanisms are at play in the cell. Individual molecules by themselves are meaningless. Cellular effects, like the change in transcription of a certain gene, or the activity of a protein by means of regulation by a another molecule, are meaningful only because of changes in concentration of the respective molecules (refer to the section on chemical equilibria in the thermodynamics discussion). We have already discussed the concept of affinity (previously referred to in terms of binding equilibria). There is some confusion when biologists describe molecular switches to laymen, particularly in terms of allostery. An example is that a protein called calmodulin serves to “switch on” a protein called calmodulin dependant kinase. Calmodulin itself is regulated by 4 molecules of Ca2+. What precisely do we mean by switch on? Not the same thing as in every day life. A molecular switch is not a light switch. Instead of being simple binary, they serve to greatly increase or decrease the affinity of the protein for another molecule by means of inducing a conformation change in the protein which of course changes the geometry of the active site for that molecule and therefore the nature of the noncovalent interactions it forms. This changes the respective probability of binding to that other molecule (refer back to thermodynamics section on equilibria). Hence we see that cellular processes are emergent and stochastic, and depend on laws of large numbers and probability.

That is all.