HEMATOPOIESIS
Resistance to inflammation underlies enhanced
fitness in clonal hematopoiesis
S. Avagyan^1 †, J. E. Henninger^2 †, W. P. Mannherz^3 , M. Mistry^4 , J. Yoon^4 , S. Yang^5 , M. C. Weber^5 ,
J. L. Moore^5 ,L.I.Zon5,6*
Clonal hematopoiesis results from enhanced fitness of a mutant hematopoietic stem and progenitor cell
(HSPC), but how such clones expand is unclear. We developed a technique that combines mosaic mutagenesis
with color labeling of HSPCs to study how acquired mutations affect clonal fitness in a native environment.
Mutations in clonal hematopoiesisÐassociated genes such asasxl1promoted clonal dominance. Single-cell
transcriptional analysis revealed that mutations stimulated expression of proinflammatory genes in mature
myeloid cells and anti-inflammatory genes in progenitor cells of the mutant clone. Biallelic loss of one such
immunomodulator,nr4a1, abrogated the ability ofasxl1-mutant clones to establish clonal dominance. These
results support a model where clonal fitness of mutant clones is driven by enhanced resistance to
inflammatory signals from their mutant mature cell progeny.
C
lonal hematopoiesis reflects an imbal-
ance of hematopoietic stem and progen-
itor cell (HSPC) output. Imbalances may
result from mutations that enable com-
petitive outgrowth during aging, selec-
tive resistance to chemotherapy, or founder
effects in HSPC transplantation ( 1 ). Clonal he-
matopoiesis with recurrent mutations in genes
such asDNMT3A,TET2, andASXL1is asso-
ciated with increased risk for developing hema-
tologic malignancies ( 2 , 3 ). It is not known
how mutations confer competitive advantage
to stem cells. Although a number of gene mu-
tations induce clonal hematopoiesis in pre-
clinical models using transplantation assays
( 4 , 5 ), prospective clonal competition in a na-tive environment has not been studied. Here,
we used an endogenous color labeling system
of HSPC clones in zebrafish, called Zebrabow,
to study the effect of germline and acquired
mutations on stem cell clonality.
We previously used Zebrabow to evaluate
HSPC composition and found that cytotoxic
irradiation and transplantation reduced clonal
diversity ( 6 ). To determine whether clonal changes
could result from genetic alterations, we eval-
uated the clonal composition in a zebrafish
runx1germline mutant that generates fewer
HSPCs during development ( 7 ). Embryos from an
intercross ofrunx1W84X/+;Tg(ubi:Zebrabow-M)
andrunx1W84X/+;Tg(draculin(drl):CreERT2zebra-
fish were treated with 4-hydroxytamoxifen
(4-OHT) at 24 hours post-fertilization (hpf) to
induce color labeling of HSPCs, and clonal
composition was assessed in adult marrow768 5 NOVEMBER 2021¥VOL 374 ISSUE 6568 science.orgSCIENCE
(^1) Dana-Farber/Boston Children’s Cancer and Blood Disorders
Center, Boston, MA, USA.^2 Whitehead Institute for
Biomedical Research, Cambridge, MA, USA.^3 Harvard
Medical School, Boston, MA, USA.^4 Harvard Chan
Bioinformatics Core, Boston, MA, USA.^5 Boston Children’s
Hospital, Boston, MA, USA.^6 Howard Hughes Medical
Institute, Harvard Medical School, Boston, MA, USA.
*Corresponding author. Email: [email protected]
†These authors contributed equally to this work.
A
BC D
0
20
40
60
80
100
Cluster size (%)
Control
3
Mutant Control
0
0.2
0.4
0.6
0.8
1.0
Gini coefficient
Mutant
Control
Cluster size (%)
Mutant
0
20
40
60
0
20
40
60
B
R G
Grow to
adulthood
Serial blood and
tissue sampling
(3 and 7 mpf)
Clonal competition
Inducible color
labeling at 24 hpf
Marrow cell color analysis,
DNAseq, scRNAseq
(8 mpf) Sorting of
dominant clone
by FACS
DNAseq,
scRNAseq
CRISPR gRNAs
- cas9 mRNA
Somatic mosaic mutants
TWISTR: Tissue editing With Inducible Stem cell Tagging via Recombination
drl creERT2
ubi
Fig. 1. Mutations of clonal hematopoiesis induce clonal dominant states with
clonal skewing.(A) Schematic of TWISTR. gRNA, guide RNAs; hpf/mpf, hours/months
post-fertilization; DNA-seq, DNA sequencing; scRNA-seq, single-cell RNA sequencing;
FACS, fluorescence-activated cell sorting. (B) Example of Zebrabow analysis in control
and mutant zebrafish. Dot plots show cluster size of each color as percent of marrow
myelomonocytes. (C) Contribution of clusters in myelomonocytes of control (n= 107)
and mutant (n= 105) zebrafish. Dashed line marks three standard deviations of mean
cluster size of control zebrafish. P< 0.0001 (Mann-Whitney U test comparing the
largest clusters per zebrafish). (D) Gini coefficient in control (n= 107, median 0.4322)
and mutant (n= 105, median 0.5121) zebrafish. P< 0.0001 (Mann-Whitney U test).
RESEARCH | REPORTS