Nature | Vol 579 | 12 March 2020 | 299Data Fig. 1d), which suggests that βarr1 interacts with the 7TM core of
M2Rpp only in a lipid bilayer^15.
We optimized M2Rpp–βarr1 complexes for cryo-EM using radioli-
gand competition binding experiments and negative-stain electron
microscopy^16. The final complex—iperoxo/LY2119620-liganded M2Rpp
in MSP1D1E3 lipid nanodiscs, bound to βarr1 and the antibody fragment
Fab30—was monodisperse and displayed an affinity for the agonist that
was 350-fold greater than that of the receptor alone (Extended Data
Fig. 1e, f ). Fab30 was included to stabilize the βarr1–V2Rpp interaction
and to aid in cryo-EM particle alignment^7. We chose MSP1D1E3 nano-
discs with a diameter of 12 nm, because preliminary cryo-EM analysis
of an M2Rpp–βarr1 complex in smaller MSP1DH5 nanodiscs (around
9-nm diameter) indicated a mixture of ‘hanging’ conformations, in
which βarr1 interacted only with V2Rpp, and ‘core’ conformations, in
which βarr1 also engaged the 7TM bundle^17 (Extended Data Fig. 1g, h).
We previously observed such conformations in our electron micros-
copy analysis of complexes formed by βarr1 and the β 2 -adrenergic
receptor^18. Low-resolution cryo-EM maps of MSP1DH5 M2Rpp–βarr1
indicated conformational variability even among particles adopting
the core interaction, with βarr1 ‘rocking’ with respect to M2Rpp and
the βarr1 C-edge occasionally approaching the lipid bilayer (Extended
Data Fig. 1h). Given that hydrophobic residues in the C domain of visual
arrestin interact with the membrane^19 , we proposed that the extended
lipid surface of larger MSP1D1E3 nanodiscs could stabilize complexes
for studies at higher resolution.
In complexes in 12-nm nanodiscs, the interactions of βarr1 with the
receptor 7TM core and the lipid surface were more uniform (Extended
Data Fig. 1i). Nevertheless, variability in the size and tilt of the lipid
nanodisc relative to the βarr1–Fab30 density presented a challenge
for structure determination. Low-resolution analysis confirmed small
differences among particle classes in the angle of βarr1 relative to the
receptor 7TM, indicating inherent flexibility (Extended Data Fig. 2a).
Through three-dimensional (3D) classification we identified a class of
approximately 145,000 particle projections (17.4% of 831,443 particles
with a well-defined βarr1–Fab30 region) that showed solid density
for the 7TM portion in one position. Other 3D classes lacked defined
receptor density, possibly due to a combination of specimen distor-
tion at the air/water interface during cryo-sample preparation^20 and
potential projection misalignments due to the large nanodiscs. 3D
refinement of the well-defined class enabled us to obtain a 3D map of
the M2Rpp–βarr1 complex with a global indicated resolution of 4 Å
(Fig. 1d, Extended Data Fig. 2b). This map showed relatively lower reso-
lution in the 7TM region, including poor density of the highly mobile
transmembrane helix 1 (TM1), but nevertheless enabled a confident
comparison with M2R active-state structures. Subsequent refinement
focused on βarr1 and its interface with M2Rpp, and yielded a map of
3.6 Å resolution displaying well-resolved features, which were used for
model building in this region (Extended Data Fig. 3).Topography of the M2R–βarr1 complex
The cryo-EM structure of the M2Rpp–βarr1 complex in nanodiscs
reveals a multimodal interaction network, in which the N domain of
βarr1 engages the phosphorylated C terminus of the receptor, the inter-
domain loops of βarr1 engage the 7TM core and ICL2 of the receptor,
and the C domain of βarr1 engages the phospholipid bilayer (Fig. 2a–c).
Topologically, the central crest region interacting with the 7TM bun-
dle occupies a similar position to the Ras domain of G proteins. Com-
parison of the structure of M2Rpp–βarr1 with our recent structure of
the M2R–Go complex (PDB: 6OIK) reveals that the 7TM bundle of the
receptor adopts similar active-state configurations in both complexes,
characterized by the opening of the TM6 cytoplasmic portion^3 (Fig. 2d).
Relatively small shifts in the cytoplasmic ends of TM5, TM6 and TM7 are
probably required to accommodate βarr1, although we cannot exclude
effects arising from the use of lipidic (M2Rpp–βarr1) compared with
detergent (M2R–Go) environments. The similarity of the transducer-
bound conformations of M2R in the two structures is consistent with
the similar allosteric enhancement of agonist binding to M2R by βarr1TM2 TM1TM3TM4TM6
TM790ºM2RM2R–GoGo α 5
Rho–Arr1abc
d eȕarr1TM6TM7
TM1H8V2RppC domainN domainH8TM1TM4C domainN domain
V2Rpp
M2R (βarr1) Finger loopβarr1
nger loopTM1H8TM2TM5TM6ICL2 TM5TM6TM7M2R–ȕarr1C domainN domainGα RasTM5Lipid
bilayer TM1TM2TM3
TM6TM5 TM7ICL2
Finger loopN domainFinger loopH8V2RppC-edge
loopsTM5TM6
H8TM7TM1Middle loopC-loopV2RppFig. 2 | Structure of the M2Rpp–βarr1 complex.
a–c, Orthogonal views of the M2Rpp–βarr1
structure coloured by subunit (orange, M2Rpp;
teal, βarr1; grey sticks, model lipid bilayer).
d, Superposed structures of M2Rpp–βarr1 and
M 2 R– Go complexes (M2R, grey; Gαo Ras domain,
pink; PDB: 6OIK) aligned by M2R. Go βγ subunits
are omitted for clarity. The top right image shows
the cytoplasmic view of M2R transmembrane
alignment in the absence of transducers. The
bottom right inset shows the similar insertion
depth of the finger loop of βarr1 and the α5-helix
of Go into the TM bundle of M2R. e, Superposition
of M2Rpp–βarr1 and Rho–Arr1 structures
(rhodopsin, grey; arrestin1, green; PDB: 4ZWJ).
The inset shows an enlarged view of the finger
loops of βarr1 (M2Rpp) and visual arrestin
(rhodopsin).