How do proteins 'know' where to go?

Question

I've just found once again this famous animation I've been curious about for many years: https://youtu.be/WFCvkkDSfIU?t=213

Here's a screenshot from the animation:

The green blobs (proteins, I believe) are moving from outside the scene to "the crowded area" (where unzipping happens). Then some of them (paler-green ones) are leaving in the same way as they came.

I am curious about how they are doing it!

I suspect that the "crowded area" emits some molecules (or perhaps ions), and "green blobs" might be literally electromagnetically attracted to the source due to density gradient of those ions. As for the "paler ones", they might be gaining on some other molecules (after contact), which probably disables their attraction to the "gradient". And the reason why they're so "determined" while leaving remains unclear, but I can guess that the same "gradient" molecules might be just pushing on them (as far as shown).

I would like to watch or read more about such details! And also about two-legged proteins from the end of that video — the ones that look like giant fighting robots from the Star Wars 1st trilogy.

Surely, these couple of things alone would take at least few month of studying.

Answer 1

Even though this animation is very well-known and the narrator says it is "... an accurate representation of the actual DNA replication machine ...", be very careful of its visual appeal. It is, after all, only a fancy animation for educational purposes. Some aspects of the process are shown correctly, but many are simplified, omitted or prettified, and can thus be misleading.

For example, a key ingredient not being shown in this animation (for obvious reasons) is water. DNA and the accompanying proteins do not float in free space, they are immersed in water. But not only that – intracellular molecular enivornments are far more crowded with other proteins than shown in the animation. See for example this whole-cell 3D model of a Mycoplasma bacterium.

Molecules don't know where to go. They move randomly¹, bump into other molecules, and generally tend to disperse evenly across the solution. If you put a little drop of food color in a glass of water, it will eventually spread throughout the water. The same "force" drives macromolecules (such as enzymes shown in the animation) to move around. But when two molecules which are somehow "compatible" just happen to by chance come very very (!) close to each other², molecular forces can help them position in the right way and stick them together. You can watch a simulation³ of molecular dynamics showing a small molecule finding its way into a large protein.

Although your hypothesis about some (mysterious) ions signalization and electromagnetic attraction is somehow reasonable, this is not the correct explanation for the process in animation⁴. First, electric forces in solution with ions are weak and short-ranged due to screening. Second, they could not be specific enough to attract only some molecules and not the others.

I put some keywords such as Brownian motion, diffusion, macromolecular crowding, molecular dynamics ... with links in my text which you can use to start exploring the topic of physical biology. Any textbook on cell biology or physical biology will be a good start, but the choice very much depends on your prior knowledge.

Oh, and the two-legged proteins are molecular motor proteins.

---

_{¹ There is also a directed transport of substances inside of a cell, but this is a different phenomenon.
² Macromolecular crowding actually hinders the movement (diffusion) of molecules around the solution (smaller diffusion coefficient), but it helps them stick together (larger binding constant).
³ Mondal et al. (2018)
⁴ Moving along a gradient of some chemical does happen, but it usually involves membranes or it happens at larger scales.}

Answer 2

This answer is specific to the "two-legged" proteins from the end of the video (motor proteins).

The animation shows identical proteins moving in perfect lockstep, but really there is wide variance in the time between steps, as seen in this graph of the motion of kinesin from M. J. Schnitzer and S. M. Block, Kinesin hydrolyses one ATP per 8-nm step, doi:10.1038/41111:

Here's a plot of the motion of myosin V from a different paper, showing a similar pattern.

This video shows what I think is a somewhat more accurate picture of the motion of kinesin, though it still shows all steps as taking a similar time.

The animations in the question were created by Drew Berry, who is also the presenter. My impression from his voiceover ("It's that mechanical. It's molecular clockwork. [...] This is all derived accurately from the science.") is that he didn't know that his depiction was inaccurate, though that was certainly known in the field at that time (the papers I linked were published years earlier). He doesn't seem to have ever done any biological research, and is primarily an artist. As such I would be suspicious of the accuracy of everything in his videos, but I'm not qualified to evaluate them.

Answer 3

There have been some suggestions that biological molecules may 'know where to go'.

Barry Honig's group at Columbia analyzed the electrical charge arrangements in protein and proposed that by placing charges throughout the protein in specific configurations that electrical fields could be focused and direct diffusion to accelerate the function of some proteins esp in finding their binding partners and reaction substrates.

An early paper for their electrical field analysis package DelPhi proposed that Super Oxide Dismutase had a strong electrical field that would strongly attract the negatively charged superoxide ion into its binding pocket. In particular the hydrophobic core's electrical permittivity in contrast to the water around the protein has a pronounced effect on the electrical field around the protein.

This is still not common analysis because such models are often compromised compared to the realities at atomic scales but they are probably qualitatively valid. Honig's work was most likely motivated by the regular measurements that diffusion in biology is faster than it has a right to be. In particular Super Oxide Dismutase's reaction constant is orders of magnitude above diffusion.

I am looking for the original work above and will bring it in. Another more relevant case to the question is that DNA binding proteins also [find their binding sites faster than diffusion would allow]3. DNA is highly negatively charged (albeit with positively charged ions surrounding it in solution) certainly its reasonable to believe that the orientation of the protein to its intended binding must be important and that the cylindrical shape of the field around the DNA would explain some of these observations. Certainly the advantage of such an adaptation in the protein would be highly preferred in an evolutionary environment.

This topic is covered in the case of several molecules in this reference. Here i've scraped a figure from the paper showing the positive charges around

PartII

I just got to attend a short conference and i have another interesting take on this.

The current thinking is that the ribosome and RNA polymerase both have a bubble of proteins that condense around them that are specifically involved in translation or transcription respectively. These 'condensates' are not compartments in the cell but self assemble because of protein protein interactions. This would make the functions in the cell partitioned because of protein self- assembly. Evidence is still coming together but another example of how biology does not submit to Occam's Razor very well.

Reference : "Honig and Nicholls, "Classical Electrostatics in Biology and Chemistry" Science v268 p1114 (1995)