Article
https://doi.org/10.1038/s41586-019-1367-0Somatic mutations and cell identity
linked by Genotyping of Transcriptomes
Anna S. Nam1,2,3,16, Kyu-tae Kim2,3,4,16 , ronan chaligne2,3,4,16, Franco izzo2,3,4, chelston Ang2,3,4, Justin taylor^5 ,
robert M. Myers2,3,4,6, Ghaith Abu-Zeinah4,7, ryan Brand2,3,4, Nathaniel D. Omans2,3,4,8, Alicia Alonso^9 , caroline Sheridan^9 ,
Marisa Mariani^9 , Xiaoguang Dai^10 , eoghan Harrington^10 , Alessandro Pastore^5 , Juan r. cubillos-ruiz^11 , Wayne tam^1 ,
ronald Hoffman^12 , raul rabadan^13 , Joseph M. Scandura3,7, Omar Abdel-Wahab^5 , Peter Smibert14,17 & Dan A. landau2,3,4,15,17*Defining the transcriptomic identity of malignant cells is challenging in the absence of surface markers that distinguish
cancer clones from one another, or from admixed non-neoplastic cells. To address this challenge, here we developed
Genotyping of Transcriptomes (GoT), a method to integrate genotyping with high-throughput droplet-based single-
cell RNA sequencing. We apply GoT to profile 38,290 CD34+ cells from patients with CALR-mutated myeloproliferative
neoplasms to study how somatic mutations corrupt the complex process of human haematopoiesis. High-resolution
mapping of malignant versus normal haematopoietic progenitors revealed an increasing fitness advantage with myeloid
differentiation of cells with mutated CALR. We identified the unfolded protein response as a predominant outcome of CALR
mutations, with a considerable dependency on cell identity, as well as upregulation of the NF-κB pathway specifically in
uncommitted stem cells. We further extended the GoT toolkit to genotype multiple targets and loci that are distant from
transcript ends. Together, these findings reveal that the transcriptional output of somatic mutations in myeloproliferative
neoplasms is dependent on the native cell identity.Somatic mutations underlie the development of clonal outgrowth,
malignant transformation^1 and subclonal diversification^2 –^5.
Nonetheless, clonally derived populations often lack cell-surface
markers that distinguish them from normal cells or that can help to
distinguish subclones, which limits our ability to link the clonal archi-
tecture of malignant populations with transcriptional read-outs. For
example, although myeloproliferative neoplasms result from recurrent
somatic mutations in CALR, JAK2 and MPL^6 ,^7 , the mutated clone often
represents a subset of bone marrow progenitors that lack distinctive
surface markers to distinguish them from non-neoplastic progen-
itors. Thus, we are unable to study the effect of myeloproliferative
neoplasm mutations in the context of the haematopoietic progenitor
subtype identity.
Although advanced methods have been developed to capture both
transcriptional information and genotype at the single-cell level^8 ,^9 ,
these methods often lack the throughput required to study complex
systems such as haematopoietic differentiation. Droplet-based sequenc-
ing enables the transcriptomic profiling of thousands of cells^10 ,^11 , and
can potentially also provide genotypic information regarding coding
mutations (Extended Data Fig. 1a). However, current methods—by
design—provide sequence information for only a short fragment at
the transcript end, which limits the ability of these techniques to jointly
genotype somatic mutations. To overcome this challenge, we devel-
oped GoT to link the genotyping of expressed genes to transcriptional
profiling of thousands of single cells. We applied GoT to CD34+ cells
from patients with myeloproliferative neoplasms, which revealed thatmyeloproliferative neoplasm mutations in haematopoietic progenitor
cells do not lead to uniform transcriptional outputs—but instead show
a strong dependence on the progenitor cell identity.Somatic genotyping in droplet scRNA-seq
To link genotypes to single-cell RNA sequencing (scRNA-seq)
in high-throughput droplet-based platforms, we modified the
10x Genomics platform to amplify the targeted transcript and locus
of interest (Fig. 1a, Extended Data Fig. 1b, c, Methods). We then
investigated amplicon reads for mutational status, and linked the
genotype to single-cell gene-expression profiles using shared cell bar-
codes (Extended Data Fig. 2a, b). We tested the ability of GoT to co-
map single-cell genotypes and transcriptomes in a species-mixing
experiment, in which mouse cells with a mutant human CALR trans-
gene were mixed with human cells with a wild-type human CALR
transgene^12 (Fig. 1b). Consistent with precision genotyping, the vast
majority of cells with transcripts aligned to the mouse genome showed
mutant CALR whereas cells with transcripts aligned to the human
genome showed wild-type CALR (96.7% of cells matched the expected
species) (Fig. 1b, Extended Data Fig. 2c–g).
CALR mutations have previously been demonstrated to activate
MPL (which results in megakaryocytic proliferation)^7 ,^12 –^16 , but how
the mutations perturb the early differentiation of haematopoietic
stem and progenitor cells (HSPCs) is largely unknown. We therefore
applied GoT to CD34+ bone marrow cells from five patients with
CALR-mutated essential thrombocythaemia, who had not been treated(^1) Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York, NY, USA. (^2) New York Genome Center, New York, NY, USA. (^3) Sandra and Edward Meyer Cancer Center, Weill
Cornell Medicine, New York, NY, USA.^4 Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medicine, New York, NY, USA.^5 Human Oncology and Pathogenesis
Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA.^6 Tri-Institutional MD-PhD Program, Weill Cornell Medicine, Rockefeller University, Memorial Sloan Kettering Cancer Center,
New York, NY, USA.^7 Richard T. Silver MD Myeloproliferative Neoplasms Center, Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medicine, New York, NY, USA.
(^8) Tri-Institutional Training Program in Computational Biology and Medicine, Memorial Sloan Kettering Cancer Center, Cornell University, Weill Cornell Medicine, New York, NY, USA. (^9) Epigenomics
Core Facility, Weill Cornell Medicine, New York, NY, USA.^10 Oxford Nanopore Technologies, New York, NY, USA.^11 Department of Obstetrics and Gynecology, Weill Cornell Medicine, New York, NY,
USA.^12 Division of Hematology and Medical Oncology, Department of Medicine, Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA.^13 Department of Systems
Biology, Columbia University Medical Center, New York, NY, USA.^14 Technology Innovation Lab, New York Genome Center, New York, NY, USA.^15 Institute for Computational Biomedicine, Weill
Cornell Medicine, New York, NY, USA.^16 These authors contributed equally: Anna S. Nam, Kyu-Tae Kim, Ronan Chaligne.^17 These authors jointly supervised this work: Peter Smibert, Dan A. Landau.
*e-mail: [email protected]
18 JUlY 2019 | VOl 571 | NAtUre | 355