heart morphogenesis, all in agreement with
anatomical annotations (figs. S14A and S15B).
Next, we sought to improve the clustering res-
olution by integrating data across H3K4me3
and H3K27ac histone marks at gene resolu-
tion. The granularity of 2D representation of
the data obtained from the integrated analysis
was further improved (fig. S14F). To assign cell
types to each cluster, we integrated the spatial-
CUT&Tag data (H3K4me3 and H3K27ac) with
the mouse embryo cell atlas from scRNA-seq
( 5 ) (fig. S14, G to K). For example, chondrocytes
and osteoblasts were mainly in the embryonic
facial prominence, and radial glia and inhib-
itory neuron progenitors were observed in
the forebrain (fig. S14, H and J). Although
H3K4me3 and H3K27ac had fewer clusters
than H3K27me3 at the 20-mm resolution, we
found that the clusters that appeared to be
homogeneous could be further deconvoluted
into subpopulations.
Finally, spatial-CUT&Tag with 20-mm pixel
size was applied to the P21 mouse brains, and
unsupervised clustering revealed distinct spa-
tial features (Fig. 4, A to C). We set out to
explore the spatial patterns of specific marker
genes to distinguish cell types and compared
them with the gene expression pattern in the
single-cell transcriptomic atlas ( 23 ) (Fig. 4,
D and E, and figs. S16 and S17). For exam-
ple,Sox10showed high GAS in cluster 2 of
H3K4me3, andItpr2had low CSS in cluster 6
of H3K27me3, indicating that these clusters
were enriched with oligodendrocyte lineage
cells. Cells of these clusters were particularly
enriched in a stripe-like structure that corre-
sponds to the corpus callosum (Fig. 4, D and
E). For cluster 3 of H3K4me3 and H3K27me3,
Adcy5was activated andRbms3was repressed,
suggesting that medium spiny neurons were
enriched in these clusters. Some clusters that
appeared to be homogeneous could be further
deconvoluted into subpopulations with dis-
tinct spatial distributions (Fig. 4F). For exam-
ple, cluster 2 of H3K27me3 could be further
divided into two clusters.Cux2,amarkerofthe
superficial cortical layers 2 and 3, had lower
H3K27me3 signal in subcluster 1. By contrast,
Blc11b, a marker of the deeper cortical layers 4
to 6 presented higher H3K27me3 signal. Al-
though Polycomb has been previously shown
to play a role in the establishment of cortical
layers at embryonic stages ( 24 , 25 ), our data
suggest that H3K27me3 is also involved in
maintaining cortical layer identities at post-
natal stages. To examine the interplay between
active and repressive marks and to infer the
potential H3K4me3/H3K27me3 bivalency, we
identified all active promoters specific to in-
dividual populations marked by H3K4me3 and
plotted the signals of H3K4me3 and H3K27me3
(fig. S18). As expected, H3K27me3 signals were
depleted when the promoter was enriched in
H3K4me3 in the respective population. How-ever, H3K27me3 signals were also observed
around a few marker genes in oligodendrocytes
and medium spiny neurons.
To further identify cell types, we integrated
spatial-CUT&Tag data with the published
scCUT&Tag ( 8 ) and scRNA-seq dataset ( 23 ).
This revealed that microglia, mature oligo-
dendrocytes, medium spiny neurons, astro-
cytes, and excitatory neurons were enriched
in clusters 1, 2, 3, 4, and 7, respectively, in
the H3K4me3 dataset, and subpopulations
of neurons could also be identified (Fig. 4, G
to J, and figs. S16 and S17). Moreover, the
integration of spatial-CUT&Tag with scRNA-
seq or scCUT&Tag could allow prediction of
theregioninwhichaspecificcelltypein
scRNA-seq or scCUT&Tag is localized (Fig. 4,
G to J). We identified that mature oligo-
dendrocytes (MOL1) are abundant in the
corpus callosum, whereas medium spiny neu-
rons (MSN2) are present in the striatum, and
TEGLU3 excitatory neurons are present in
deeper cortical layer 6, which is in agree-
ment with previous results ( 23 ) but was deter-
mined herein by epigenetic modification states.
TEGLU8 excitatory neurons have been shown
to populate cortical layer 4 ( 23 ), and indeed
we observed that the corresponding epige-
netic state of this neuronal population was
distributed in a more superficial cortical layer
than TEGLU3 (Fig. 4J). We found that a sub-
population of nonactivated microglia (MGL1)SCIENCEscience.org 11 FEBRUARY 2022¥VOL 375 ISSUE 6581 685
Fig. 4. Spatial mapping and integrative analysis of mouse brain.(A) Image
of mouse brain tissue section and the region of interest for spatial mapping.
(BandC) Unsupervised clustering analysis and spatial distribution of each
cluster (20-mm pixel size). (DandE) Spatial mapping of gene activity by
H3K4me3 and gene silencing by H3K27me3 modification for selected marker
genes. (F) Refined clustering identified subpopulations in neurons with distinct
spatial distributions and marker genes. (GandH) Integration of scCUT&Tag
( 8 ), scRNA-seq ( 23 ), and spatial-CUT&Tag data. (I) List of cell types in
scRNA-seq. (J) Spatial mapping of selected cell types identified by label
transferring. Scale bar, 500mm.RESEARCH | REPORTS