Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How many times does an average protein fold?

  1. Sep 27, 2016 #1
    Does anyone know how many times would an average protein fold in its lifetime?
    And how long do proteins live on average?

    Also, another quick question, somewhat related, any knowledge on how many "cascades" of different proteins making a conformational change on different new proteins can there be up to? Creating back-to-back strand of interactions, until some "process" is finalized?

    Many thanks to any thoughts!
  2. jcsd
  3. Sep 27, 2016 #2

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    Interesting question. AFAIK, proteins fold into a 'metastable' configuration, and during folding there are many 'chaperone' proteins that help. This is not universally true: amyloid conformations of proteins are a true minimum energy configuration, yet result in dysfunctional proteins (often, they form plaques). About 30% of new proteins are misfolded and are degraded in the proteasome.

    Here's a good review article:
    http://palgrave.nature.com/nature/journal/v426/n6968/full/nature02261.html [Broken]
    Last edited by a moderator: May 8, 2017
  4. Sep 27, 2016 #3
    Great, thanks Andy! Very interesting! 30% is a very large number!
    Very interesting about "helping/chaperone" proteins, I still wonder if it is known how many back-to-back foldings bunch of proteins can cause in order to finish a task. (on average at least)
  5. Sep 28, 2016 #4
    I just found these few resources that talk about "Signal transduction" and "signaling cascades"

    And also, in regards to the protein longevity, the article below states that "protein half-life is just 90 minutes, while for mammals it may be more like 1 or 2 days"

    Still curious about how many times a protein can on average change its conformation back and forth in its lifetime.
  6. Sep 28, 2016 #5


    User Avatar
    Science Advisor
    2017 Award

    Protein lifetimes vary substantially for different proteins. Some proteins are degraded almost immediately after synthesis. Some, like the crystallin proteins that make up the lens of our eyes, are produced in utero and last our entire lifetimes (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3570024/).
  7. Sep 28, 2016 #6
    Great, thanks Ygggdrasil! I think this basically answers my last question too, about how many times any given protein might fold, which is probably anywhere from once, up to a very large number of times. :)
  8. Sep 28, 2016 #7

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    I think it's worth making the distinction between 'folding'- as in acquiring secondary/tertiary/quaternary structure, and 'conformation change', which is different and usually applied protein-ligand binding. Proteins fold (or mis-fold) once; channel proteins can open and close (conformational changes) rapidly and often. I guess conformational changes can be applied to tertiary and quaternary structure, tho...
  9. Sep 28, 2016 #8
    That is a great distinction, thanks Andy!
    In my own words, to make sure I got this one right, is it safe to conclude that a protein "folding" in a way is the overall assembly of a protein? If primary (and I guess secondary) "foldings" get "unfolded", that would be the end of the protein, as there is no molecular machinery to put it back together, (humpty dumpty comes to mind randomly)
    Meanwhile, the tertiary and quaternary structures are sort of the "active" or "passive" states of the protein that can lead to different "conformational changes"?
  10. Sep 29, 2016 #9

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    My understanding is that 'folding' refers to the creation of secondary structure during translation, transforming a linear sequence of amino acids into a 3-dimensional functional molecule. This process is, in the cell, essentially irreversible. In the lab, proteins are 'denatured' (unfolded) using heat, salts, and detergents to allow biochemical analysis: electrophoretic separation, mass spectroscopy, etc.

    Conformational changes, by contrast, occur within the cell as a matter of normal function. Conformational changes are reversible, but I should say I'm somewhat uncomfortable with the usual presentation: a protein and ligand bind *here*, and the protein changes shape *there*. How exactly does that happen? What does 'shape' (other than equipotential surfaces) even mean for a macromolecule?

    And there are lots of post-translational modification that occur- various molecular groups get stuck onto the folded protein, these can modify the overall structure.
  11. Oct 3, 2016 #10
    So clarified! Thanks so much @Andy Resnick!

    In one of my previous posts here about intermolecular forces, @Fervent Freyja referenced Force Fields, the algorithms to represent complex molecules forces. Just as an example, I remember seeing "AMBER" referenced as protein and DNA force field. Could it be that the different "states" of these force fields could be different conformation states of any given protein? In other words, might that address the "shape" issue and frame it more as a force field associated with that molecule's current state?

    Also, just wondering, does the secondary folding happen in the Golgi apparatus?
    Last edited: Oct 4, 2016
  12. Oct 4, 2016 #11

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    I'm not that familiar with Amber:

  13. Oct 5, 2016 #12

    Fervent Freyja

    User Avatar
    Gold Member

    Those force fields are models and used more for simulations in molecular dynamics, they aren't real descriptions. You cannot layer them together that way. As Andy noted, there are different modifications that can effect the 'shape'.

    I'm not sure where secondary structure is formed. I'm under the impression that modifications can occur while in the ribosome all the way to the cytoplasm. Maybe someone else knows?

    To clarify some terminology here, a folded protein is usually considered tertiary structure. There are databases for known protein folds, around 1400 folds are classified at http://www.cathdb.info/ and http://scop.berkeley.edu/.

    For future reference, what grade level are you at?
  14. Oct 5, 2016 #13
    Thanks for your response Fervent Freyja. Basically, once a protein is folded into the tertiary structure, it is ready to take on different conformational changes, depending on its interactions with the surrounding molecules.

    My understanding was that once a molecule assumes a specific structure, or folding, or a conformational change, its way of interacting with its surrounding changes, with the change in its overall "interatomic potential", due to the forces of all of its molecules. I was using "force field" as a shorthand to describe the intermolecular forces between all the molecules, even if the actual algorithms might not be the actual descriptions but just simulations.

    Perhaps there is a better word or expression to use? That was the attempt I was going for in the thread when I first heard the force fields concept ( https://www.physicsforums.com/threa...ns-intermolecular-forces.878355/#post-5517998 )

    Thanks again for your responses!
    Last edited: Oct 5, 2016
  15. Oct 5, 2016 #14


    User Avatar
    Science Advisor
    2017 Award

    When thinking about the conformational changes of proteins, biophysicists like to talk about the concept of an energy landscape. As you may have leaned in chemistry class when discussing thermodynamics, systems tend to want to find a state of minimum free energy, and the difference in free energy between different states can tell you about the relative amounts of the different states at equilibrium (i.e. the equilibrium constant). Chemists often visualize the energy landscape of a chemical chemical reactions as something like the picture below:
    Where the reactants are associated with a certain free energy, the products are associated with a lower free energy (with the difference being the ΔG of the reaction), and the two are separated by an energy barrier (Ea, or the activation energy). The x-axis of the plot represents the "reaction coordinate" and in this case it represents the whether the intermediate is more reactant-like or product-like.

    Biophysicists view protein folding from a the similar view of an energy landscape, but instead of a one-dimensional reaction coordinate, protein folding occurs in a very highly multidimensional space (one can think of a polypeptide as a flexible chain where each amino acid has two "joints" around which the chain can rotate to achieve the different conformations). Here, each point of the energy landscape represents the thermodynamic stability (i.e. the free energy) of a particular protein conformation. Here is one artists visualizaiton of the energy landscape of protein folding (sometimes referred to as a "folding funnel"). High energy structures (i.e. unstable) are at the top, and the low energy conformations are at the bottom:
    Similarly, conformational changes can then be thought of as the meandering of protein around on the energy landscape. Because the energy landscape depends on the interaction energies between all of the atoms in the protein, changing the environment, for example, by adding a ligand that interacts with the protein, can change the energy landscape and affect both the most stable conformation at equilibrium as well as the "excited" states that are accessible:
    For example, the image above shows a protein that can exist in two different conformations: an open conformation (O) and a closed conformation (C). In the absence of ligand when the protein is unbound (U), the open state has lower energy, so the protein mostly stays open. However, the barrier between the two states is low enough that thermal energy can sometimes excite the protein into the higher energy closed state. However, the presence of the ligand changes the thermodynamics such that in the bound state (B), the closed conformation now has the lower energy, and the open conformation is the excited state.

    If you are interested more in the topic, the Molecular Driving Forces textbook by Ken Dill is a good reference to learn the basics.
  16. Oct 5, 2016 #15
  17. Oct 21, 2016 #16
    Just had a follow up question to this thread. (might almost be a new topic, but it seems to tie in well to the last response)

    When and how do these ligands that affect the conformational state of proteins get produced? Are they produced at the same time the proteins (that react to them) get coded? Or is it totally random, or even opposite, that proteins got developed in response to ligands that come into a cell?

    I seem to mainly gather information on different DNA and different proteins, but ligands seem to be as much involved in creating the diversity of different forms of life.

    For example, would an anti-body be considered a "ligand"? The creation of anti-bodies is directly linked to the functioning of white blood cells and their proteins, in response to a specific target that they need to "search for". Or am I off track here completely?

    Many thanks again!
  18. Oct 22, 2016 #17


    User Avatar
    Science Advisor
    2017 Award

    Are you asking about the evolution of ligand-receptor interactions? Probably the best studied case of how protein-ligand interactions evolved are the steroid hormone receptors, such as the estrogen, progesterone, and testosterone receptors. Evolutionary biologists have shown the ancestral animal genome encoded only one hormone receptor, the estrogen receptor, and that the other receptors evolved later. The next receptor to evolve was the progesterone receptor, and interestingly, progesterone is actually an intermediate produced during estrogen synthesis. So, the cell exploited this per-existing hormone precursor to evolve a second axis of hornome signalling (http://www.pnas.org/content/98/10/5671.long).

    Antibodies are generally considered receptors, and the antigens they recognize are considered the ligands. Antibodies are somewhat of a special case of ligand-receptor interactions. Immunity is not inherited, but instead is "learned" by the body as part of the adaptive immune system. Immune cells generate many essentially random antibodies through a process called V(D)J recombination, in which the immune cells re-shuffle the DNA in the genes encoding for the antibody.
  19. Oct 22, 2016 #18
    Fantastic! Thanks Ygggdrasil!
    And right, an antibody would be a receptor.

    In your estrogen example, I am gathering that "steroid hormone" is a ligand.
    Would a ligand like this be coded at the same "time", or at the same "place" as the ligand receptor? In other words, when a gene encodes an estrogen receptor, there would have to be another gene in one of the chromosomes somewhere that is encoding for the ligand?

    I would imagine yes, and that we are all both born with both estrogen and testosterone receptors, and depending on which genes are expressed, we would have certain different amounts of estrogen and testosterone "ligands" in our bodies?

    And then, there are different smells from the outside world that are not made in a body but the body can have receptors to? Like olfactory receptors?

    Finally, are ligands just another group of proteins that just have the name ligands because of their functionality? Perhaps that is where my missing pieces with forming this picture lie?

    Many thanks again!
    Last edited: Oct 22, 2016
  20. Oct 22, 2016 #19


    User Avatar
    Science Advisor
    2017 Award

    Ligand is a general term for a molecule that a protein binds. Some ligands are small molecules. This is the case for the steroid hormones (small molecules derived from cholesterol). Other ligands can be small peptides (chains of tens of amino acids), while some ligands are complete proteins.

    While protein and peptide ligands are directly encoded by the genome, small molecules are indirectly encoded. The genome encodes the proteins responsible for synthesizing the small molecule ligand. Usually quite a few proteins are involved in the biosynthesis of small molecule ligands.

    Some receptors sense intracellular signals, so the ligands and receptors are produced in the same cell. Many receptors are involved in intercellular communication, so the ligand and receptor are produced in separate cells. Sometimes the communication is between neighboring cells that are physically touching (e.g. at neural synapses) and other times the communication can be between very distant cells (e.g. insulin produced in the pancreas binding to insulin receptors in the liver to modulate glucose uptake). Some receptors sense ligands produced by entirely different organisms (for example, I used to study the receptor in legume roots that helps sense signals produced by the nitrogen fixing bacteria that live symbiotically in the root nodules). Olfactory receptors are another example of a receptor that can sense many different kinds of ligands (mostly small molecules) that could have biological or non-biological origins.
    Last edited: Oct 23, 2016
  21. Oct 22, 2016 #20
    One could not wish for a more thorough answer!
  22. Oct 23, 2016 #21
  23. Oct 25, 2016 #22
    If I may indulge into another continuing thought/question that ties into all this:

    Is there a term or a name for the progress of a group of proteins that belong to and are coded by the same genome, that end up having back to back interactions after they get triggered by an "external" molecule that doesn't belong to the same genome?

    For example, a molecule of certain "scent" hits the nose and then that triggers a chain reaction of all the proteins, and ligands, neurotransmitters, and perhaps even production of certain new proteins, until that "stream" of interactions comes to a "completion", at least as far as the molecules that belong to the same genome are concerned.

    I think it is an "intermolecular interactions" or interactome, and the overall signal is a "signal transduction" but maybe that's not it?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted