Schrieket al.,Science 375 , eabf7470 (2022) 11 February 2022 3 of 12
Fig. 2. MARCH1-
deficient mice
harbor MZ B cells
expressing cDC
surface proteins.
(A) Left: Representa-
tive flow cytometry
plots of CD24 versus
CD8 expression in
CD11c+wild-type and
March1–/–low-density
splenocytes. Right:
Frequencies and
total numbers of
CD11cintCD24+CD8int
cells. Graphs display
data pooled from
three independent
experiments, with each
symbol representing
an individual mouse
(n= 2 or 3 per experi-
ment); bars denote
mean ± SD. ****P<
0.0001 [independent-
samplesttest with
Welch’s correction
(no assumption
of equal variances),
two-tailedPvalue
(95% CI)]. (B) Flow
cytometry analysis
of the indicated surface
molecules on splenic
CD11cintCD24+CD8int
cells (blue), cDCs
(green), or B cells (red)
ofMarch1–/–mice.
Histograms are repre-
sentative of at least
two independent
experiments with
two or three individual
mice per experiment.
(CtoE) RNA-seq
of sort-purified
CD11cintCD24+CD8int
cells fromMarch1–/–
mice and B cells
and cDC1s from wild-
type andMarch1–/–
mice. (C) Top: Heat-
map showing 6510 dif-
ferentially expressed
genesacrossthe
five groups. Bottom:
Two-dimensional scaling
plot of the top 500 differentially expressed genes. (D) Barcode plots showing enrichment of genes in the B cell receptor signaling pathway (BioCarta) (left) and B cell
activation (GO:0042113) (right) based on gene set analysis comparingMarch1–/–CD11cintCD24+CD8intcells and cDC1s. (E) Reads per kilobase million (RPKM) for
genes encoding characteristic B cell (top) or cDC1 (bottom) surface markers inMarch1–/–cDC1s, B cells, and CD11cintCD24+CD8intcells. For RNA-seq analysis in (C) to
(E), mRNAs from sort-purified cells of four biological replicates with pooled spleens from five mice were sequenced in technical replicates. (F) Representative gating
strategy for the identification of splenic follicular (FO), MZ, and progenitor MZ (MZP) B cells, based on the surface expression of B220, CD93, IgM, CD21/35, and
CD23 (all gated on CD19+alive cells), in whole splenocyte preparations of wild-type andMarch1–/–mice (top and middle rows) or among the CD11cintCD24+CD8intcells
from low-density splenocytes ofMarch1–/–mice (bottom). Data are from at least two independent experiments with two or three individual mice per experiment.
A
CD11c+
43.41.7 22.324.6
CD11c+
****0102030010203040CD11cintCD24+CD8intCD24 CD24CD11c CD11cCD8 CD8FSC-A FSC-AWT March1−/−number (×105 )% of CD11c+March1−/−March1+/+March1−/−March1+/+B
MHC IIXCR1CD11bB220BST-2CD11cClec9ASirpαCD19Siglec HFLT3CD8CD4CD62LF4/80CD86IgMCD3CD54IgDLy6GDEC205C
-6 0 6logFC
B cell
cDC1
B cell
cDC1
March1−/−CD11cintCD24+CD8int6510 genesWT0.00.5-0.5-1.0
-2 -1 0 1 2 3 4
Leading logFC dim1Leading logFC dim2March1−/− B cellsMarch1−/− cDC1
WT B cellsWT cDC1March1−/− CD11cintCD24+CD8intE
FMO March1−/−B cells March1−/−CD11cintCD24+CD8int March1−/−cDC(1/2)
March1−/−pDC (BST2-, Siglec H), mac (F4/80), T cell (CD3), neutrophlil (Ly6G)cDC1cDC2B cellpDC,mac,
T cell,
neutrophilcDCEnrichment-10.72-2.18-0.88-0.43-0.170.030.210.410.671.317.4602.1B cell activationD
-9.00-2.19-0.89-0.43-0.170.030.200.400.671.307.29BCR signaling pathway01.9EnrichmentMarch1−/−CD11cintCD24+CD8intMarch1−/−cDC1March1−/−CD11cintCD24+CD8intMarch1−/−cDC1Cd19 Ighd Ighm Cr2RPKMRPKMCd8a Itgax Xcr1 Flt3020040060002004006008001000010002000300040005000010020030040050001002003000100200300050100150200050100150F
cDC1B cellCD11cintCD24+CD8int
cDC1B cellCD11cintCD24+CD8int
cDC1B cellCD11cintCD24+CD8int
cDC1B cellCD11cintCD24+CD8intcDC1B cellCD11cintCD24+CD8int
cDC1B cellCD11cintCD24+CD8int
cDC1B cellCD11cintCD24+CD8int
cDC1B cellCD11cintCD24+CD8intMarch1−/− B cellWT B cellMarch1−/− CD11cintCD24+CD8intCD21/35
CD93/AA4 IgM CD23B220 B220CD21/35
CD93/AA4 IgM CD23B220 B220CD21/35
CD93/AA4 IgM CD23B220 B220CD93+
83.2
CD93−
16.5CD93+
73.2
CD93−
26.5CD93+
94.3
CD93−
5.4783.8FO9.28MZFO
74.118.0MZFO
22.1MZ
72.763.8MZ MZP
34.4MZ
63.5 MZP34.1MZ
85.3
MZP
14.3RESEARCH | RESEARCH ARTICLE