Science - USA (2022-04-15)

(Maropa) #1

Capolupoet al.,Science 376 , eabh1623 (2022) 15 April 2022 6 of 12


AB

Cycle G1/S

Cycle G1

Cycle G2

Cycle M
Cycle M/G1

Middle

Basal

Inflammatory1

Vascular Fibrogenic1

Fibrogenic2

Contractile

Inflammatory2
Vascular
fibrogenic

Smooth Muscle

Fibrolytic2

Fibrolytic1
UMAP1

UMAP2

BasalCycle G1Cycle G1/SCycle G2Cycle MCycle M/G1MiddleContractileFibrogenic1Fibrogenic2Smooth MuscleVascularfibrogenicInflammatory1Inflammatory2Fibrolytic1Fibrolytic2Vascular
Basal
Cycle G1/SCycle G1
Cycle G2Cycle M
Cycle M/G1Middle
Contractile
Fibrogenic1Fibrogenic2
V. fibrogenicSm. Muscle
Immuno1Immuno2
Fibrolytic1
Fibrolytic2Vascular

Basal Proliferative Fibrogenic InflammatoryFibrolyticVascular

CAV1
S100A10
ADIRFMCM4MCM3HELLS
HIST1H4CHIST1H1B

RRM2KPNA2PLK1AURKACCNB1PTTG1CDC20TUBA1B
HNRNPA1
HNRN

PA2B1GTSE1UBE2CTOP2AGAPDHMYL9PFN1COL1A1
LBH
PS

AT1
CALD1CHRM2
COL12A1RSPO4OLFM2COL12A1
THBS2
COL12A1 P
PAP2BMXRA5SOD2CCL2TNFAIP2ISG15
MX1IFIT1
NEAT1CTSKMMP14MMP1SOD2STC1PTGIS
FN1FGF7

% of c
ells

expression

in
group

PROLIFERATIVE

BASAL

INFLAMMATORY FIBROGENIC

FIBROLYTIC

VASCULAR

Diffusion
map

PAGA
graph

DC1

DC3

DC2

C

D

H
ACTA2

LMNA

0 40

UMI counts

0 60

UMI counts

C
h
Tx

B

A
CT

A2

ACTA2

ShTxB

1 a

LM

N
A

LM

N
A

C
hT

xB

Sh

TxB2

e

IJ

E F

G

Triple+ ShTxB1a+/2e+

ChTxB+ ShTxB2e+

A
vg

. Z
-s
core


0.75

Signatures -0.75

IGF2
ACANINMT
HAPLN3ACVR2A
IGFBP7SFRP1
VCAM1MEG3
SERPINB2
KRTAP1-5NEAT1
ADAM33APP
MMP14TWIST2
CPS1
EFNA1DSG2
CCAT1KRT17
SLC7A2NPTX1
FAM83A
CELF4HAS2
CAPZA1GREB1
FAM76APEG3
B3GALT1CHIT1

C
hTxB/ShTxB1a

+

ChTxB

+

ShTxB2e

+

ShTxB1a

+/2e

+

Top Genes

RNA-seq
condition

Av
g. R
PM

0

Max

Avg. Z-score

% cells

| Z-score | > 0.2

Fibrogenic 1

V. fibrogenic

Fibrogenic 2
Sm. Muscle

Fibrolytic 1

Inflammatory 2

Fibrolytic 2

Contractile

Inflammatory 1

Basal

Vascular

Cycle M/G1
Middle

Cycle G1
Cycle G1/S
Cycle G2
Cycle M

-0.25

0

0.25

15%

30%

60%

Triple

+

ChTxB

+

ShTxB2e

+

ShTxB1a

+/2e

+

Signatures

States

C

hT

xB

+

Triple

Sh

T

xB2e

+

S

h
T
xB1a

+/2e

+

(^10080)
(^6040)
20
Fig. 4. Lipotype mapping to transcriptional cell states.(A) UMAP embedding
analysis of scRNA-seq of 5652 individual dHFs colored by the assigned cluster.
(B) Gene expression dot plot of cluster marker genes. Genes for each cluster were
identified using the Wilcoxon rank-sum test. (C) Diffusion map visualization of single
dHF cells from (A) highlighting the axes of transcriptional variation among the different
cell states. (D) PAGA applied to scRNA-seq data of control dHFs. Nodes indicate
cell type states, and the length of edges indicates the degree of similarity between
states, with shorter edges corresponding to greater state similarity. (E) Heatmap
reporting the average gene expression of enriched genes for each of the FACS-sorted
lipotype populations (bulk RNA-seq data). For each lipotype, the top eight genes,
ranked by fold change, are shown. (F) UMAP embedding colored by the different
lipotype gene signature scores. The 250 top differentially expressed genes were
used to calculate the signature score. (G) Dot plot colored by the average lipotype
z-score of cells of the different clusters. Size of the dots represents the number
of cells with magnitude of the score >0.35. (H) PAGA applied to scRNA-seq data of
control dHFs and based on the ShTxB2e+, ShTxB1a/2e+, ChTxB+, and triple+
lipotype signatures. Nodes are positioned corresponding to cell states in (D).
Color of nodes corresponds toz-score signatures, with a positivez-score (red)
indicating a greater correspondence of the particular cell type to the particular
lipotype state. Color bar is the same as in (G). (I) UMAP embedding of dHFs
colored by the expression of the two canonical markers for fibrogenic (ACTA2) and
basal (LMNA) cell states. (J) Confocal micrographs of cells stained with ShTxB1a
(green), ShTxB2e (red), and ChTxB (blue) and counterstained by antibodies
against ACTA2 and LMNA (magenta). Scale bar, 100mm.
RESEARCH | RESEARCH ARTICLE

Free download pdf