The autonomic nervous system confuses me

The difference between anti-muscarinic and anti-cholinergic toxicity, organophosphate poisoning, whatever the hell atropine is doing, the concept of a “nicotinic AchR” — none of these made any sense to my poor brain as a medical student. On an anesthesiology rotation I sat down with one of my attendings at lunch and asked him to explain these things to me, not so I could write a disastrous blog post years later, but so that I could understand what was going on with the neo/glyco reversal combination. He spent 30 minutes talking to me. He drew multiple diagrams. I had no idea what was going on at the end of that lunch.

So, here goes nothing: the autonomic nervous system, and its receptors, for those of us desperately holding onto our last remaining synapse. Or neuron.

Some general anatomical tomfoolery

The autonomic nervous system is broadly described as having three parts: the sympathetic nervous system, the parasympathetic nervous system, and the enteric nervous system. The sympathetic nervous system corresponding roughly to a state of physiologic acceleration, and the parasympathetic to a state of physiologic deceleration. The enteric nervous system is its own strange beast, being arguably the oldest evolutionarily of all of our nervous systems, and scares me enough that I don’t particularly want to touch it today. My apologies to Julia Kaltschmidt, who certainly has no idea who I am. Thank the lord.

Ganglia shmanglia

The sympathetic and parasympathetic nervous systems consist of a set of pre-ganglionic neurons that run from the brain to a set of ganglia, and then synapse upon a set of post-ganglionic neurons that are made to do the dirty work of actually telling organs what to do. In this sense the ganglia can be thought of as signal integration centers, that parse “higher-order” commands from the brain into more concrete directions for particular organs. Somewhere out there, a neuroscientist has just collapsed from an aneurysm after the completion of that sentence.

Parasympathetic ganglia tend to lie close to their target organs; for example, there are named parasympathetic ganglia for several organs (ciliary ganglion for the eye, pterygopalatine ganglion for the lacrimal gland) and ganglia near their target organs that can be identified downstream of various plexuses. In contrast, sympathetic ganglia tend to either be within the spinal cord, with post-ganglionic fibers innervating multiple target organs, although there are exceptions to this rule. A quick look comparing the Wikipedia figures for the sympathetic and parasympathetic nervous systems may prove useful – preganglionic fibers are in red, while postganglionic fibers are in black. The sympathetic nervous system has more black stuff near target organs; the parasympathetic nervous system, more red stuff.

Neurotransmitters aplenty

This gives us 2 sets of synapses for the sympathetic and parasympathetic nervous systems: pre-ganglionic and post-ganglionic. Understanding these synapses clears up a lot of the anti-muscarinic anti-cholinergic mumbo jumbo that confused me throughout the litany of DUMBELS/SLUDGE mnemonics I endured in medical school.

Everything preganglionic uses acetylcholine (which was used for cell-cell communication since before the development of neurons themselves) – and, specifically, nicotinic acetylcholine receptors. They’re called nicotinic because (shocker) they bind nicotine, and they act as ligand-gated ion channels. Postganglionic parasympathetic synapses use muscarinic acetylcholine receptors (named because they bind muscarine more than nicontine), which act as GPCRs (more on that in a moment). Postganglionic sympathetic synapses mostly use norepinephrine as their signaling molecule, which binds $\alpha$ and $\beta$ receptors (which are GPCRs), but can also from time to time use acetycholine or dopamine.

These muscarinic receptors, in turn, have subtypes – 5 of them, to be precise. Subtypes M1, M3, and M5 use G${}_{\mathsf{q}}$ GPCRs, while M2 and M4 use G${}_{\mathsf{i}}$ receptors. M1 receptors are involved in analgesia (honestly not sure what they do), M2 receptors slow cardiac conduction and reduce inotropy, M3 receptors cause smooth muscle contraction/emesis/various glandular secretions, M4 receptors have some hematologic effects that are beyond me, and M5 receptors do some CNS shenanigans. Basically: M2 reduces chronotropy/inotropy via G${}_{\mathsf{i}}$ GPCRs, and M3 induces secretory things via G${}_{\mathsf{q}}$ GPCRs causing smooth muscle contraction.

On the sympathetic side, your sympathetic receptors are organized as follows: ${\alpha }_1$ receptors are G${}_{\mathsf{q}}$ GPCRs, ${\alpha }_2$ receptors are $G_{\mathsf{i}}$ GPCRs; all 3 $\beta$ receptors (${\beta }_1$, ${\beta }_2$, and ${\beta }_3$) are G${}_{\mathsf{s}}$ GPCRs, but ${\beta }_2$ and ${\beta }_3$ also couple with G${}_{\mathsf{i}}$. The effects of these are pretty aggressively pounded into anyone’s head on their ICU rotation, but for the sake of partial completeness: ${\alpha }_1$ causes vasoconstriction, ${\alpha }_2$ opposes general sympathetic tone, ${\beta }_1$ increases chronotropy/inotropy, and ${\beta }_2$ induces bronchodilation. ${\beta }_3$… we don’t talk about ${\beta }_3$. Something to do with the bladder and mirabegron.

Here’s a little ol’ table with all this from here¹1. Of course, in full disclosure: I hate tables, can never memorize them, but I suppose it’s useful as a reference.:

So, post-ganglionically-speaking, we have a bunch of scattered GPCRs triggered by Ach or norepinephrine. Why do we care?

GPCRs

It may be useful at this time to remind yourself (read: myself) how GPCRs work. GPCRs are receptors comprised of 7 $\alpha$-helices embedded in the cell membrane – the receptor bit – attached to a series of G proteins on the intracellular side. When a ligand (acetylcholine, norepinephrine, etc) binds to the receptor, this causes the G-protein complex to dissociate away from the receptor and do fun cell signaling things. Here’s a fun diagram of this from my old friend Addgene.

The downstream effect of G${}_{\mathsf{s}}$ proteins is to increase intracellular cAMP via activation of adenylyl cyclase; the effect of G${}_{\mathsf{i}}$ proteins is to decrease cAMP via inhibition of the same protein; and the effect of G${}_{\mathsf{q}}$ proteins is to dodge this whole thing and activate phospholipase C/IP${}_3$/DAG, in turn driving up intracellular Ca${}^{2+}$. There are many other types of G proteins; I do not study G proteins and thus this is where my knowledge of them will end, at least for now.

Why do we care about this in the context of sympathetic/parasympathetic stimulation? Well, it helps explain why things happen. For example: ${\alpha }_1$ receptors are G${}_{\mathsf{q}}$-coupled, so they increase intracellular Ca${}^{2+}\Rightarrow$ vasoconstriction. Same thing with M3 muscarinic receptors. On the other hand, ${\alpha }_2$ receptors are G${}_{\mathsf{i}}$-coupled, and so they inhibit the effects of all the $\beta$ receptors by reducing intracellular cAMP. And so on and so forth.

Toxicity

So, how does this help us (read: me) understand muscarinic vs cholinergic toxicity? Well, muscarinic toxicity can be thought of as overactivation of your M1-5 receptors: you get bradycardia, bronchoconstriction, and lots o’ secretions. Then, cholinergic toxicity is this plus the effects of the nicotinic receptors – neuromuscular junction shenanigans, mixed sympathetic/parasympathetic findings, CNS symptoms. Organophosphate poisoning is cholinergic because of inhibition of acetylcholinesterase; muscarinic toxicity is more often seen in muscarine overdose (lol), and pilocarpine/bethanechol (M3 agonists) toxicity.

From a treatment perspective, atropine is a competitive and reversible muscarinic antagonist – which means, on its own, it will not treat the full range of symptoms from cholinergic toxicity. In contrast, neostigmine is a acetylcholinesterase inhibitor that does NOT cross the blood-brain barrier (in contrast, physostigmine does enter the CNS; it lacks neo’s quaternary nitrogen atom). We then combine that with glycopyrrolate, a competitive muscarinic antagonist (like atropine!) to suppress the muscarinic toxicity while letting the neo help rebuild acetylcholine at the neuromuscular junction. We use glyco rather than atropine because a) atropine enters the CNS, which we’d like to avoid), and b) it lasts longer, so its kinetics are better matched to neostigmine. In contrast, atropine is a reasonable match for the alternative acetylcholinesterase inhibitor edrophonium.