Mechanisms of Drug-receptor Interaction
This chapter answers parts from Section C(iv) of the 2017 CICM Primary Syllabus, which asks the exam candidate to "to explain receptor activity". Apart from a vague mention in Question 1 from the first paper of 2013 ("list the different mechanisms of drug actions") this matter has never appeared in any of the past papers. However, the topic is ubiquitous and the diagram of different receptor types can be found in just about any pharmacology textbook, usually in Chapter 1. It is therefore not inconceivable that the college examiners will one day flip to that page and decide to ask the trainees to "list the different mechanisms of drug-receptor interactions, giving examples".
The following (extensive, rambling, incoherent) discussion of drug targets was interesting to write, which does not guarantee that it will be interesting to read. Moreover it trespasses the already badly-trampled boundary of exam-relevant "required reading", into unexaminably apocryphal territory. The time-poor candidate will be well-served by reading no further than the grey box which follows. In summary, the following broad mechanism-based categories of receptor drug targets can be identified and linked to a familiar example:
- Soluble extracellular enzymes (dabigatran, perindopril)
- Cell surface
- Cell surface molecules (abciximab)
- Transmembrane nonenzymes (cytokines, interferon-γ)
- Transmembrane proteins with active domains eg. receptor kinases (insulin, imatinib)
- Ligand-gated ion channels (nicotine, suxamethonium)
- Voltage-gated ion channels (lignocaine, verapamil)
- G-protein coupled receptors (dobutamine, metoprolol)
- Soluble intracellular enzymes (glyceryl trinitrate, milrinone)
- Nuclear receptors (corticosteroids, thyroxine)
- Nucleic acids (doxorubicin, azithromycin)
Probably the best overall resource for this is Goodman and Gilman's Chapter 3. This textbook is probably available to most people as a part of some sort of university library collection, whether electronic or papyrus. One might need to reach for a non-canonical resource: Chapter 2 from Golan et al ("Principles of Pharmacology", 4th ed) covers receptor types fairly well. If for some bizarre reason obscure paywalled articles are somehow available but well-established textbooks are not, the affected candidate may wish to read Feener & King's "The biochemical and physiological characteristics of receptors" (1998), which is probably more related to cell biology than to pharmacology. In general, it appears quite difficult to find this information presented in a painless and accessible way. If one trades "painless" for "accessible", one may come up with guidetopharmacology.org, which is a massive (comprehensive) resource of drugs and drug targets, funded and developed by a collaboration between The British Pharmacological Society (BPS) and the International Union of Basic and Clinical Pharmacology (IUPHAR).
Major receptor-effector relationships
Nothings says "painless and accessible" like a complicated diagram which is supposed to clarify matters.
Receptor drug targets
Hypothetically, if one wished to discuss each drug receptor in a separate page, how many entries would one make? Overington et al (2006, "How many drug targets are there?) proposed a number after asessing over 20,000 FDA-approved pharmaceutic products: "all current drugs with a known mode-of-action act through 324 distinct molecular drug targets", they concluded. This seems almost doable. However, in the interest of time, these are grouped into their dominant classes for the purposes of this discussion. Of the drugs in common use, more than 50% target only four cardinal receptor families: G-protein coupled receptors, nuclear receptors, ligand-gated ion channels and voltage-gated ion channels.
These are pore-like transmembrane proteins which alter the local permeability of the cell membrane to ions. Typically, these channels are fairly selective to which ion they are open for.
Ligand-gated ion channels
The binding of a ligand opens the pore, and without the presence of a ligand the channel is closed. In this state some amino acid side-chains block the pore. The ligand tends to bind to the extracellular domain of the channel. A classical example is acetylcholine; when two acetylcholine molecules bind the receptor it undergoes a conformational change and the pore opens. A handy example of a drug ligand is suxamethonium. Many others exist. IUPHAR/BPS have an awesome list of receptor families, endogenous ligans for which include serotonin, GABA, glycine, glutamate, inositol triphosphate, and zinc. Generally, these channels mediate rapid millisecond-level ionic fluxes across membranes, eg. of the sort that might be required for neurotransmission. That said, some non-excitable tissues also express these cell surface proteins, which hints at some other sort of function.
Second messenger regulated ion channels
These are also "ligand-gated" but the ligand is a second messenger of some sort, and the way to activate these channels pharmacologically is to influence the effector of that second messenger system. Classic examples include ion channels which open in response to intracellular calcium or inositol triphosphate (IP3). In case you were wondering, that's how you get a hen trachea to secrete chloride for you.
Voltage-gated ion channels
These channels are closed and undergo a conformational change when the transmembrane voltage difference reaches some threshold value. Of these those of greatest exam significance are probably the voltage-gated sodium channels, because altering their function is something which anaesthetists do constantly, and the CICM exam has been borrowing extensively from the ANZCA collection of past papers. The exam candidate will want to focus on sodium channels because they will at some stage or another be expected to explain their function, and possibly discuss the fact that drug binding to these targets can be state-selective (i.e. a drug may only bind to a channel in its closed or open state, but not both).
The wholistic post-exam intensivist may need to relinquish this sodium-centric worldview. Pharmacogenomic studies of the voltage-gated receptor protein family have yielded a massive tree of phylogenetic relationships (reproduced here shamelessly without permission but all to the greater glory of Yu & Catterall, 2004). In short, the voltage-gated sodium channels are only one half of a little branch of this tree, designated in blue on the diagram. They share that branch with voltage gated calcium channels. There are also voltage-gated calcium channels, hydrogen ion channels and potassium channels. In addition, numerous non-gated (eg. inward-rectifying or "funny" channels) are members of the same receptor family.
Broadly speaking, drug action on ion channels is immediate. Voltage or ligands can produce a conformation change which occurs over such timeframes as to make other seemingly instantaneous cellular molecular events appear lazy and sluggish. For instance, Chakrapani et al (2005) tried to establish the speed of acetylcholine receptor opening in the mouse neuromuscular junction, and found that 0.86 μs (microseconds) was approximately the maximum rate of conformational change (time it took for a pore to go from closed to open upon binding acetylcholine). In comparison, carbonic anhydrase (generally held to be the fastest-working enzyme in the body) takes around 10 μs to crack one molecule of CO2, although one can crank this up to 1 μs if one raises the pH to 9 and lowers the temperature to 25°C (Lindskog, 1997). The theoretical limit of protein folding speed (in case you were wondering) is 20 ns, or 0.02 μs (Bieri et al, 1999).
The origins of ion channels from an evolutionary perspective traverse biology and even proto-biology, entering into weird speculative realms. Eukarya, Bacteria and Archaea all share ion channels which have conserved domains, suggesting that sodium potassium and chloride channels were already well-established and diverse in the last common ancestor of all living things (Pohorille et al 2005; Moran et al, 2015).
These are receptors with seven transmembrane regions, which have their extracellular domain as the receptor. They are all made of seven -helix domains which stretch back and forth across the cell membrane, and this stitch-like structure is common to them all. In case you'd ever wanted to see what one looks like in detail, you're out of luck (the size of it stretches the capabilities of scanning EM) but here is a blurry zoom image from He et al (2017), showing off the 54x91 angstrom complex of a G-protein coupled receptor with rhodopsin, a light-sensitive receptor protein from the retina.
These receptors are known to medical students as receptors for various drugs, and so one might develop the impression that adrenergic neurotransmission is somehow their chief role. But in actual fact these are mainly sensory receptors. Of the 800+ G-protein coupled receptors the majority are in some way related to sense organs, and 50% of them (over 400) are involved specifically in olfaction. It would be fair to say that G-protein coupled receptors are olfactory receptors, of which some have been repurposed to perform neuroendocrine roles.
This sensory role dominance reveals to us the fact that these receptors were probably among the first chemical probes our ancestors used to perceive their environment. In fact these receptors are so ancient that they certainly pre-date the last common ancestor of all eucaryotes, who very likely already had a highly functional and complex system of G-protein coupled receptors. de Mendoza et al (2014) traced the genetics of GPCR classes through vertebrate vomeronasal receptors, insect odorant receptors, nematode chemoreceptors, to fungi chemoreceptors and plant cytokine signalling. It appears some of this sensory function remained so vital and fundamental to survival in the last three billion years that we still use these receptors for our sense of smell.
Anyway, to return to actual pharmacology. The G-protein coupled receptor is bound to a GTPase protein (hence the "G") which hydrolyses GTP (guanine triphosphate) into GDP. When bound to GTP, these proteins become activated, which then allows them to regulate the activity of second messenger systems and amplify the signal of receptor activation. Because some time is required for this secondary amplification, these receptors tend to function over timescales of seconds to minutes.
These are usually hormone receptors, and their role is generally to regulate gene transcription. These receptors, when activated, will bind directly to some sort of "response elements" in the promoter regions of their specific genes. They will do this when activated by a whole range of different endogenous ligands. Once it binds its ligand, the receptor will usually undergo some sort of conformational change which recruits other proteins into a huge multimeric complex, or instead destabilises and deactivates such a complex.
In either case, the onset time is clearly not going to be immediate. Either you have just activated genetic transcription which over hours will lead to synthesis of new proteins, or you have turned off such synthesis and now must wait for the proteins to be degraded. Ina any case, hours is the timeframe. To be precise, Becker (2013) offers 6-8 hours for the antiinflammatory effect of corticosteroids, which is likely representative.
Of the groups of intranuclear receptors, there are two main families: steroid and nonsteroid. The steroid receptors are generally present in the extranuclear cytoplasm where they are complexed with chaperone proteins; when an agonist binds them they are transported into the nucleus where they do all their work. Nonsteroid receptors are generally more properly intranuclear, in the sense that they are generally found inside the nucleus as heterodimers (bound with other intranuclear receptors and transcription factors).
Nuclear receptors are a relatively recent part of the cellular toolkit, in evolutionary terms. They appear to be quite specific for metazoans, and cannot be found among procaryotes or plants. Escriva et al (1998) found that much of their genetic diversification has already occurred before the vertebrate-arthropod split, with several groups of receptor which are unique to the vertebrate lineage.
In case anybody ever has any use for it, and also to demonstrate how deep this rabbithole goes, here is a list of all the receptor families - reproduced directly from guidetopharmacology.org, with links to their overview pages. For the time-poor exam candidate, to click "Show all receptor families" below may have an effect similar to the Total Perspective Vortex.