Structural biology

Muscarinic receptors become crystal clear

Muscarinic acetylcholine receptors mediate many physiological responses of the nervous system. Structures of two of these receptors yield insight into how they bind drugs and their mechanism of action. See Letters p.547 & 552

G-protein-coupled receptors (GPCRs) are the darling drug targets of many pharmaceutical and biotech companies. This largest superfamily of cell-membrane receptors affects many aspects of life, including mood and behaviour, the immune system and the senses. In this issue, Haga et al.1 and Kruse and colleagues2 describe the crystal structures of two GPCRs — the M2 and M3 muscarinic acetylcholine receptors, which belong to the same GPCR family but couple to different effector proteins. The results not only advance our understanding of the structure and molecular pharmacology of this receptor family, but also contribute to our knowledge of GPCRs and membrane proteins in general.

Muscarinic acetylcholine receptors (mAChRs) are expressed on most target organs of the autonomic branch of the peripheral nervous system, which controls unconscious physiological responses such as heart rate, digestion, respiration and urination. They are also expressed in the central nervous system, where they modulate circuits that control movement and contribute to processes such as learning and memory. Drugs that target these receptors are being used and/or tested for conditions that include abnormal heart rate, asthma, overactive bladder, Alzheimer's disease, Parkinson's disease and schizophrenia.

Mammals have five subtypes of mAChR (M1–M5), which are divided into two functional groups: M2 and M4 preferentially couple to the Gi family of G proteins, whereas M1, M3 and M5 couple to the Gq family. The receptors affect different aspects of body function. For instance, M2 decreases heart rate by controlling certain potassium-ion membrane channels, and M3 stimulates hormone secretion and relaxes airway smooth muscle. Understanding the intricate structural details of these receptors should help in the design of drugs that target specific mAChRs without producing undesirable side effects.

Solving crystal structures of GPCRs is notoriously difficult because of the proteins' natural flexibility. A trademark of these receptors is their seven transmembrane domains (TM1–TM7), which give rise to intracellular and extracellular loops. Of these, the third intracellular loop is particularly large and mobile. To solve the structures of M2 and M3, respectively, Haga et al. (page 547) and Kruse et al. (page 552) replaced this loop with phage T4 lysozyme, a protein that promotes crystal formation. As with other GPCRs previously crystallized by this approach, the modification did not alter the receptors' ability to bind agonist ligands (molecules that activate the receptors, such as the neurotransmitter acetylcholine) or antagonist ligands (molecules that block receptor activation).

Haga et al.1 describe the structure of M2 bound to the muscarinic blocker 3-quinuclidinyl benzilate. They report that the structure of inactive M2 is similar to that of other inactive GPCRs, particularly in the transmembrane domains. But M2 differs most from other GPCRs at its extracellular surface and in having a 33-ångström channel that contains the ligand binding pocket and extends beyond it. The ligand is oriented in the binding pocket by an aspartate amino-acid residue in TM3 and an asparagine residue in TM6. It also interacts with a lid formed by an 'aromatic cage' consisting of multiple tyrosine and tryptophan residues (located in TM3, TM6 and TM7). The authors found similar aromatic cages in three non-GPCR proteins that bind acetylcholine, which suggests that the aromatic cage is a common motif for binding this ligand.

Kruse et al.2 determine the structure of M3 bound to tiotropium — a bronchodilator and mAChR blocker. Overall, the structures of inactive M3 and M2 are similar. For instance, a characteristic of mAChRs seems to be an outward bend in TM4, which is not seen in other GPCRs. But the authors also identify a few notable differences in the structures of inactive M2 and M3. One is the presence of a phenylalanine residue (rather than, as in M2, a leucine) in the second extracellular loop of M3, which creates a space in the receptor's binding pocket. This small difference in the structure of the binding pocket may facilitate the development of drugs that have increased selectivity for a specific mAChR subtype. The relative position of TM7 in the two receptors also varies, possibly due to a difference in the TM2 amino acids with which TM7 interacts.

Another difference between M2 and M3 is in the position of TM5, especially at the cytoplasmic end of this domain. Specific TM6 residues that interact with TM5 at the cytoplasmic end determine the receptor's coupling selectivity for various G proteins. This difference may be a factor in the coupling selectivity of other GPCRs, as the TM5–TM6 distance in the M2 receptor is longer than that in M3 and similar to that in other Gi-coupled GPCRs, whereas in M3 this distance is similar to that in other Gq-coupled GPCRs (Fig. 1).

Figure 1: Differences between receptor subtypes.

G-protein-coupled receptors have seven transmembrane (TM) domains that span the cell membrane, giving rise to three intracellular loops (IL1–1L3) and three extracellular loops (EL1–EL3). Haga et al.1 and Kruse and colleagues2 report the crystal structures of two such receptors, M2 and M3, which are muscarinic acetylcholine receptors. The intracellular ends of TM5 and TM6 are farther apart in M2 (blue) and other Gi-coupled receptors than in M3 (red) and other Gq-coupled receptors. This and other structural differences between M2 and M3 may contribute to variations in the association and dissociation rates of drugs targeted to the two receptors.

Kruse et al. used molecular-dynamics simulations to investigate the binding of tiotropium to mAChRs. Although this blocker binds to the acetylcholine binding site, the simulations indicate that it pauses at a separate (allosteric) site during both association and dissociation from the mAChR. Tiotropium dissociation from M3 is slower than from M2, perhaps because the second extracellular loop in M3 is less mobile. Exploiting such differences in the extracellular surfaces of mAChRs may again contribute to the development of subtype-specific drugs, an endeavour that has previously been impeded by the close structural similarity of the ligand binding regions in the transmembrane core of mAChRs.

These latest advances inevitably raise further questions. For example, what are the differences in the receptors' structure on binding to antagonists, full agonists and 'biased' agonists (which elicit only a subset of physiological responses3)? Also, for the G-protein-interacting regions of GPCRs to be sufficiently ordered for crystallization, the receptor must be bound to its G protein4. Yet replacement of the third intracellular loop with the phage T4 lysozyme eliminates receptor coupling to G proteins. The conformational changes involved on binding to the G protein are therefore unclear. Crystallization of more intact mAChRs in complex with their cognate G proteins is thus required for detailed information about the pathways of receptor–G-protein coupling. These are some of the challenges we face in our attempts to better understand mAChRs.


  1. 1

    Haga, K. et al. Nature 482, 547–551 (2012).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Kruse, A. C. et al. Nature 482, 552–556 (2012).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Reiter, E., Ahn, S., Shukla, A. K. & Lefkowitz, R. J. Annu. Rev. Pharmacol. Toxicol. 52, 179–197 (2012).

    CAS  Article  Google Scholar 

  4. 4

    Rasmussen, S. G. F. et al. Nature 477, 549–555 (2011).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Rebecca L. Kow.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kow, R., Nathanson, N. Muscarinic receptors become crystal clear. Nature 482, 480–481 (2012).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing