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

Endorphin-morphine chemical/physical connection?

  1. Jul 9, 2013 #1
    Endorphins in the body act as chemical signals to the morphine-opium receptors (MOR) in the brain and morphine derivatives also react to this receptor. For example alpha-endorphin is a peptide with 10 amino acids and morphine has an aromatic 5-ring non-planer structure.

    What set of physical/chemical properties of morphine resemble physical/chemical properties of some part of alpha-endorphin that allows morphine to substitute as endorphin and react with the receptor like endorphin?

    Edit: I think I can make the question more precise:

    Does the conformational and electrical properties of morphine resemble the same properties of some portion of the endorphin peptide and if so, is that particular region of the endorphin molecule involved with the bonding at the active site of the receptor?

    Edit 2: Morphine has a phenanthrene backbone. Phenanthrene is a ring with three fused benzene rings. And so are there other (non-morphine derived) phenanthrene derivatives capable of exerting similar effects as morphine?
    Last edited: Jul 10, 2013
  2. jcsd
  3. Jul 10, 2013 #2
    Typically, protein-ligand interactions rely on a handful of interactions due to the non-covalent nature of the interactions. You usually need some key amino acid side chains to be arranged appropriately in space in order to favorably interact with some part of the ligand. Proteins and ligands/substrates are extremely diverse and much depends on the specific system one is studying but, in general, you are correct. For some type of analog to bind to a protein you need the "bare-bones" factors (IE geometry and electronics) to be present in the analog as in the natural ligand.

    Here is a pdb of the μ-opioid receptor bound with a ligand (abstract says its from a mouse). You can read the paper and take a look at the structure with some type of pbd viewing program (jmol, pymol, swiss pdb viewer etc).


    This should shed some light on your specific question.
  4. Jul 10, 2013 #3
    Hi Yanck,

    Your link is causing lots of security concerns on my PC so I stopped it. I'd like to view in 3D, both the receptor or at least its active site, and then have morphine come in and bond to it showing which amino acids are participating in the bonding and their conformation relative to morphine's key role. I bet it's the phenanthrene interaction and I suspect other phenanthrenes agonize the mu-opiate receptor as well.

    I'll add to the mix:

    Fentanyl also is a potent mu-opioid receptor and it doesn't have the phenanthrene backbone. It is aromatic with three rings though.
    Last edited: Jul 10, 2013
  5. Jul 11, 2013 #4
    The link is legit, it is to a protein data bank with the exact thing you are looking for. You can try and google it yourself, the pdb code is 4DKL. In the link that I found the receptor is bound to an antagonist (morphinan). If you download any of a number of free pdb file viewing programs, listed in my first response, you can see exactly what you are asking (except no movement of course). You can also get the paper and read about the interactions, I'm sure there will be discussion about the binding site in the text (also found within my link). If you want to learn a bit more about the pdb archives you can check out wwpdb.

    If you still have concerns you can check out other databases such as the Structural Biology Knowledgebase. You should be able to search for the pdb file there.
  6. Jul 11, 2013 #5
    Outstanding then Yanick. At lest one of us is current in biochemistry. Can you explain to me how to download the paper? I can't figure it out but will try some more. Also, where in the 3D image is the antagonist? It's not obvious to me. There's a lot of different things in that nice 3D picture which I assume is the receptor. I realize a lot of it is the tertiary structure of the protein but looks like there are other things in the picture as well. You know what they are? May I ask how well do you understand that picture?
    Last edited: Jul 11, 2013
  7. Jul 11, 2013 #6
    Link to the paper is here. I'm not sure if you will have access since it's a Nature paper and requires a subscription.

    Seeing the ligand requires knowing how to use whatever program you may be using. I don't know how to use all of them but in general there should be some type of list of residues or other objects which you can display or not. If you let me know which program you are using to view the pbd file I may be able to help more.
  8. Jul 11, 2013 #7
    I'm using JMol Java viewer. I don't see any option for residues. May take me a while to figure it out.

    I'm studying the structure. I think the antagonist is the small-chain gray-colored with 4 red colors, molecule. It's showing them at several locations. Would be nice to have the paper. I'll try and find it without a subscription.

    Edit: Ok I got the paper. The antagonist is beta-funaltrexamine and the picture below is from the paper I hope I can post that here. It beautifully shows the interactions of the antagonist with various part of the protein.

    I was wondering if someone could help me interpret a specific part: at the top, the cyclopropane ring of funaltrexamine is hydrogen bonding to the two amino acids Tyrosine and Tryptophan. How would we explain that hydrogen bonding in terms of electrostatics?

    Edit 2: Also, I would like to be able to rotate funaltrexamine inside the receptor so I could see more clearly the bonding. That would be a nice improvement to that work. I suspect though they don't have that set up.

    Attached Files:

    Last edited: Jul 11, 2013
  9. Jul 12, 2013 #8
    I've never used Jmol so I can't help you there, but it looks like you got the paper which should help. I skimmed the text last night and the paper is very interesting, it has gone on my "to read" list which is much too long already.

    The cyclopropane ring and the Tyr/Trp interaction is probably not an H-bond, but some other non-covalent interaction. I've read of some people claiming H-bonding in things like methane clathrates but AFAIK alkyl groups do not H-bond. Not all of the interactions shown in the figure are H-bonds and not all H-bonds may be optimal since H-bonding has covalent and ionic character (so its not just dipole/electronics which are important but geometry as well).

    You should also have seen that the antagonist is covalently bound to the protein via Lys233, that is the ticket to it being an irreversible antagonist and also imposes a restriction on the conformations/geometries available to the antagonist in the binding pocket.

    AFAIK you can't rotate the ligand within the protein since it is in one pdb file, the elements (or whatever they're called) are fixed. I'm also not a crystallographer and don't know all of the in's and out's of making and manipulating pdb files, so maybe there is a way but I don't know of any. I've messed around with docking simulation's and they allow you to do something like what you want but not exactly and they require a good amount of time and effort to get going.
  10. Jul 12, 2013 #9
    Hi Yanick,

    Seems the mu-opioid receptor is a rhodopsin-type g-protein coupled receptor, i.e., made up of those seven alpha helices we see in the 3D plot which I think pass through the vesicle membrane seven times. Apparently these g-protein types are used throughout mammalian biochemistry.

    Would be nice to see a precisely-complicated picture of one of the receptors embedded in it's membrane matrix, then have an animation of how an agonist reacts with the extracellular projections to effect a conformational change which I think then affects the intracellular projections causing a cellular response. I read the intracellular response is coupled with the conversion of GDP to GTP through special G-proteins. I assume then the high-energy GTP can be further coupled into an energy-requiring cellular response such as in the synthesis of a neurotransmitter or hormone, or initiate changes in ion flow (nerve response).

    Really interesting part of biochemistry in my opinion. :)
    Last edited: Jul 12, 2013
  11. Jul 12, 2013 #10


    User Avatar
    Science Advisor
    2017 Award

    I don't know the details of the mu-opiod receptor, so the following is more of a general description of G-protein coupled receptor (GPCR) signalling:

    GPCRs are generally thought of as very conformationally flexible molecules that exist in the plasma membrane. The various ligands of the GPCRs – both natural endogenous ligands and synthetic drugs – act by binding to the receptor and stabilizing different conformations of the receptor. Certain conformations of the receptor position the intracellular portions of the receptor such that they can activate the G-proteins associated with the receptor.

    Thus, the action of a specific ligand depends on which conformations of the receptor the ligand stabilizes. Ligands that stabilize conformations where the intracellular portions are incompatible with G-protein activation act as antagonists, blocking activity of the receptor. Ligands that stabilize the active conformation of the receptor, where the intracellular portions are capable of activating the G-protein, act as agonists that turn on signalling from the receptor.

    Which conformations a ligand stabilizes depend largely on the shape of the molecules and how these shapes fit within the ligand binding pocket. The ligand binding pocket can either be the binding site for the endogenous ligand or it can be an allosteric site that exists on another part of the protein. Although sometimes these allosteric sites have evolved in the protein as binding sites for other endogenous ligands that can modulate the activity of the receptor, some allosteric pockets exist as a consequence of the imperfect way in which proteins fold and have no endogenous ligands.

    Much like the GPCRs, the G-protein activity depends on their shape, which in turn depends on the bound nucleotide. When bound to GDP, the G-proteins adopt an inactive conformation. When the G-proteins exchange the GDP for GTP (for example, when activated by GPCRs), the G-protein changes shape and its new shape allows it to interact with and activate other signalling proteins (such as adenylate cyclase or the PI3Ks) which can cause other changes within the cell. Eventually the G-protein, either spontaneously or through interacton with a GTPase activating protein (GAP), will hydrolyze GTP to GDP, turning off the G-protein. Thus, the high energy nature of the bound GTP is not involved in G-protein signalling; rather, it just provides a mechanism for eventually turning off activity of the G-protein.

    Although it would be nice to look at all the conformations that GPCRs can adopt in their native membrane environment, the structural biology of GPCRs remains a fairly young field of research. We have a relatively small number of GPCR structures available and most of the structures are of receptors in their inactive states. It might be possible to look at some of these conformational changes computationally, however, such simulations are currently very computationally intensive. Furthermore, unlike soluble proteins where we have fairly good force fields to calculate the interactions between atoms in proteins in aqueous environments, the force fields we have for membrane environments are less reliable. Figuring out these question and others – how receptor conformation is linked to ligand binding, how various lipid molecules can modulate the conformations of the receptor, how to more effectively model receptor-ligand interactions – are active areas of research that will hopefully see important strides forward in the coming years.
    Last edited: Jul 12, 2013
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook