Cerebrovascular vulnerability and fibrosis in human brain aneurysms
nature neuroscience
Article
https://doi.org/10.1038/s41593-026-02326-9
Cerebrovascular vulnerability and fibrosis in
human brain aneurysms
Received: 26 August 2024
Accepted: 6 May 2026
Published online: xx xx xxxx
Check for updates
Jerry C. Wang 1,16, Chang N. Kim 1,2,3,4,16, Shubhang Bhalla1,
Lea Scherschinski5,6, Adnan Gopinadhan 1, Santhosh Arul1, Damian Sanchez1,
Tyler D. Schriber5,6, Amanda C. M. Apolonio7,8, Belda Gülsuyu1,
Muhammet M. Öztürk1, John P. Andrews 1, Joseph Kim1,
Behnam Rezai Jahromi 9, Mika Niemelä9, Martin Lehecka9, Aunoy Poddar1,
Thomas Wälchli10,11, Joshua S. Catapano5,6, Rajeev D. Sen12, Michael R. Levitt 12,
Daniel L. Cooke13, Kazim Narsinh13, Ruchira M. Jha 14, Tomoki Hashimoto5,6,
S. Paul Oh5,6, Eric J. Huang 15, Edward F. Chang 1, Daniel A. Lim1,2,
Adib A. Abla1, Andrew C. Yang 7,8, Tomasz J. Nowakowski 1,2,3,4,
Michael T. Lawton 5,6 & Ethan A. Winkler 1
Brain aneurysms are a cerebrovascular disease that results in a severe
type of stroke. The cell-specific molecular pathology underlying their
formation and rupture is unknown. Here we profile 227,663 neurovascular
cells, including 52,946 aneurysmal cells, from a total of 14 adult human
brain aneurysms and 11 control vessels. Our atlas of human brain
aneurysms, as well as cell-resolution spatial transcriptomics, revealed
that pathological cerebrovascular remodeling occurs with the loss
of structurally supportive smooth muscle cells and the emergence of
activated perivascular fibroblasts, which re-populate the vascular wall
and express multiple genes linked to aneurysm risk. Fibrotic changes
coincide with fibroblast–myeloid cell signaling pathways and an influx
of specialized macrophages that are rarely detected in non-aneurysmal
cerebrovasculature and that express destabilizing vascular cell programs.
Thus, we reveal an unrecognized interplay between cerebrovascular
fibrosis and myeloid inflammation during disease progression,
substantially advancing our understanding of the cellular drivers and
mechanisms underlying this devastating cerebrovascular disease that will
inform translational development.
Stroke is the second leading cause of death worldwide1,2. Aneurysmal
subarachnoid hemorrhage (aSAH) is an especially severe form of stroke,
third frequent after ischemic infarction and spontaneous intracerebral
hemorrhage. Most often, aSAH results from a saccular aneurysm, a
sac-like bulge in the vessel wall, at branch points of cerebral arteries
comprising the circle of Willis at the base of the brain3. Brain aneurysms
occur due to a loss of supportive structures, such as vascular smooth
muscle cells or extracellular matrix (ECM) proteins, which can be caused
A full list of affiliations appears at the end of the paper.
Nature Neuroscience
by genetic factors, environmental factors and blood flow patterns4–6.
Further weakening of the arterial walls can lead to rupture, resulting in
bleeding into the brain or the subarachnoid spaces. Molecular profiling
of samples from human brain aneurysms has revealed common pathways in extracellular matrix remodeling and inflammation involved
in the development of the disease7,8; however, the specific molecular
changes that occur in different cell types during the progression of
human brain aneurysms are not fully understood.
e-mail: ;
Article
https://doi.org/10.1038/s41593-026-02326-9
a
b
c
Condition
OL
Aneurysm
Control
Control
n=5
AC
EC
FbM
MG
pvMφ
aMφ
PICA
n=2
Circle of Willis
d
CLDN5
PODXL
CNN1
MYH11
SSTR2
IGFBP5
APOD
FBLN1
LUM
POSTN
FAP
LAMB1
PRR16
CD3D
CD8A
GNLY
CD79A
JCHAIN
FCER1A
CD1C
CCL3
MRC1
LYVE1
S100A9
S100A8
ACP5
MMP9
ADAM28
P2RY12
CX3CR1
AQP4
GFAP
KCNC2
SNAP25
PLP1
OLIG1
LHFPL3
pDC
NK
Control
n=3
e
Avg. Exp.
FB
aFB
SMC
TC
UMAP
MCA
n=2
cDC
Mo
Neu
mFB
BC
ACoA
n=3
UMAP
PCoA
n=1
OPC
CLDN5
(EC)
CNN1
(SMC)
IGFBP5
(FbM)
APOD
(FB)
POSTN
(aFB)
LAMB1
(mFB)
CD3D
(TC)
FCER1A
(cDC)
MRC1
(pvMφ)
S100A9
(Mo)
ACP5
(aMφ)
P2RY12
(MG)
2
1
0
% Exp.
UMAP
EC
SMC
FbM
FB
aFB
mFB
TC
NK
BC
pDC
cDC
pvMφ
Mo
aMφ
MG
AC
Neu
OL
OPC
0
25
50
75
Fig. 1 | Cell atlas of human brain aneurysms. a, Schematic showing location
and number of donors for samples profiled with single-cell or single-nucleus
sequencing. Inset shows circle of Willis. b, Uniform Manifold Approximation and
Projection (UMAP) of 100,409 transcriptomes from unruptured brain aneurysms
(n = 37,560 transcriptomes, 8 donors) and control cerebrovasculature (n = 62,849
transcriptomes, 8 donors) colored by condition. c, The same as b except colored
by cell type annotation. d, Dot plot showing expression of cell population marker
genes. e, Expression distribution of cell population marker genes projected on
UMAP embeddings from all transcriptomes in b and c. Gray, low expression;
orange–red, high expression; ACoA, anterior communicating artery; PCoA,
posterior communicating artery; MCA, middle cerebral artery; PICA, posterior
inferior cerebellar artery; EC, endothelial cell; SMC, smooth muscle cell; FbM,
fibromyocyte; FB, fibroblast; aFB, activated fibroblast; mFB, myofibroblast;
TC, T cell; NK, natural killer cell; BC, B cell; pDC, plasmacytoid dendritic cell; cDC,
conventional dendritic cell; pvMϕ, perivascular macrophage; Mo, monocyte; aMϕ,
APC5+ macrophage; MG, microglia; AC, astrocyte; Neu, neuron; OL, oligodendrocyte;
OPC, oligodendrocyte precursor cell; Avg. Exp., average (mean) expression. Panel a
created in BioRender; Winkler, E. https://BioRender.com/2atrfp2 (2026).
Single-cell and single-nucleus RNA sequencing (scRNA-seq and
snRNA-seq, respectively) has recently defined cell type composition
in the human cerebrovasculature and defined immune interactions
in arteriovenous malformations9–12. In this study, we used single-cell
and spatial transcriptomics to investigate the changes in cellular
composition and transcriptional programs that arise in human brain
aneurysms. Our results reveal a previously unrecognized role for activated perivascular fibroblasts and a specialized macrophage population in the progression of aneurysm pathology. This atlas of human
brain aneurysms and mechanistic investigations provide important
understanding of human brain aneurysms that can inform future
translational development.
limited to several milligrams of tissue, thereby challenging applications
of single-cell genomics. We leveraged expertise across four high-volume
treatment centers and performed scRNA-seq or snRNA-seq on unruptured brain aneurysm tissues isolated from eight patients undergoing
neurosurgical treatment (Supplementary Table 1). We used two control
tissues based on consensus recommendations14,15: surgically acquired
pial cortical vasculature (n = 5) and postmortem non-aneurysmal circle
of Willis arteries (n = 3). Fresh tissues were dissociated with established
techniques to perform scRNA-seq (n = 2 aneurysm and n = 5 control
donors)9. To extend our analyses to a larger cohort of aneurysms, we performed snRNA-seq on fresh frozen aneurysm tissues (n = 6 don (...truncated)
