Switchable annulation paths to diverse N-bridged bicyclic scaffolds via ligand-directed dicarbonylation
Article
https://doi.org/10.1038/s41467-026-74395-0
Switchable annulation paths to diverse
N-bridged bicyclic scaffolds via liganddirected dicarbonylation
Received: 20 January 2026
Check for updates
1234567890():,;
1234567890():,;
Accepted: 2 June 2026
Yu-Kun Liu1, Peng Yang1, Le-Cheng Wang1,2, Zhi-Peng Bao1,2 &
Xiao-Feng Wu 1,2
N-Bridged bicyclic scaffolds bearing 3-azabicyclo[3.2.0]heptane and 3-azabicyclo[3.1.1]heptane cores have emerged as privileged bioisosteres of piperidine and meta-substituted pyridine, offering enhanced conformational
rigidity and improved pharmacokinetic profiles. However, existing synthetic
approaches are typically restricted to single scaffolds, limiting scaffold diversity. Here, we report a ligand-controlled, palladium-catalyzed tandem dicarbonylation of readily available cyclobutenols and amines, enabling divergent
access to both 3-azabicyclo[3.2.0]heptane and 3-azabicyclo[3.1.1]heptane
scaffolds from the same substrate. This strategy overcomes key challenges
associated with strained-ring systems-including ring-opening side reactions
and complex regio-, stereo-, and diastereoselective control-through fine-tuning of the catalyst environment. The method demonstrates excellent selectivity, broad substrate compatibility, and operational simplicity,
accommodating 20 derivatives of bioactive molecules. This work provides a
modular and general platform for the synthesis of structurally distinct Nbridged heterobicycles, addressing a critical gap in the development of bioisosteric nitrogen-containing scaffolds.
Analysis of FDA-approved small-molecule drugs from 2013 to 2023
highlights the predominance of nitrogen-containing heterocyclesparticularly six-membered ones-which appear in the majority of newly
approved therapeutics1,2. Within this context, piperidine- and pyridinebased scaffolds are regarded as among the most prevalent
N-heterocyclic frameworks in pharmaceutical chemistry3–7. In recent
years, replacing piperidine and pyridine rings with bridged bicyclic
scaffolds in drug molecules and bioactive compounds has attracted
increasing attention (Fig. 1a)8–10, as this strategy is widely recognized to
block metabolically vulnerable sites and increase molecular rigidity,
thereby improving physicochemical properties and pharmacokinetic
profiles11–14. Notably, 3-azabicyclo[3.2.0]heptanes and piperidine both
adopt intrinsically three-dimensional architectures, whereas 3-azabicyclo[3.1.1]heptanes conform to the ‘escape-from-flatland’ design
principle. As a result, these scaffolds represent promising bioisosteres
of piperidine and meta-substituted pyridine, respectively
(Fig. 1b)9,10,15–17. Consequently, the development of modular and efficient strategies for the construction of structurally diverse and bioisosteric nitrogen-containing ring scaffolds from readily accessible
starting materials is of urgent importance.
3-Aza-BCHep scaffolds exhibit pronounced conformational
rigidity, making them inherently difficult to access through conventional alkylation-based methods18. Consequently, most synthetic
approaches toward 3-azabicyclo[3.2.0]heptane frameworks rely on
cycloaddition reactions, including (3 + 2)19,20and (2 + 2)21–28 processes
under metal, thermal, or photocatalytic conditions (Fig. 1c, left).
Nevertheless, these strategies are often constrained by the requirement for activated substrates and with limited dipolarophile scope. In
contrast, bicyclo[1.1.0]butanes (BCBs) have emerged as privileged
strain-release intermediates for constructing bridged bicyclic
1
Leibniz-Institut für Katalyse e. V., Rostock, Germany. 2Dalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics, Chinese Academy of
e-mail:
Sciences, Dalian, Liaoning, China.
Nature Communications | (2026)17:5130
1
Article
https://doi.org/10.1038/s41467-026-74395-0
a) Bioactive molecules containing 3-azabicyclo[3.2.0]heps and 3-azabicyclo[3.1.1]heps
F
F
Cl
O
F
F
O
N
CO2H
N
N
N
N
NH
O
H 2N
N
N
N
N
H
N
NMe
N
N
N
O O
S
Me
Belaperidone
Ecenofloxacin
antischizophrenia agent
antibacterial agent
N
H
KIF18A inhibitor
3D-analogue of Rupatadine
IC50 = 43 nM
b) Bioisosteres of piperidine and pyriding
125˚
125˚
Bioisosteres
3D
109˚
Bioisosteres
2D
Het
N
N
N
112˚
5.0 Å
4.8 Å
meta-Substituted Pyridines
flat aromatic ring
3-Azabicyclo[3.1.1]Heps
sp3-hybridized N atom
3D-shape
4.9 Å
Piperidine
3D saturated ring
3-Azabicyclo[3.2.0]Heps
rigidified 3D similarity
Escaping
from flatland
N
N
6-Membered
N-heterocycle
c)Synthetic approaches to 3-aza-bridged bicyclic frameworks and retrosynthetic analysis
+
N
(2+2)
Cycloaddition
Highly strained
four-membered ring
+
C1 + C1
N
(2+2)
Cycloaddition
N
O
Readily available
N-sources
O +
d) Pathways for multisite-selective dicarbonylation
1 2
3 4
1
3 4
4
2
CO
2
3
4
or
2
1
3 4
Subsequent
carbonylation
2
CO
and
3
3
1
4
1
2
Specific
dicarbonylation
3
1 2
4
CO
4
Subsequent
carbonylation
1 2
3 4
Initial
carbonylation
Common
intermediate
Multisite
Monocarbonylation
non-BCB routes are scarce
3
1
CO
or
Initial
carbonylation
X = C, N, O
LiAlH4
Then isomerization
e) Our concept: Divergent dicarbonylation via a common intermediate
1
2
CN
O
X
* N *
C1 source
+
H 2N
N
or
1,3-Dipolar
&
&
&
(3+3)
Cycloaddition
or
(3+2) Cycloaddition
Vinyl azides then ring expansion
N3
X
N
+
&
(3+2)
Cycloaddition
Multisite
Multiple
dicarbonylation
f) Ligand-controlled tandem dicarbonylation for divergent synthesis of 3-aza-bridged bicycles frameworks (This work)
O
O
N
O
Pd
CO
2
1
3 OH
4
Pd
NH2
+
NiXantphos
2,3-Cyclization
5-exo-trig
r.r. > 19:1, dr > 20:1
Broad coupling partners
> 100 examples; 20 bioactive derivatives
MePhos
2 or 2'
1
Readily available
Highly strained
N-sources
four-membered ring
N
O
1,3-Cyclization
6-endo-trig
r.r. > 19:1
Readily avaliable substrates
Divergent synthesis of 3-aza-Heps
Excellent regio- and diastereselectivity
Excellent functional group compatibility
Fig. 1 | The applications and preparation of N-bridged bicyclic scaffolds, and
site-selective dicarbonylation strategies for their synthesis. a Bioactive molecules containing 3-azabicyclo[3.2.0]heps and 3-azabicyclo[3.1.1]heps.
b Bioisosteres of piperidine and pyriding. c Synthetic approaches to 3-aza-bridged
Nature Communications | (2026)17:5130
CO
bicyclic frameworks and retrosynthetic analysis. d Pathways for multisite-selective
dicarbonylation. e Our concept: Divergent dicarbonylation via a common intermediate. f Ligand-controlled tandem dicarbonylation for divergent synthesis of 3aza-bridged bicycles frameworks (This work).
2
Article
frameworks29–40. Notably, (3 + 3) cycloadditions with 1,3-dipoles41–47
and dipolar (3 + 2) annulations of vinyl azides have enabled access to 3azabicyclo[3.1.1]heptane architectures48,49, the latter typically proceeding via subsequent isomerization (Fig. 1c, right). Beyond BCBbased approaches, only a few non-BCB strate (...truncated)