Single-Cell Analysis of the Multicellular Ecosystem in Viral Carcinogenesis by HTLV-1

Research Article

Junji Koya1, Yuki Saito1,2, Takuro Kameda3, Yasunori Kogure1, Mitsuhiro Yuasa1,4, Joji Nagasaki5, Marni B. McClure1, Sumito Shingaki1, Mariko Tabata1,6, Yuki Tahira3, Keiichi Akizuki3, Ayako Kamiunten3, Masaaki Sekine3, Kotaro Shide3, Yoko Kubuki3, Tomonori Hidaka3, Akira Kitanaka7, Nobuaki Nakano8, Atae Utsunomiya8, Yosuke Togashi5, Seishi Ogawa9, Kazuya Shimoda3, and Keisuke Kataoka1,10

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on October 1, 2021. Copyright 2021 American Copyright 2021 by AssociationAmerican for Association Cancer Research. for Cancer Research. abstract Premalignant clonal expansion of human T-cell leukemia virus type-1 (HTLV-1)– infected cells occurs before viral carcinogenesis. Here we characterize premalignant cells and the multicellular ecosystem in HTLV-1 infection with and without adult T-cell leukemia/ lymphoma (ATL) by genome sequencing and single-cell simultaneous transcriptome and T/B-cell receptor sequencing with surface protein analysis. We distinguish malignant phenotypes caused by HTLV-1 infection and leukemogenesis and dissect clonal evolution of malignant cells with different clinical behavior. Within HTLV-1–infected cells, a regulatory T-cell phenotype associates with premalignant clonal expansion. We also delineate differences between virus- and tumor-related changes in the nonmalignant hematopoietic pool, including tumor-specific myeloid propagation. In a newly generated conditional knockout mouse model recapitulating T-cell–restricted CD274 (encoding PD-L1) gene lesions found in ATL, we demonstrate that PD-L1 overexpressed by T cells is transferred to surrounding cells, leading to their PD-L1 upregulation. Our findings provide insights into clonal evolution and immune landscape of multistep virus carcinogenesis.

Significance: Our multimodal single-cell analyses comprehensively dissect the cellular and molecular alterations of the peripheral blood in HTLV-1 infection, with and without progression to leukemia. This study not only sheds light on premalignant clonal expansion in viral carcinogenesis, but also helps to devise novel diagnostic and therapeutic strategies for HTLV-1–related disorders.

INTRODUCTION Besides viral carcinogenesis, recent studies described somatic mutations in normal tissues, such as the hematopoietic and Oncogenic viral infections, such as human T-cell leukemia gastrointestinal systems, pointing to a critical role of prema- virus type-1 (HTLV-1), are responsible for 10% to 15% of human lignant clonal expansion in diverse oncogenic processes (5, cancers worldwide (1). These viruses can induce carcinogenesis 6). However, the biological properties and molecular basis by various mechanisms, such as chronic inflammation, cellu- of premalignant clonal expansion are not well characterized. lar transformation, and immune suppression (2). Accumulat- HTLV-1 is a retrovirus infecting approximately 5 to 20 million ing evidence suggests that clonal expansion of infected cells people worldwide, particularly in endemic regions (7). This precedes malignancy and contributes to tumorigenesis (3, 4). virus preferentially targets CD4+ T cells and is associated with a variety of diseases, ranging from inflammatory disorders to adult T-cell leukemia/lymphoma (ATL). ATL is an aggressive

1Division of Molecular Oncology, National Cancer Center Research Insti- T-cell neoplasm evolving after a long latency of >30 to 50 tute, Tokyo, Japan. 2Department of Gastroenterology, Keio University years. Even before ATL onset, premalignant HTLV-1–infected School of Medicine, Tokyo, Japan. 3Division of Hematology, Diabetes, and clones bearing driver mutations can be detected (4), offering Endocrinology, Department of Internal Medicine, Faculty of Medicine, Uni- a unique opportunity to functionally dissect premalignant 4 versity of Miyazaki, Miyazaki, Japan. Department of Pathology, Graduate clonal expansion in multistep viral carcinogenesis. However, School of Medicine, The University of Tokyo, Tokyo, Japan. 5Chiba Cancer Center, Research Institute, Chiba, Japan. 6Department of Urology, Graduate as only a small proportion (usually <1%–5%) of peripheral School of Medicine, The University of Tokyo, Tokyo, Japan. 7Department of blood mononuclear cells (PBMC) of asymptomatic carriers Laboratory Medicine, Kawasaki Medical School, Kurashiki, Japan. 8Depart- (AC) contain integrated proviral DNA (8), the phenotypical ment of Hematology, Imamura General Hospital, Kagoshima, Japan. and functional evaluation of HTLV-1–infected cells is difficult. 9Department of Pathology and Tumor Biology, Graduate School of Medicine, Kyoto University, Kyoto, Japan. 10Division of Hematology, Department of Efficient cytotoxic T-lymphocyte (CTL) response to HTLV-1 Medicine, Keio University School of Medicine, Tokyo, Japan. has been reported to limit the proviral load and the risk of Note: Supplementary data for this article are available at Blood Cancer associated diseases (9). By contrast, ATL patients are severely Discovery Online (https://bloodcancerdiscov.aacrjournals.org/). immunocompromised, frequently contracting opportunistic J. Koya and Y. Saito contributed equally to this article. infections (10). These observations emphasize the relevance K. Kataoka is the senior author of this article. of immune dynamics in HTLV-1–related diseases. This view Corresponding Author: Keisuke Kataoka, Division of Molecular Oncology, is supported by genetic studies showing frequent alterations National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo in immune molecules, such as structural variations (SV) trun- 104-0045, Japan. Phone: 81-3-3542-2511; Fax: 81-3-5565-0727; E-mail: cating the CD274 3′-untranslated regions (UTR) in ATL (11, [email protected] 12). However, although many efforts have been devoted to Blood Cancer Discov 2021;2:1–18 CTL characterization, the entire composition and functional doi: 10.1158/2643-3230.BCD-21-0044 heterogeneity of the immune microenvironment in HTLV-1 This open access article is distributed under the Creative Commons Attribution- infection and ATL remain elusive. NonCommercial-NoDerivatives License 4.0 International (CC BY-NC-ND). Recent advances in single-cell analysis provide an avenue to ©2021 The Authors; Published by the American Association for Cancer Research elucidate phenotypic features at a cellular resolution, allowing

SEPTEMBER 2021 blood CANCER DISCOVERY | OF2

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on October 1, 2021. Copyright 2021 American Association for Cancer Research. RESEARCH ARTICLE Koya et al. profound understanding of tumor heterogeneity and microen- ples were grouped into the same clusters (Fig. 1B and C). vironment in cancer (13). Particularly, cellular indexing of tran- By contrast, nonmalignant cells, including those from ACs scriptomes and epitopes by sequencing (CITE-seq) is a recently and HDs, were classified into the same clusters according to developed tool for exploring transcriptomic and surface prot- their lineage, suggesting their transcriptomic similarity among eomic states in the same single cell (14). Also, single-cell T/B- patients. Almost all malignant clusters expressed HTLV-1 HBZ cell receptor (TCR/BCR) sequencing (scTCR/BCR-seq) enables (Fig. 1C). Although almost no cells were classified into malig- detection of T/B-cell clonal expansion patterns through identi- nant clusters in ACs or HDs, malignant cells accounted for cal TCR/BCR sequence detection (15), which can be utilized to 15% to 99% (median 82%) of the total cells in ATL, which were track clonal evolution in lymphoid malignancy. lower in smoldering than in other subtypes (Supplementary Here, to investigate the cellular and molecular architecture in Fig. S1D and S1E). Single-cell mutation and CNA estimation HTLV-1 infection and ATL, we performed multimodal single-cell revealed somatic alterations almost exclusively in malignant analysis, evaluating transcriptome and many surface markers. clusters, consistent with the bulk sequencing and SNP array Combined with scTCR-seq, our approach deciphers functional results (Supplementary Fig. S1F and S1G), confirming success- heterogeneity of HTLV-1–infected premalignant cells and the ful separation of malignant from nonmalignant cells. hierarchy and clonal evolution of malignant cells. Simultaneous mRNA and protein assessment delineates discordant regula- Lineage-Specific Regulation of mRNA and tion of their expression levels, particularly in myeloid lineage. In ADT Levels addition, we reveal variegated alterations in the immune micro- Protein abundance is a more direct determinant of cellular environment, including a novel mode of PD-L1 upregulation in functions but cannot necessarily be inferred from mRNA surrounding cells by CD274 SV-harboring tumors. level due to posttranscriptional processes (16). CITE-seq analysis in healthy, virally infected, and leukemic states enables a RESULTS detailed comparison of mRNA and ADT level changes in various cell types. At the single-cell level, ADT was less prone to Landscape of the Multicellular Ecosystem in dropout than mRNA, conferring higher sensitivity for many HTLV-1 Infection and ATL surface markers (Supplementary Fig. S2A; Supplementary To delineate the multicellular ecosystem of HTLV-1 carrier Table S2). As expected, more than two thirds of genes showed state and ATL, we constructed transcriptomic, surface phe- strong positive correlations between mRNA and ADT levels notypic, and immune repertoire maps using droplet-based when averaging expression within clusters (Fig. 2A), sug- CITE-seq combined with TCR/BCR-seq at the single-cell gesting that overall expression of many surface markers is level. For CITE-seq analysis, we simultaneously generated 10x transcriptionally regulated. These included surface markers Genomics 5′ single-cell RNA-sequencing (scRNA-seq) and anti- showing a moderate to high expression in multiple cell types, body-derived tag sequencing (scADT-seq) libraries from 233,093 including ENTPD1/CD39 and CD28, for which mRNA and PBMCs obtained from 30 ATL patients (34 samples, including 4 ADT levels were highly correlated within and across lineages sequential ones), 11 HTLV-1 ACs, and 4 healthy donors (HD; (Fig. 2B; Supplementary Fig. S2B). Lineage-specific markers, Fig. 1A; Supplementary Table S1). Using targeted sequencing such as MS4A1/CD20 and CD14, were also observed, showing (targeted-seq), whole-exome sequencing (WES), and single- elevated expression of both mRNA and ADT levels in a single nucleotide polymorphism (SNP) array of bulk samples, cell type (Fig. 2C; Supplementary Fig. S2B). somatic mutations and/or copy-number alterations (CNA) However, there were important exceptions: Several surface were examined for all ATL samples, most (n = 32) of which markers, such as major histocompatibility complex (MHC) also underwent bulk RNA-seq (Supplementary Table S1). class I molecules and PTPRC/CD45, exhibited weak positive, no, Among the analyzed 102 ADTs, we removed 32 (31%) because or even negative correlations between mRNA and ADT levels of insufficient separation of known positive and negative (Fig. 2A and D). For these markers, some cell types displayed lineages and/or poor sensitivity (Supplementary Table S2). substantial upregulation of ADT levels despite low to moderate On average, 1,422 informative genes [4,477 unique molecular mRNA levels, whereas other cell types showed low ADT abun- indexes (UMI)] and 64 antibodies (2,745 UMIs) were detected dance irrespective of mRNA level. Among these markers, mye- per cell (Supplementary Table S2). TCR and BCR sequences loid cell–specific dissociated ADT elevation was observed for were detected in 85% and 96% of T and B cells, respectively. 11 markers, including CD47, CD69, and ITGB1/CD29, whereas These quality control parameters were comparable among CCR6/CD196 and CR2/CD21 showed dissociated ADT upregu- patient groups (Supplementary Fig. S1A). lation only in B cells (Fig. 2D; Supplementary Fig. S2B), sug- Dimensionality reduction analysis applied to scRNA-seq gesting lineage-specific posttranscriptional regulation. identified a total of 38 clusters, consisting of 30 malignant As such dissociated ADT upregulation was frequently and 8 nonmalignant ones as determined by TCR clonality observed between myeloid and NK cells, we compared the (Fig. 1B; Supplementary Fig. S1B; Supplementary Table S3). ratio of ADT to mRNA level between these cell types for Nonmalignant clusters consisted of two CD4+ T-cell clus- surface markers including those showing strong positive ters, two CD8+ T-cell clusters, two B-cell clusters, one natural correlations (Fig. 2E). This analysis identified an additional killer (NK)–cell cluster, and one myeloid cell cluster based on six markers, including ITGAL/CD11a and ITGB2/CD18, mRNA and ADT levels of canonical lineage markers (Supple- showing higher ADT levels in myeloid cells than NK cells mentary Fig. S1C). Within the malignant clusters, different with equal or higher mRNA levels (Fig. 2F; Supplementary ATL samples were segregated into different clusters, sugges- Fig. S2B), which included a lineage marker or functional pro- tive of interpatient heterogeneity, although sequential sam- tein in myeloid cells (17). The dissociated ADT upregulation

OF3 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

A Bulk analysis WES Cryopreserved Targeted-seq PBMCs RNA-seq Surface protein–derived DNA Surface proteins (ADT) Single-cell TotalSeq-C analysis Integrated analysis 102 antibodies mRNA-derived cDNA Whole transcriptome Multimodal single-cell data 233,093 cells + 3 isotype controls Genotyping information 30 ATL patients (34 samples) NNNNNAAAAAAA NNNNNAAAAAAA Clinical information 11 ACs (11 samples) TCR repertoire 4 HDs (4 samples) 10× chromium single cell 5′ library and gel bead kit BCR repertoire Functional assay Live cell sorting

B C All PBMCs (233,093 cells) Sample Cell type

NM07 NM06 10 10 NM08 10 0 0 M29 NM05 M09 M23 M10 NM02 −10 −10 M05 M25 0 M13 NM04 NM03 M26 M20 M28 M07 −100 10 −100 10

UMAP2 M21 NM01 M14 M06 M16 Disease subtype HBZ expression M18 M30 M02 M04 10 10 M12 M08 M27 M11 −10 M15 M01 0 0 M24 M22

M19 UMAP2 2 M17 M03 −10 −10 1 −10 010 UMAP1 0 −100 10 −100 10 Nonmalignant cluster Malignant cluster scTCR-seq scBCR-seq NM01 (CD4+ T) M01 (ATL02, ATL02−2) M16 (ATL22−2) NM02 (CD8+ T) M02 (ATL14, ATL16, ATL27) M17 (ATL20) NM03 (naïve CD8+ T) M03 (ATL09) M18 (ATL13) 10 10 NM04 (cytotoxic CD4+ T) M04 (ATL28) M19 (ATL07) NM05 (NK) M05 (ATL06, ATL06−2) M20 (ATL10) NM06 (IgD+ B) M06 (ATL23) M21 (ATL24) NM07 (IgD− B) M07 (ATL18) M22 (ATL15) 0 0 NM08 (myeloid) M08 (ATL25) M23 (ATL19) M09 (ATL08) M24 (ATL12) M10 (ATL01, ATL01−2) M25 (ATL05) M11 (ATL03) M26 (ATL17) −10 −10 M12 (ATL29) M27 (ATL26) M13 (cycling cells) M28 (ATL21) M14 (ATL30) M29 (ATL04) M15 (ATL11) M30 (ATL22, ATL22−2) −100 10 −100 10 UMAP1

Sample Disease subtype Cell type HD01 AC07 ATL04 ATL13 ATL22−2 Acute Malignant HD02 AC08 ATL05 ATL14 ATL23 Chronic CD4+ T HD03 AC09 ATL06 ATL15 ATL24 Smoldering CD8+ T HD04 AC10 ATL06−2 ATL16 ATL25 AC NK AC01 AC11 ATL07 ATL17 ATL26 HD B AC02 ATL01 ATL08 ATL18 ATL27 scTCR-seq Myeloid AC03 ATL01−2 ATL09 ATL19 ATL28 Monoclonal scBCR-seq AC04 ATL02 ATL10 ATL20 ATL29 Polyclonal Detected AC05 ATL02−2 ATL11 ATL21 ATL30 Undetected Undetected AC06 ATL03 ATL12 ATL22 Unevaluated Unevaluated

Figure 1. Entire landscape of single-cell CITE-seq and TCR/BCR-seq. A, Scheme of the overall study design. Targeted-seq, targeted sequencing; WES, whole-exome sequencing. B, UMAP plot of 233,093 cells from PBMCs of 4 HD, 11 AC, and 34 ATL samples; 30 malignant (M) and 8 nonmalignant (NM) clusters determined by TCR clonality are shown in different colors. C, UMAP plots (same as A) colored by sample, cell type, disease subtype, HBZ mRNA level, scTCR-seq, and scBCR-seq. See also Supplementary Fig. S1.

SEPTEMBER 2021 blood CANCER DISCOVERY | OF4

A Correlation between mRNA and ADT B Strong positive correlation C Across multiple cell types Single cell type 150 Pearson cor = 0.81 Pearson cor = 0.85 Pearson cor = 0.90 Pearson cor = 0.92 q = 1.8 × 10−58 q = 1.9 × 10−67 30 q = 1.8 × 10−85 q = 1.6 × 10−98 HLA−C/HLA-ABC ITGA6/CD49f 6 10 PTPRC/CD45 CCR6/CD196 40 HLA−B/HLA-ABC CR2/CD21 20 LAMP1/CD107a NCR1/CD335 4 CXCR3/CD183 SELL/CD62L CD28 CD20 CD14 CD69 ITGB7 CD58 CD39 5 20 100 2 10

Normalized ADT level 0 0 0 0

( q -value) B2M CD47 CCR5/CD195 0 0.25 0.5 0.75 01234 0510 15 0510 15

10 HLA−A/HLA-ABC HAVCR2/CD366 CX3CR1 ENTPD1 CD28 CD14 MS4A1 ITGB1/CD29 Normalized mRNA level −Log D 50 Weak positive, no, or negative correlation SIGLEC7/CD328 6 Pearson cor = −0.37 Pearson cor = −0.32 Pearson cor = −0.32 Pearson cor = −0.31 q = 3.9 × 10−7 q = 2.9 × 10−7 q = 5.6 × 10−7 q = 7.2 × 10−7 6 3 ITGAM/CD11b 2 CD244 4 CD99 4 2 10

ITGB2/CD18 B2M CD47 0 ITGAL/CD11a 2 CD45 1 HLA−ABC 2 1 −0.5 0 0.5 1 Pearson correlation coefficient 0 0 0 0 Weak positive, no, or negative correlation 0 100 200 0255075 0510 15 024 Myeloid specific B specific Others B2M HLA−C PTPRC CD47 Strong positive correlation B specific Myeloid specific Pearson cor = −0.13 Pearson cor = −0.033 15 Pearson cor = 0.29 6 Pearson cor = 0.33 E q = 0.045 q = 0.68 q = 4.2 × 10−6 q = 1.7 × 10−7 Up in NK Up in myeloid 20 15 ADT/mRNA ratio 10 25 15 4 ITGAM/CD11b 10 10 CD69 CD21 HLA−C/HLA-ABC CD29 5 CD196 2 PTPRC/CD45 CD47 LAMP1/CD107a 5 5 ITGAL/CD11a 20 HLA−A/HLA-ABC Normalized ADT level 0 0 0 0 ITGB2/CD18 CD244 CD69 HLA−B/HLA-ABC B2M 0 36912 0510 15 20 25 0123 00.2 0.40.6 0.8 ITGB7 CD69 ITGB1 CR2 CCR6 SIGLEC7/CD328 CD99 ITGB1/CD29 Normalized mRNA level 15 CX3CR1 HAVCR2/CR366 F Myeloid-specific regulation in strong positive correaltion -value)

( q CD58 Pearson cor = 0.44 Pearson cor = 0.45 Pearson cor = 0.52 Pearson cor = 0.57 10 q = 1.0 × 10−12 q = 3.8 × 10−13 q = 2.8 × 10−18 150 q = 5.2 × 10−22 10 6 7.5 6 −Log 100 4 5 4 CD18 CD11b CD11a

CD244 50 5 2 2.5 2

Normalized ADT level 0 0 0 0 01230510 15 00.5 1 012 0 ITGAL ITGB2 CD244 ITGAM Normalized mRNA level 0 1234 + + Log2 (FC of ADT/mRNA ratio between myeloid and NK) Cell type Malignant CD4 T CD8 T NK B Myeloid Patient group HD AC ATL

Figure 2. Comparison of mRNA and ADT levels across cell types. A, Correlation coefficients and their significance for normalized mRNA and ADT levels. Genes with weak positive, no, or negative correlations (q ≥ 1 × 10−10) were depicted separately. Dots are colored if genes are under lineage-specific posttranscriptional regulation; otherwise, they are colored gray. The inset (dotted box) represents an enlarged view of the solid box. B–D, mRNA and ADT levels for representative genes showing a strong positive correlation across multiple cell types (B) and in a single cell type (C), and a weak positive, no, or negative correlation (D), with shapes and colors indicating patient group and cell type. E, Volcano plot comparing the ratio of ADT to mRNA levels between myeloid (NM08) and NK (NM05) cells. Fold change (FC) was calculated as the ADT/mRNA ratio in myeloid cells divided by that in NK cells when −10 the average mRNA level was ≥0.2 in both cell types. ADTs with q < 10 and log2(FC) ≥ 1 were considered significant and colored by correlation significance. Mann–Whitney U test. F, mRNA and ADT levels of representative genes with dissociated ADT upregulation in myeloid cells. B–D and F, Each dot represents an average expression value for each cell type (≥30 cells) in each sample. Pearson correlation (cor) test with Benjamini–Hochberg correction. Fitted linear regression line is shown. See also Supplementary Fig. S2. in myeloid cells was validated using other public CITE-seq T-cell clusters, as distinguished by TCR clonality. Within the data sets (Supplementary Fig. S2C–S2E). Our approach sys- nonmalignant CD4+ T-cell cluster (NM01, 35,668 cells), we tematically delineated lineage-specific regulation of surface identified four subclusters. Subclusters CD4-3 and -4 were marker expression, demonstrating the involvement of post- characterized by HBZ, FOXP3, and CADM1 mRNA levels and transcriptional regulatory mechanisms, operating particu- the CD45RO+CD25+CD7−CD194+ immunophenotype, sug- larly in myeloid cells. gestive of the regulatory T-cell (Treg)–like features (Fig. 3A and B; Supplementary Fig. S3A–S3D; Supplementary Table S4). Phenotypic and Transcriptomic Characterization We therefore referred to these subclusters as HTLV-1–infected/ + of Malignant and Nonmalignant CD4 T Cells Tregs, as previously reported (18). Subcluster CD4-1 and CD4-2 Then, we comprehensively investigated phenotypic and showed typical naïve CD45RA+CD7+CD62L+ and effector/ transcriptomic features of malignant and nonmalignant CD4+ memory (EM) CD45RO+CD127+ immunophenotypes,

OF5 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

A B C D Nonmalignant CD4+ T + HBZ mRNA Nonmalignant CD4 T (NM01) Naïve EM Infected/Treg *** ** 0.5 5 CD4-4 5 Infected/Treg2 0.5 0.6 * * * ** CD4-2 0.4 EM 0 *** 0.4 0.3 0 CD4-1 0 Naïve

UMAP2 0.2 UMAP2 CD4-3 0.2 Infected/Treg1 0.1 –5 –5 0 0 −4 048 −4 048 HD AC ATL HD AC ATL HD AC ATL HD AC ATL UMAP1 UMAP1 Fraction of cells in nonmalignant Fraction of cells in nonmalignant

E F G Nonmalignant CD4+ T-cell clones *** RNA velocity of nonmalignant CD4+ T-cell clusters HD AC ATL 5 *** 1 HD ex AC 0.75 ATL *** *** *** 0 0.5 ** UMAP2 0.25

−5 ind STARTRAC−expa 0 −4 048 Naïve EM Infected/Treg Naïve EM UMAP1 Infected/TregMalignant

HI J K mRNA (681) Up in malignant (681) Up in uninfected (91) Up in malignant (21) Up in uninfected (6) Malignant > infected/Treg PD-L1 CD73 Malignant > uninfected CST7 1 (176) LAG3 CD194 HLA-DR CD28 LGALS1 CD74

) CD39 ITGB1 HLA-DRA 1 ANXA2 CD278 0 2 CD71 CD86 351 HLA-DRB1 CD99 CD38 S100A11 CD25 CD99 PD-1 TIGIT CD69 ISG15 0 −1 CD127 CADM1 CD99

0 PELI1 (FC of ADT level

(FC of mRNA level) 154 TIGIT 2

2 Malignant = infected/Treg COTL1 −1 Activated CD3 46 130

Lo g Stimulatory CD45RA (505) Lo g 13 Inhibitory 1 Top 5 immune-related pathways 24

CD7 erage normalized level −2 Treg related

Infected/Treg > uninfected Av 0 200 400 600 800 01020 Malignant > infected/Treg 0 Rank Rank • Hypergeometric test • and • P = 2.8 × 10−273 −1 • and • P = 1.8 × 10−247 LGALS1 • and • P = 4.2 × 10−48 Naïve EM Infected/TregMalignant

Figure 3. Cellular and molecular features of malignant and nonmalignant CD4+ T cells. A, Subclustering of nonmalignant CD4+ T cells (NM01). B, Normal- ized HBZ mRNA level (after ALRA imputation) on UMAP plots in A. C and D, The fraction of total nonmalignant CD4+ T cells (NM01; C) and naïve (CD4-1), EM (CD4-2), and infected/Treg (CD4-3 and -4) CD4+ T cells (D) in nonmalignant cells (NM01–08) for each sample. E, T-cell clones (present in ≥5 cells) on UMAP plots in A. Each color represents a distinct TCR clonotype. F, Clonal expansion levels of each CD4+ T cluster (≥30 cells) for each sample. G, Steady- + state RNA velocity of nonmalignant CD4 T clusters from HD, AC, and ATL samples. H and I, Differentially expressed mRNAs [q < 0.01 and log2(FC) > 0.4; H] and ADTs [q < 0.01 and log2(FC) > 0.3; I] between uninfected (CD4-1 and -2) and malignant (M01–30) clusters. FC, fold change. J, Overlap of upregulated mRNAs among comparisons between uninfected (CD4-1 and -2), infected/Treg (CD4-3 and -4), and malignant (M01–30) clusters. Hypergeometric test. K, Gene-expression patterns among CD4+ T clusters with higher (top) or comparable (bottom) mRNA levels in malignant (M01–30) compared with infected/ Treg (CD4-3 and -4) clusters among those upregulated in malignant (M01–30) compared with uninfected (CD4-1 and -2) clusters. CD99 and LGALS1 are shown in red. (continued on next page) respectively, together with corresponding mRNA levels (Sup- Among 23,457 uninfected CD4+ T cells (CD4-1 and -2) plementary Fig. S3B and S3C; Supplementary Table S4). In expressing at least one TCR (either α or β chain), 63% of cells total, about one third of nonmalignant CD4+ T cells (NM01) expressed paired productive TCR αβ chains, whereas only were classified into HTLV-1–infected/Treg clusters in ATL one or more than two TCRs were detected in 28% and 8% of and ACs, but only 4% in HDs, which were found to be HBZ cells, respectively (Supplementary Fig. S3G; Supplementary negative (Fig. 3C and D; Supplementary Fig. S3A and S3E). Table S5), consistent with previous studies (19, 20). Clonal These results suggest that the HTLV-1–infected/Treg clusters analysis by the single T-cell analysis by RNA-seq and TCR consisted mainly of HTLV-1–infected cells, although a small tracking-expansion (STARTRAC-expa) index (15) revealed number of normal Tregs were also present. The fraction of almost exclusive clonal expansions in HTLV-1–infected/Treg HTLV-1–infected/Tregs was highly correlated with HTLV-1 clusters (Fig. 3E and F; Supplementary Table S5), indicating proviral load in ACs (Supplementary Fig. S3F), confirming the a clear biological disparity between HTLV-1–infected and reliability of our estimation. uninfected cells. The STARTRAC-expa index was proportional

SEPTEMBER 2021 blood CANCER DISCOVERY | OF6

LMPQR (kD) Up in infected/Treg (224) Up in uninfected (40) Up in infected/Treg (10) Up in uninfected (6) ×105 TL-Om1 cells TL-Om1 cells LGALS1 HLA-DRB1 CD73 100 p-AKT 60 2 • Non-target * HLA-DRB5 HLA-DR 4 • CD74 sgCD99-1 AKT 60 HLA-DRA 1 CD194 * HLA-DPA1 sgCD99-2 75 1 CD99 CD25 3 ** 44 HLA-DPB1 TIGIT p-ERK1/2 –M (%) 42 CD58 2 HLA-DQB1 44 CD39 50 ERK1/2 0 0 2 • 42 CD3 *** (FC of ADT level)

(FC of mRNA level) CD127 2 CD27 p-p38 38 2 GO_ANTIGEN_ Activated 25 Cells in S–G

−1 Log PROCESSING_AND_ Stimulatory Absolute cell number

Log CD62L 1 • (n = 4) (n = 4) p38 38 PRESENTATION_OF_ Inhibitory −1 Treg related CD45RA PEPTIDE_ANTIGEN CD7 • β-Actin 45 0 0 100 200 51015 0724 48 2 -1 -2 -1 -2 Rank Rank Hours of culture CD99 CD99 CD99 CD99 Non-targetsg sg Non-targetsg sg NO ST Up in malignant (189) Up in infected/Treg (45) Up in malignant (13) Up in infected/Treg (4) Effector: PBMCs Target: BALL-1 cells Murine E.G7-OVA cells GNLY GO_LEUKOCYTE_ PD-L1 50 Mock 150 *** CELL_CELL_ADHESION CD38 CD3scFv *** * ** * 2 LAG3 GO_REGULATION_OF_ 1 40 CD3scFv/LGALS1 SELL T_CELL_ACTIVATION CD71 *** 100 CD27 CD278 CD99 1 CD28 30 PTPN6 CD25 CD86 CD69 PD-1 CTLA4 20 0 (FC of ADT level) 50 (FC of mRNA level) 0 2 2 Activated CD3 10 Tumor size (mm³) Cytotoxic activity (%) Cytotoxic Lo g

Lo g Stimulatory −1 Inhibitory Treg related CD7 0 (n = 3) 0 (n = 15) (n = 17) 0 100 200 51015 E/T ratio 0102050 Mock LGALS1 Rank Rank

Figure 3. (Continued) L–O, Differentially expressed mRNAs (L and N) and ADTs (M and O) between uninfected (CD4-1 and -2) and infected/Treg (CD4-3 and -4; L and M) clusters and between infected/Treg (CD4-3 and -4) and malignant (M01–30; N and O) clusters. H, L, and N, Red color represents upregulated mRNAs related to indicated pathways. I, M, and O, Different colors represent different sets of T-cell–related ADTs. Treg-related ADTs are shown in orange. H, I, and K–O, Numbers of significant mRNAs or ADTs are shown in parentheses.P and Q, Growth curve (mean ± SD; P) and the frequency of cells in S–G2–M phase (Q) for TL-Om1 cells transduced with indicated single-guide RNAs (sgRNA; n = 4). R, Phospho (p)-AKT, AKT, p-ERK1/2, ERK1/2, p-p38, p38, and β-Actin (loading control) expressions for the same TL-Om1 cells. S, Cytotoxic activity of PBMCs (effector: E) against BALL-1 cells transduced with indicated plasmids (target: T) at various E/T ratios (n = 3). CD3scFV, anti-CD3 single-chain variable fragment. T, Tumor volume of mock- or LGALS1-transduced E.G7-OVA tumors at 6 days after transplantation (n = 15–17). C, D, F, P, Q, S, and T, Welch t test. ***, P < 0.0005; **, P < 0.005; *, P < 0.05. See also Supplementary Figs. S3 and S4.

to the fraction of HTLV-1–infected/Tregs in ACs (Supplemen- CD25, and CD38) in malignant cells (Fig. 3I; Supplemen- tary Fig. S3F), suggesting that HTLV-1–infected cell expansion tary Table S6). Besides these reported molecules, malignant can be attributed to clonal proliferation. Overall, the propor- cells also exhibited increased expression of stimulatory mol- tion of CD4+ T cells among nonmalignant cells was elevated ecules (CD99, CD28, and CD278). These molecules were in ACs (Fig. 3C). Interestingly, uninfected CD4+ T cells, includ- frequently coexpressed in malignant cells (Supplementary ing naïve and EM cells, were increased in ACs compared Fig. S4C), suggesting their synergistic function. By contrast, with HDs (Fig. 3D), suggesting that HTLV-1 infection affects naïve T-cell markers, such as CD45RA and CD7, were down- even uninfected cells. RNA velocity and pseudotime trajectory regulated in malignant cells as reported (18). Interestingly, analyses (21) showed rapid differentiation from naïve to EM among CD4+ T-cell clusters, despite their mRNA upregula- only in ACs (Fig. 3G; Supplementary Fig. S3H). These findings tion, malignant cells exhibited decreased CD3 and CD62L suggest that HTLV-1 infection not only intrinsically but also ADT abundance (Supplementary Fig. S4D and S4E), sug- extrinsically regulates T-cell proliferation. gesting their tumor-specific posttranscriptional regulatory Next, we investigated the similarities and differences of mechanism. mRNA and ADT levels among uninfected, infected/Treg, Interestingly, many differentially expressed mRNAs and malignant clusters. First, by comparing with unin- showed sequential expression changes from uninfected to fected cells (CD4-1 and -2), we identified 681 upregulated malignant cells (Fig. 3J and K; Supplementary Fig. S4F and and 91 downregulated genes in malignant cells (M01–30; S4G). Among the upregulated genes, 176 genes showed Fig. 3H; Supplementary Table S6). These malignant signa- higher expression in malignant (M01–30) than infected/ ture gene expressions were more strongly altered in aggres- Treg clusters (CD4-3 and -4), whereas the remaining 505 sive than in indolent subtypes (Supplementary Fig. S4A gene levels were comparable (Fig. 3J and K). A substantial and S4B). Functional enrichment analysis showed upregu- proportion of these genes were upregulated in infected/Treg lation of various immune-related pathways. Comparison clusters (CD4-3 and -4) and remained high or even increased of ADT levels revealed co-upregulation of a wide range of (154 and 46 genes) in malignant cells (M01–30), suggesting immune-related molecules, including inhibitory molecules that ATL cells acquire additional malignant phenotype dur- (PD-L1, CD73, and CD39) and activation markers (CD71, ing leukemogenesis while maintaining the transcriptomic

OF7 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on October 1, 2021. Copyright 2021 American Association for Cancer Research. Multicellular Ecosystems in Viral Carcinogenesis by HTLV-1 RESEARCH ARTICLE features of HTLV-1–infected cells. Comparison between non- Functional Heterogeneity of HTLV-1–Infected malignant clusters identified 224 upregulated and 40 down- Nonmalignant and Malignant CD4+ T Cells regulated genes in infected/Tregs (CD4-3 and -4) relative to uninfected cells (CD4-1 and -2), most of which remained To evaluate the functional heterogeneity of HTLV-1– high and low in malignant cells, respectively (Fig. 3L; Sup- infected T cells, we performed subclustering of these cells, plementary Table S6). Interestingly, MHC class II pathways identifying three FOXP3+ clusters and one FOXP3− cluster were enriched in the upregulated genes, suggesting, together (Fig. 4A and B; Supplementary Fig. S5A; Supplementary with higher HLA-DR ADT level in infected/Tregs (Fig. 3M), Table S4). Although all samples contained a varying degree of that MHC class II upregulation may contribute to HTLV-1– both FOXP3+ and FOXP3− HTLV-1–infected cells, the propor- infected cell proliferation by stimulating T–T interaction tion of FOXP3+ cells was elevated in ATL patients compared and intracellular signaling (22). Importantly, some differ- with ACs and was higher in aggressive versus indolent sub- entially expressed mRNAs and ADTs, including CD73 and types (Fig. 4C and D). FOXP3+ cells displayed a higher degree CD99, showed a better discriminatory power between unin- of clonal propagation than FOXP3− cells by STARTRAC-expa fected and infected/Tregs than known molecules (Fig. 3M; index (Fig. 4E; Supplementary Fig. S5B; Supplementary Table Supplementary Table S6), pointing to their potential as an S5), suggesting that the FOXP3+ phenotype is a premalignant HTLV-1 infection marker. state likely resulting in clonal evolution. Importantly, FOXP3+ By comparison with nonmalignant infected/Tregs (CD4-3 and FOXP3− cells shared TCRs in several ATL patients and and -4), 189 upregulated and 45 downregulated genes were ACs, but no unidirectional flow was demonstrated between identified in malignant cells (M01–30; Fig. 3N; Supplemen- these cells by RNA velocity and pseudotime trajectory anal- tary Table S6). Pathways regulating T-cell adhesion and acti- yses (Fig. 4F and G; Supplementary Fig. S5C and S5D), vation, such as LAG3 and CTLA4, were significantly enriched evidently demonstrating the phenotypic plasticity between in the upregulated genes. Moreover, comparison of ADT FOXP3+ and FOXP3− states. Compared with FOXP3− cells, levels depicted elevated expression of costimulatory and coin- FOXP3+ cells exhibited the upregulation of transcriptional hibitory molecules, particularly those related to somatic alter- programs negatively regulating lymphocyte activation, such ations (PD-L1 and CD28; Fig. 3O; Supplementary Fig. S4H; as CTLA4 and LAG3, and CD73 and CD39 immunosuppres- Supplementary Table S6), suggesting further deregulation of sive proteins (Fig. 4H and I; Supplementary Fig. S5E and immune-modulatory property in malignant cells. These cells S5F; Supplementary Table S6), suggesting that FOXP3+ cells also showed coordinated upregulation of several activation resemble Tregs with highly suppressive activity. This phe- markers (CD38, CD71, and CD25), suggesting their highly notype was validated by elevation of established Treg gene activated state. Therefore, although both HTLV-1 infection signature scores (ref. 25; Fig. 4J). These findings suggest that and malignant transformation alter the activity of diverse FOXP3+ Treg–like cluster expansion precedes ATL develop- immune pathways, they affect a different set of molecules ment, favoring premalignant clonal proliferation. that are supposed to collaborate in multistep carcinogenesis Next, we phenotypically characterized clonally expanded by HTLV-1. cells within HTLV-1–infected/Treg clusters (CD4-3 and -4). CD99 is a transmembrane protein involved in T-cell acti- Indeed, many clones harbored the identical TCRs with malig- vation and adhesion, but its function is poorly understood nant clonal cells in many ATL patients (17 of 34 sam- (23). CD99 expression was higher in infected/Tregs (CD4-3 ples; Fig. 4K), confirming that ATL frequently develops from and -4) compared with uninfected cells (CD4-1 and -2) and expanded nonmalignant clones. More than one third of further elevated in malignant cells (M01–30; Fig. 3K). Bulk ACs harbored somatic mutations related to ATL and/or RNA-seq data showed a stronger CD99 expression in aggres- clonal hematopoiesis of indeterminate potential (refs. 5, 11; sive rather than indolent subtypes (Supplementary Fig. S4I). Supplementary Table S7), suggesting that these mutations Notably, CRISPR-mediated CD99 knockout inhibited cell may cooperate with HTLV-1 infection to drive premalig- proliferation in vitro and caused a decrease in cycling cells nant clonal expansions. Interestingly, multiple MHC class II in the TL-Om1 ATL cell line (Fig. 3P and Q; Supplementary pathway genes were modestly reduced in clonal cells relative Fig. S4J). Among proliferation-related signaling pathways, to nonclonal cells (Fig. 4L; Supplementary Fig. S5G; Supple- CD99 disruption suppressed AKT activation (Fig. 3R). LGALS1 mentary Table S6). This reduction was confirmed by a slight encodes Galectin-1, a secreted protein exerting pleiotropic decrease of HLA-DR ADT level (Fig. 4M), suggesting that immune functions (24). LGALS1 was upregulated in infected/ HTLV-1–infected T cells may downregulate excessive MHC Tregs (CD4-3 and -4) and remained high in malignant cells class II molecules for clonal expansion, thereby avoiding (M01–30; Fig. 3K). Bulk RNA-seq data showed equivalent MHC class II–mediated immunity or anergy (22). LGALS1 expression between aggressive and indolent sub- Although intertumor heterogeneity is evident in ATL, types (Supplementary Fig. S4K). Lentiviral overexpression of intratumor differences were also demonstrated by a principal LGALS1 on BALL-1 cells significantly attenuated the cytotoxic component analysis (PCA) of mRNA levels, which showed activity of anti-CD3/CD28–stimulated PBMCs (Fig. 3S; Sup- the significant association of PC1 and PC2 with cell-cycle plementary Fig. S4L). In addition, LGALS1 overexpression genes (Supplementary Fig. S5H and S5I). For more precise promoted tumor growth in transplantation of murine E.G7- characterization, we used G1–S and G2–M signature scores OVA T-cell lymphoma (Fig. 3T; Supplementary Fig. S4M). In (26) and found a small proportion of cycling cells in each summary, these data corroborated the single-cell data that tumor (0%–33%; mainly in M13 cluster), which were higher highly upregulated immune proteins play a pivotal role in in aggressive than indolent subtypes (Fig. 4N–P; Supple- ATL pathogenesis. mentary Fig. S5J). As expected, cycling cells showed stronger

SEPTEMBER 2021 blood CANCER DISCOVERY | OF8

A Infected/Treg CD4+ T (CD4-3 and 4) B FOXP3 mRNA J Tumor Treg signature Peripheral Treg signature FOXP3+-2 FOXP3 +-3 0.4 3 FOXP3 +-1 0.5 3 0.3 0 0 0.2 UMAP2 −3 UMAP2 0.1 0 −3 0 −6 FOXP3 − −6 012 −0.1 −8 −4 048 −8 −4 048 − + − + EM EM UMAP1 UMAP1 Naïve Naïve FOXP3 FOXP3Malignant FOXP3 FOXP3Malignant C D E K Infected/Treg Infected/Treg P = 0.015 P = 4.4 × 10−4 Representative clones P = 1.3 × 10−3 shared between malignant and infected/Treg CD4+ T cells 1 1 1 AC T cells cells T cells cells ATL + + + + FOXP3 FOXP3 0.5 0.5 0.5 10 Clone Smoldering ATL08 Chronic TRA:CAASNDYKLSF; Acute TRB:CASSDYNEQFF Fraction of STARTRAC−expa index Fraction of 0 0

0 in infected/Treg CD4 in infected/Treg CD4 − + ATL17 AC ATL TRB:CASSEGTTGNNEQFF Indolent FOXP3 FOXP3 Aggressive 0 F Shared T-cell clones UMAP2 ATL30 TRA:CAARSTGGFKTIF; G TRB:CASSEVRGTEAFF RNA velocity of infected/Treg CD4+ T cluster FOXP3 + FOXP3 − FOXP3 +-1 3 FOXP3 +-2 FOXP3 +-3 −10 Cell type 0 Malignant Uninfected CD4+ T

UMAP2 + −3 Infected/Treg CD4 T Cytotoxic CD4+ T CD8+ T Others −6 AC09 TRB:CASSSGETQYF FOXP3− −8 −4 048 −100 10 UMAP1 UMAP1 H I LM Up in FOXP3+ (37) Up in FOXP3− (40) Up in FOXP3+ (3) Up in FOXP3 − (4) Up in clonal (8) Up in nonclonal (23) Up in clonal (2) Up in nonclonal (2) GO_NEGATIVE_ CD73 CD73 1 REGULATION_OF_ GO_ANTIGEN_PROCESSING_ FOXP3 LYMPHOCYTE_ Stimulatory AND_PRESENTATION_ ID2 OF_PEPTIDE_OR_ DIFFERENTIATION Inhibitory 0.5 POLYSACCHARIDE_ANTIGEN_ CD39 1 Treg related VIA_MHC_CLASS_II LAG3 0 0 CTLA4 HLA-DPB1 CD27 CD39 CD40LG ANXA1 HLA-DQA2 HLA-DRB1 HLA-DPB1 0 (FC of ADT level) (FC of ADT level) (FC of mRNA level) HLA-DPA1 (FC of mRNA level) 2 HLA-DPA1 2 2 −1 CD6 0 2 HLA-DR Activated GO_POSITIVE_ CD74 Log Log 0.5 Inhibitory Log CD2 Log − REGULATION_OF_ CD5 CD82 −1 HLA-DRA T_CELL_PROLIFERATION Treg related CD7 CCL5 CD7 HLA-DRB5 020406080 246 0102030 1234 Rank Rank Rank Rank NOP G1–S G2–M mRNA-based cycling status Malignant Cycling: G1–S or G2–M ≥ 0.15 P = 0.011 Nonmalignant Nonmalignant Nonmalignant 0.3 Smoldering 10 10 Chronic Acute

0.2 0 0 UMAP2 UMAP2 0.1 −10 −10 Cycling Fraction of cycling cells 0 0.2 0.4 0 0.4 Noncycling 0 −10 010 −10 010 Indolent Aggressive UMAP1 UMAP1 Q RS Up in cycling (15) Up in noncycling (2) 0.5 Noncycling malignant 0.5 Cycling malignant CD71 0.4 0.4 CD278 Fisher exact test 1 5 CD38 HLA-DR 0.3 0.3 P < 1.0 × 10− PD-1 CD2 0.2 0.2 CD25 CD86 Fraction of cells CD39 CD69 Fraction of cells 0.1 0.1 expressing the ADT(s) expressing the ADT(s) 0 0 (FC of ADT level) 0 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

2 CD71 CD71 Activated CD38 • • • • • • • • • • • • • • • CD38 • • • • • • • • • • • • • • • Log Stimulatory Inhibitory CD52 CD25 • • • • • • • • • • • • • • • CD25 • • • • • • • • • • • • • • • Treg related CD73 CD69 • • • • • • • • • • • • • • • CD69 • • • • • • • • • • • • • • • 51015 048 04 Rank Number of cells (×104) Number of cells (×10³)

OF9 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on October 1, 2021. Copyright 2021 American Association for Cancer Research. Multicellular Ecosystems in Viral Carcinogenesis by HTLV-1 RESEARCH ARTICLE expression of activation markers, such as CD71 and CD38, same V-J pair (Fig. 5G). HLA-B expression loss was found in a than noncycling cells, most of which were simultaneously cluster encompassing these minor clones (ATL03M-2), where positive (Fig. 4Q–S; Supplementary Fig. S5K). Interestingly, an HLA-B nonsense mutation, together with other cluster- CD52 and CD73 were downregulated in cycling cells (Sup- specific mutations, was identified by single-cell mutation plementary Fig. S5K), suggesting their negative impact on analysis (Fig. 5H and I; Supplementary Fig. S6H). Similar cycling malignant cells. Thus, these results suggest that tran- CDR3 mutations were found in subclones from five samples scriptomic and immunophenotypic heterogeneity is related (Fig. 5D), most of which were validated by bulk WES or RNA- to cell cycling and activation in malignant cells. seq (Supplementary Table S5), suggesting that subclonal acquisition of TCR mutations is common in ATL evolution. ATL Clonal Evolution Revealed by scTCR-seq Besides such TCR subclones, subclones with different V-J To further delineate intratumor heterogeneity, we analyzed pairs were found in seven samples (21%; Fig. 5D; Supple- scTCR-seq data in malignant clusters (M01–30), where most mentary Table S5), suggesting that independent subclones expanded (= major) clones expressed two TCRs in 25 samples frequently arise in the same patient. In a sample harboring (24 αβ and 1 ββ), and only one or three TCRs were mainly the largest subclone (ATL30), transcriptomic and immu- detected in 6 (2 α and 4 β) and 3 samples (2 ααβ and 1 αββ; nophenotypic differences between independent clones were Fig. 5A; Supplementary Fig. S6A; Supplementary Table S5). confirmed (Supplementary Fig. S6I–S6K). Moreover, sequen- These numbers were proportional to the TCR distribution of tial analysis of ATL02 revealed the emergence of an inde- uninfected CD4+ T cells (CD4-1 and -2; Fig. 5B), suggesting pendent clone after acute transformation, which showed the that although malignant clones present diverse TCR expres- enhanced malignant signature, together with other samples sion, they mostly reflect the physiologic TCR variation. There after progression (Fig. 5J; Supplementary Fig. S6L–S6N). were, however, notable exceptions: In uninfected CD4+ T cells By contrast, in ATL06, whereas a dominant clone remained, (CD4-1 and -2), different TCR V-J pairs were almost uniformly several independent subclones disappeared after acute trans- found in clones expressing two α or β chain (ααβ or αββ; formation (Fig. 5K), suggesting their clonal selection. Thus, Fig. 5C), likely resulting from allelic inclusion (27). By con- independent subclones with different biological features and trast, 36% of malignant cells (M01–30) with two α or β chain clinical behavior frequently develop in ATL. expression harbored the same pairs (Fig. 5C; Supplementary Fig. S6B), suggesting TCR duplication. Indeed, TRB ampli- Dynamic Changes in the Nonmalignant fication was detected in a sample (ATL01-2) predominantly Hematopoietic Pool of HTLV-1 Infection and ATL expressing the same TRB V-J pair (Supplementary Fig. S6C). HTLV-1 infection may induce various microenvironmental In this case, two slightly different TRB complementarity- changes, whereas additional modulation can occur through determining region 3 (CDR3) sequences were detected in the ATL leukemogenesis. To uncover the complexity of the dominant clone (mainly in ATL01M-3 and -4; Fig. 5D and HTLV-1–related immune ecosystem, we evaluated nonmalig- E; Supplementary Table S5). Considering that a dominant nant populations. Among their changes, the most prominent clone harbored the former CDR3 sequence in the smoldering was a marked increase of myeloid cells in ATL, particularly phase (ATL01), these results indicate that a CDR3 substitu- in aggressive subtypes, despite a decreasing trend in ACs tion occurred after TRB duplication. TRA CDR3 deletion was (Fig. 6A and B; Supplementary Fig. S7A), pointing to the also identified in another major clone from the acute phase relevance of the myeloid lineage in the nonmalignant hemat- (ATL01-2). Different transcriptomic and immunophenotypic opoietic pool of ATL. Subclustering of the myeloid cluster patterns, such as reciprocal expression of CD73 and PD-L1, (NM08, 11,698 cells) revealed three subclusters: CD14+ clas- were shown among major clones (Fig. 5F; Supplementary sic monocytes (My-1), CD16+ nonclassic monocytes (My-2), Fig. S6D–S6G). Interestingly, sequential comparison depicted and CD1c+ dendritic cells (My-3; Fig. 6C; Supplementary that the same clones were present as a subclone in the smold- Fig. S7B–S7D; Supplementary Table S4). All these subcluster ering phase and expanded after chemotherapy, suggesting fractions, particularly of nonclassic monocytes and dendritic different therapeutic sensitivity among ATL clones (Fig. 5E). cells, were increased in ATL (Fig. 6D). Functional enrichment Another example was ATL03, harboring several minor analysis in total myeloid cells showed the upregulation of clones with additional TRB CDR3 substitutions with the interferon (IFN) γ and α pathways in ATL patients compared

Figure 4. Functional heterogeneity of HTLV-1–infected nonmalignant and malignant CD4+ T cells. A, Subclustering of infected/Treg CD4+ T clusters (CD4-3 and -4). B, Normalized FOXP3 mRNA level on UMAP plot in A. C and D, The fraction of FOXP3+ cells in infected/Treg clusters (CD4-3 and -4) for each sample grouped by disease state (C) and by disease subtype (D). E, Clonal expansion levels of FOXP3− and FOXP3+ cells for each AC and ATL sample. F, T-cell clones shared between FOXP3− (light gray) and FOXP3+ (dark gray) cells. G, Steady-state RNA velocity of infected/Treg clusters (CD4-3 and -4). + − H and I, Differentially expressed mRNAs [q < 0.01 and log2(FC) > 0.4; H] and ADTs [q < 0.01 and log2(FC) > 0.3; I] between FOXP3 and FOXP3 cells. FC, fold change. J, Tumor Treg (left) and peripheral Treg (right) signature scores for each CD4+ T-cell cluster. Lines indicate medians. K, Representative clones shared between malignant (M01–30) and infected/Treg (CD4-3 and -4) clusters on UMAP plot in Fig. 1B, colored by cell type. L and M, Differentially expressed mRNAs (L) and ADTs (M) between clonal and nonclonal cells of infected/Treg clusters (CD4-3 and -4). H and L, Red and blue colors represent upregulated and downregulated mRNAs related to indicated pathways, respectively. N and O, G1–S (left) and G2–M (right) signature scores (N) and their based cycling status (G1–S or G2–M score ≥ 0.15; O) of malignant cells on UMAP plots in Fig. 1B. P, The fraction of cycling cells in malignant cells for each ATL sample, grouped and colored by disease subtype. Q, Differentially expressed ADTs between cycling and noncycling cells. I, M, and Q, Different colors represent different sets of T-cell–related ADTs. Treg-related ADTs are shown in orange. R and S, UpSet plot showing overlap of T-cell activation ADTs expressed (normalized expression ≥ 0.5) in noncycling (R) and cycling (S) cells. C–E and P, Welch t test. H–M and Q, Numbers of significant mRNAs or ADTs are shown in parentheses. See also Supplementary Fig. S5.

SEPTEMBER 2021 blood CANCER DISCOVERY | OF10

A Malignant clonal E 1 ATL01 TCR not detected Another sample αβ TRAV25: TRA:CAGEGGSGNTGKLIF;TRB:CASSLGETQYF αα TRAJ37; TRBV3-1: TRA:CAGEGGSGNTGKLIF;TRB:CASSLGDTQYF;TRB:CASSLGETQYF ββ TRBJ2-5 TRA:C GSGNTGKLIF;TRB:CASSLGETQYF 0.5 ααβ ATL01-2 acute αββ ATL01 smoldering ααββ 6 ATL01M-3

Fraction of cells ATL01M-1 α 0 β 3 ATL01M-4 0 ATL01 ATL02 ATL03 ATL04 ATL05 ATL06 ATL07 ATL08 ATL09 ATL10 ATL11 ATL12 ATL13 ATL14 ATL15 ATL16 ATL17 ATL18 ATL19 ATL20 ATL21 ATL22 ATL23 ATL24 ATL25 ATL26 ATL27 ATL28 ATL29 ATL30 UMAP2 ATL01-2 ATL02-2 ATL06-2 ATL22-2 ATL01M-2 −3

BCTwo α chains Two β chains −6 P = 0.012 P = 1.7 × 10−9 (23,099) (34) (1,316) (8) (717) (13) −10 −5 05 1 1 UMAP1 αβ F CD73 ADT PD-L1 ADT αα 6 ββ 0.5 ααβ 0.5 3 αββ ααββ 0

Fraction of clones α

UMAP2 3 0 β genes (%) the same V-J 0 Fraction of clones harboring −3 1 2 1 −6 0 0 Uninfected Malignantmajor UninfectedMalignantclonal UninfectedMalignantclonal −10 −5 05 UMAP1 D Malignant clonal Major clone type αβ ββ α β ααβ G H 1 ATL03 Up in ATL03M-2 (119) Up in ATL03M-1 (102) TRBV3-1:TRBJ1-2 HCST TRB CASSPTGGYGYTF STMN1 ANXA1 CASSPTGGYGYIF 1 TBL1XR1 CVSSPTGGYGYTF 0.5 EEF2 ATL03M-2 0 C12orf75 2.5 CES1 ITGA4 Fraction of cells -1 CCL3L1 0 −Log (q-value) THADA 0 10 HLA-DRB1

(FC of mRNA level) 300 2 UMAP2 SYNE2 CCR7 -2 −2.5 200 GPR25 Log ATL01 ATL03 ATL05 ATL06 ATL08 ATL09 ATL11 ATL12 ATL13 ATL14 ATL15 ATL16 ATL20 ATL21 ATL22 ATL24 ATL25 ATL26 ATL27 ATL28 ATL29 ATL30 ATL23 ATL18 ATL19 ATL02 ATL10 ATL17 ATL04 ATL07 100 ATL01-2 ATL06-2 ATL22-2 ATL02-2 ATL03M-1 −5 −3 HLA−B Independent clones Clones with additional substitutions/indels −10 −5 05 050 100 150 200 Major and its identical clones UMAP1 Rank

I Shared Cluster specific (ATL03M-1) Cluster specific (ATL03M-2) Cluster specific (ATL03M-2) STAT3 P = 1.0 CARD11 P = 4.5 × 10−3 CCR4 P = 5.7 × 10−3 HLA-B P = 5.8 × 10−3 2.5

0 UMAP2 −2.5 WT WT WT WT G1981T (D661Y) C1299A (S433R) C988T (Q330*) G723A (W241*) −5 −10 −5 05 UMAP1

J K ATL02 ATL06 ATL02 chronic ATL02-2 acute ATL06 chronic ATL06-2 acute ATL06M-1 10 ATL02M-1 Disappearing clones 5 5 ATL02M-2 ATL06M-2 0

UMAP2 0 −5 Emerging clones −5

−10 −5 05 −4 04 UMAP1 UMAP1

TRBV6-1:TRBJ1-2 TRAV26-1:TRAJ44;TRBV2:TRBJ2-5 TRAV13-1:TRAJ3;TRBV19:TRBJ1-5 TRAV5:TRAJ31;TRBV2:TRBJ1-4 TCR not detected TRAV8-3:TRAJ7;TRBV5-1:TRBJ2-7 TCR not detected TRAV26-1:TRAJ34;TRBV20-1:TRBJ2-3 Another sample TRAV21:TRAJ39;TRBV19:TRBJ2-1 Another sample TRAV12-2:TRAJ17;TRBV19:TRBJ1-5

OF11 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on October 1, 2021. Copyright 2021 American Association for Cancer Research. Multicellular Ecosystems in Viral Carcinogenesis by HTLV-1 RESEARCH ARTICLE with HDs (Fig. 6E; Supplementary Table S6). Although ADT and MAIT cells were decreased, particularly in ATL (Fig. 6P). levels largely reflected the subcluster fraction shift, upregula- Despite the lack of clear velocity trajectories in HDs and tion of activation markers (including CD64) and immune ACs, ATL patients showed a strong directional flow from checkpoints (such as PD-1) was observed in ATL (Fig. 6F; naïve to EM and eventually EMRA, which were considered Supplementary Table S6), suggesting the invigoration of to be less proliferative (Supplementary Fig. S8D and S8E). increased myeloid lineage. Of note, CD69, an early activation There were a few clonal CD8+ T cells in HDs, whereas ATL marker, was downregulated in both ACs and ATL patients, exhibited a more significant clonal expansion (Supplemen- suggesting differential regulation of CD69 from other activa- tary Fig. S8F and S8G; Supplementary Table S5). Although tion markers. no remarkable pathway was upregulated in ATL, TNFα B cells were decreased in ATL compared with HDs and ACs signaling was significantly downregulated (Fig. 6Q; Supple- (Fig. 6B), suggesting impaired humoral immunity. Within mentary Table S6). Furthermore, ATL-derived CD8+ T cells the B-cell clusters (NM06 and NM07, 24,907 cells), three showed stronger expression of activation markers, such as distinct subclusters emerged: CD38+IgM+ transitional (B-1), CD69, CD25, and HLA-DR (Fig. 6R; Supplementary Table CD27−IgD+ naïve (B-2), and CD27+IgD− switched-memory S6). Thus, ACs have decreased EM cells, which is further (B-3) B cells (Fig. 6G; Supplementary Fig. S7E–S7G; Sup- aggravated by ATL development despite induction of activa- plementary Table S4). Among these, switched-memory B tion markers. cells were most significantly decreased in ATL (Fig. 6H), sug- To elucidate whether ATL cells themselves can induce gesting attenuated germinal center reaction. Like myeloid phenotypic changes in nonmalignant cells, we performed cells, ATL patients showed stronger expression levels of IFN coculture assay with normal PBMCs (Fig. 6S). Unlike control pathway genes and a lower CD69 level, together with ADT cells, ATL cell lines significantly enhanced CD64 expres- levels corresponding to the subcluster changes (Fig. 6I and sion in myeloid cells and increased the frequency of acti- J; Supplementary Table S6). IFN pathway activation was also vated (CD25+HLA-DR+ and CD69+HLA-DR+) CD8+ T cells observed in NK and CD8+ T cells (Supplementary Fig. S7H), (Fig. 6T–V). These results highlight the substantial differ- suggesting that this is considered a pervasive phenomenon in ence between virus- and tumor-induced immune responses, immune cells in ATL. including selective increase in myeloid cells and decrease in B Subclustering of the NK cell cluster (NM05, 20,662 cells) cells and aggravation of HTLV-1–induced dysfunction of NK identified three subclusters: CD56bright (NK-1), CD328 and CTL responses in ATL. (SIGLEC7)+ mature (NK-2), and CD328− mature (NK-3) NK cells (Fig. 6K; Supplementary Fig. S7I–S7K; Supplementary Transfer of Genetically Overexpressed PD-L1 Table S4). Although the proportion of total NK cells was Proteins into Nonmalignant Cells comparable, ACs and ATL patients showed a larger frac- To disclose the effect of genetic alterations on the nonma- tion of CD328− NK cells than HDs (Fig. 6B and L). These lignant hematopoietic cells, we focused on CD274 SVs lead- results were confirmed by decreased CD328 ADT level in total ing to PD-L1 overexpression (12). In this cohort, there were NK cells in ATL. ATL-derived NK cells also showed MHC eight samples harboring CD274 SVs, detected by targeted- class II pathway activation and increased CX3CR1 ADT level seq and/or RNA-seq (Fig. 7A; Supplementary Table S1). (Fig. 6M and N; Supplementary Table S6). Therefore, NK cell These showed decreased proportion of CD8+ T cells, par- dysfunction seems to occur in ACs and be further enhanced ticularly EMRA cells (Fig. 7B and C), confirming the pre- in ATL patients (28, 29). vious result in solid cancers (12). In addition, CD274 SVs Similarly, within CD8+ T cells, four typical subclusters restrained clonal expansion of CD8+ T cells (Fig. 7D). Con- (NM02 and NM03, 21,595 cells) were present: naïve (CD8- spicuously, in patients with CD274 SVs, PD-L1 ADT level 1), EM (CD8-2), EM reexpressing CD45RA (EMRA; CD8-3), was increased in nonmalignant cells, particularly in B and and mucosal-associated invariant T (MAIT; CD8-4) cells myeloid cells, despite almost no mRNA level change (Fig. 7E; (Fig. 6O; Supplementary Fig. S8A–S8C; Supplementary Supplementary Fig. S8H), suggesting that PD-L1 overex- Table S4). ACs and ATL patients had or tended to have a pression in tumor cells can upregulate PD-L1 level in sur- smaller fraction of total CD8+ T cells than HDs (Fig. 6B). rounding cells. To further elucidate this phenomenon, we Although the frequency of EMRA cells was maintained, EM analyzed a newly generated mouse model recapitulating

Figure 5. ATL clonal evolution revealed by scTCR-seq. A, The fraction of malignant clonal cells (≥5 cells) expressing different numbers of productive TCR α and β chains in each ATL sample. B, The fraction of clones expressing different numbers of productive TCR α and β chains in uninfected CD4+ T-cell and most expanded (= major) malignant clones. C, The fraction of clones harboring the same V-J gene pairs in uninfected CD4+ T and malignant cell clones expressing two productive α (left) or β (right) chains. B and C, Numbers of TCR clonotypes are shown in parentheses. D, The fraction of malignant clonal cells from major and its identical clones (gray), clones with additional substitutions/indels (red), or independent clones with different V-J pairs (blue) for each ATL sample. E and F, Distribution of expanded TCR clonotypes in ATL01 (left) and ATL01-2 (right; E) and normalized CD73 (left) and PD-L1 (right) ADT levels (F) on UMAP plots of malignant cells from both samples. G, Distribution of expanded TCR clonotypes on UMAP plot of ATL03 malignant cells. E and G, Underlined, bold text indicates CDR3 mutations. H, Differentially expressed mRNAs [q < 0.01 and log2(FC) > 0.4] between ATL03M-1 and ATL03M-2 malignant clusters. Red color gradient indicates q-value. FC, fold change. I, Somatic STAT3, CARD11, CCR4, and HLA-B mutations and their references on UMAP plots in G, detected by scRNA-seq data. Asterisk (*) indicates stop codon. WT, wild-type. C and I, Fisher exact test. J and K, Distribu- tion of expanded TCR clonotypes in ATL02 (left) and ATL02-2 (right; J) and in ATL06 (left) and ATL06-2 (right; K) on UMAP plots of malignant cells from both samples of each patient. Red dotted lines surround clones emerging (J) or disappearing (K) after progression. E, G, J, and K, Black dotted lines surround subclusters. See also Supplementary Fig. S6.

SEPTEMBER 2021 blood CANCER DISCOVERY | OF12

A Nonmalignant CD4+ T CD8+ T BNK Myeloid B 1 CD8+ T NK B Myeloid

0.6 *****

0.5 0.4

Fraction of cells 0.2 Fraction of cells 0 0 HD AC ATL HD AC ATL HD AC ATL HD AC ATL AC01 AC02 AC03 AC04 AC05 AC06 AC07 AC08 AC09 AC10 AC11 HD01 HD02 HD03 HD04 ATL01 ATL02 ATL03 ATL04 ATL05 ATL06 ATL07 ATL08 ATL09 ATL10 ATL11 ATL12 ATL13 ATL14 ATL15 ATL16 ATL17 ATL18 ATL19 ATL20 ATL21 ATL22 ATL23 ATL24 ATL25 ATL26 ATL27 ATL28 ATL29 ATL30 ATL01-2 ATL02-2 ATL06-2 ATL22-2 HD AC ATL C D E Up in ATL (100) Up in HD (36) F Up in ATL (13) Up in HD (4) Myeloid (NM08) My-1 My-2 My-3 *** 2 IFITM1 CD16 PD-L1 CD64 HLA-DR 0.6 NFKBIA 0.5 IFITM3 CD45RA 4 MT2A CD1c LY6E CD123 PD-1 CD71 My-1 *** *** 1 TNFAIP3 0 0.4 HLA−A 0 Classical monocyte My-2 * * Nonclassical MX1 ISG15 CD18 CD33

UMAP2 monocyte 0 HLA−C (FC of ADT level) CD328 (FC of mRNA level)

−4 0.2 2 2 My-1 marker

My-3 Log −1 My-2 marker Log −1 HM_IFNγ_RESPONSE & Dendritic cell LYZ My-3 marker −8 HM_IFNα_RESPONSE CD69 0 SULT1A1 Fraction of cells in nonmalignant −5 0510 AC AC AC 050 100 51015 UMAP1 HD ATL HD ATL HD ATL Rank Rank G H I Up in ATL (59) Up in HD (36) J Up in ATL (4)Up in HD (9) B (NM06 and NM07) B-1 B-2 B-3 0.5 IFITM1 HM_IFNγ_RESPONSE & IgM IFI44L HM_IFNα_RESPONSE PD-L1 XAF1 0.5 CD38 0.4 1 LY6E 5 * * EPSTI1 B-2 0 B-3 MX1 Naïve 0.3 * BTG1 CD20 Switched-memory 0 UBE2L6 ISG15 CD44 Podoplanin 0 PSME2 CD45RA HLA−C CD31 UMAP2 0.2

STAT1 (FC of ADT level) CD19 CD21 (FC of mRNA level) 2 B-1 2 −1 0.1 −1 JUN CD73 −5 Transitional Log B-1 marker Log KLF6 B-2 marker PPP1R15A B-3 marker CD69 0 FOS −5 05Fraction of cells in nonmalignant 0255075 510 HD AC HD AC HD AC K UMAP1 L ATL ATL ATL M Rank N Rank NK (NM05) NK-1 NK-2 NK-3 Up in ATL (44) Up in HD (43) Up in ATL (8) Up in HD (3) HLA-DRA 0.5 CX3CR1 HLA-DR NK-1 ** 1 HLA-DQA1 0.5 HLA-DPB1 PD-1 4 CD56bright *** CD74 CD25 0.4 0 LAG3 HLA-DPA1 0 NK-2 HLA-DRB1 0 NK-3 CD328+ mature 0.3 -

UMAP2 CD328 mature (FC of ADT level) CD7 CD16 0.2 (FC of mRNA level) -1 GO_ANTIGEN_PROCESSING_ 2 2 −0.5 −4 PRESENTATION_OF_PEPTIDE_ NK-1 marker Log

0.1 Log POLYSACCHARIDE_ANTIGEN_ NK-2 marker −2 VIA_MHC_CLASS_II JUN CD328 0 −4 048Fraction of cells in nonmalignant 0255075 396 HD AC HD AC HD AC O UMAP1 P ATL ATL ATL Q Rank R Rank CD8-1CD8-2 CD8-3 CD8-4 Up in ATL (44) Up in HD (46) Up in ATL (7)Up in HD (1) CD8 +T (NM02 and NM03) 0.5 GZMB CD69 8 1 GNLY CX3CR1 HLA-DR CD8-4 0.4 0.4 PD-L1 * * TNFAIP3 CD25 CD183 MAIT NFKBIA 4 * * CD38 PRF1 CD69 0.3 0 JUNB 0.2 CD8-2 EIF1 EM 0 CD8-1 NR4A2 UMAP2 CD8-3 0.2 IER2 Naïve (FC of ADT level) 0 (FC of mRNA level)

EMRA -1 ZFP36 2 2 PPP1R15A CD8-2 marker −4 0.1 DUSP1 Log Log HM_TNFα_SIGNALING_ FOS -0.2 CD8-3 marker CD8-4 marker −2 VIA_NFKB JUN CD45RA 0 −5 0510 Fraction of cells in nonmalignant 0255075 2 468 HDAC HDAC HDAC HDAC UMAP1 ATL ATL ATL ATL Rank Rank STMyeloid UVCD8+ TCD8+ T

* (%) *** (%) ** + Normal CD4+ T * + Non-ATL (Jurkat) *** *** 9 80 60 ATL (TL-Om1, ATL43) Normal PBMCs * ** 7 HLA-DR 60 * HLA-DR ** + + 40 5 * 40 3 * 20 20 72 h 1 Ratio of CD64 intensity (n = 4) 0 (n = 4) 0 (n = 4) FACS Normal Jurkat TL-Om1 ATL43 Normal Jurkat TL-Om1 ATL43 Normal Jurkat TL-Om1 ATL43 Frequency of CD69 + Frequency of CD25 + + CD4 T Non-ATL ATL CD4 T Non-ATL ATL CD4 T Non-ATL ATL

OF13 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

CD274 3′-UTR truncation [Cd274 3′-UTR conditional knock- Another notable finding is frequent TCR CDR3 mutations, out (cKO); Fig. 7F; Methods]. We crossed these mice with enabling tracking of clonal evolution in T-cell malignancies. CD4-cre transgenic mice and validated recombination of Moreover, the emergence of independent clones is common the Cd274 allele and resultant Pd-l1 overexpression in CD4+ in ATL. scTCR-seq analysis disclosed different clinical behav- T cells (Fig. 7G). These mice exhibited almost no pheno- iors, such as acute transformation and therapeutic resist- type compared with wild-type (WT) mice (Supplementary ance, between malignant clones. Simultaneous mRNA and Fig. S8I and S8J). Importantly, Cd274 3′-UTR truncation in ADT level analysis uncovered posttranscriptional regulation CD4+ T cells caused a moderate elevation of Pd-l1 expression of surface markers, including dissociated protein upregula- in myeloid cells independent of the mRNA level (Fig. 7H; tion for myeloid markers and ATL-specific CD3 and CD62L Supplementary Fig. S8K), validating the human single-cell protein reduction. Although various posttranscriptional pro- data results. To investigate whether overexpressed PD-L1 cesses can be related, including translation rate and protein proteins in tumor cells are released into surrounding cells, turnover control (16), our data provide the fundamental we performed PBMC coculture (Supplementary Fig. S8L). basis for understanding cell-type–specific mRNA and protein Compared with control protein, PD-L1–GFP fusion protein expression regulation. was more efficiently transferred from HEK293T and Jurkat Our study revealed the dynamic changes in the nonma- T cells into B and myeloid cells and expressed on the cell lignant hematopoietic pool of HTLV-1 infection and ATL, surface (Fig. 7I and J; Supplementary Fig. S8M). In addition, demonstrating their differential effects on immune cells. the conditioned medium obtained from PD-L1–GFP fusion– Although hampered NK and CTL responses in HTLV-1 overexpressing HEK293T cells induced GFP expression in infection are further deteriorated, selective alterations occur the treated cells, unlike only GFP overexpression (Fig. 7K), in myeloid and B lineage in ATL, represented by myeloid suggesting that the biological property of PD-L1 promotes expansion, particularly of nonclassic monocytes and den- the intercellular protein transfer at least partly through indi- dritic cells. Together with recent single-cell studies (30), our rect mechanisms. Importantly, the transferred PD-L1–GFP findings point out an increasingly important role of various fusion inhibited the proliferation of activated CD8+ T cells myeloid subsets and their potential as a therapeutic target. (Fig. 7L and M). Together, genetically overexpressed PD-L1 Also noteworthy is the demonstration of genetically over- in tumor cells can serve as a major contributor of PD-L1 expressed PD-L1 in tumor cells as a major source of PD-L1 upregulation in nonmalignant hematopoietic pool. upregulation in nonmalignant cells, providing a clue for the intricate mechanisms regulating PD-L1 expression in the tumor microenvironment. Generally, PD-L1 expression in DISCUSSION tumor cells is induced by IFNγ, which leads to suppression Through multimodal single-cell analysis, we have deline- of antitumor immunity (31). It is intriguing to speculate that ated the phenotypic and functional heterogeneity of HTLV-1– upregulated PD-L1 in malignant cells can be transferred into infected normal and leukemic cells. HBZ expression clearly the microenvironment to help shape the adaptive immune demarcates HTLV-1–infected cells from uninfected CD4+ resistance. Our data combined with previous studies (32, cells, enabling their phenotypic characterization. Beyond 33) imply that somatically overexpressed proteins can spread known markers, several proteins including CD73 and CD99 through intercellular communication to remodel the tumor can show better discriminating potential. Within HTLV-1– microenvironment. infected cells, there are two distinct but interchangeable subsets: FOXP3− and FOXP3+ cells. Although cell of origin of METHODS ATL remains controversial, the latter FOXP3+ Treg–like phe- Detailed materials and methods are available in the Supplemen- notype accelerates HTLV-1–infected cells into premalignant tary Data. clonal expansion characterized by excessive MHC class II attenuation. By contrast, ATL cells express a diverse array of Data and Code Availability immune molecules, such as CD99 and LGALS1, which can scRNA-seq, scADT-seq, scTCR/BCR-seq data as well as newly be a potential target for immunotherapy. These observations sequenced bulk RNA-seq, targeted-seq, and WES data have been provide unique insights into multistep viral carcinogenesis, deposited in the European Genome-phenome Archive (EGA) under distinguishing the malignant phenotype driven by HTLV-1 accession number EGAS00001004936. All other data sets generated infection and ATL development. or analyzed during this study are included in the published article.

Figure 6. Dynamic changes in the nonmalignant hematopoietic pool of HTLV-1 infection and ATL. A and B, The fraction of each cell type in nonmalignant cells is shown for each sample (A) and compared among patient groups (B). C–R, Evaluation of myeloid (NM08; C–F), B (NM06 and NM07; G–J), NK (NM05; K–N), and CD8+ T (NM02 and NM03; O–R) cells. C, G, K, and O, Subclustering of myeloid (C), B (G), NK (K), and CD8+ T (O) cells. D, H, L, and P, The fraction of cells in each myeloid (D), B (H), NK (L), and CD8+ T (P) subcluster in nonmalignant cells (NM01–08) among patient groups. E, I, M, and Q, + Differentially expressed mRNAs [q < 0.01 and log2(FC) > 0.4] between HD- and ATL-derived myeloid (E), B (I), NK (M), and CD8 T (Q) cells. Red and blue colors represent upregulated and downregulated mRNAs related to indicated pathways respectively. FC, fold change; HM, hallmark. F, J, N, and R, Dif- + ferentially expressed ADTs [q < 0.01 and log2(FC) > 0.3] between HD- and ATL-derived myeloid (F), B (J), NK (N), and CD8 T (R) cells. Each color represents each related subcluster. E, F, I, J, M, N, Q, and R, Numbers of significant mRNAs or ADTs are shown in parentheses.S, Experimental design for coculture assay. T–V, Relative CD64 expression in myeloid cells (T) and the fraction of CD25+HLA-DR+ (U) and CD69+HLA-DR+ (V) cells in CD8+ T cells cocultured with normal CD4+ T, Jurkat (non-ATL), TL-Om1, or ATL43 cells (ATL). B, D, H, L, P, and T–V, Welch t test. ***, P < 0.0005; **, P < 0.005; *, P < 0.05. See also Supplementary Figs. S7 and S8.

SEPTEMBER 2021 blood CANCER DISCOVERY | OF14

A BCD Single cellSingle cellSingle cell Single cell + + Malignant CD4 T CD8 T NK B Myeloid CD8-1 CD8-2 CD8-3CD8-4 CD8+ T

l 1 *** 1 60 * ** leve ** 0.2 40 0.5 0.5 20 0.1

STARTRAC−expa index STARTRAC−expa 0 Average PD-L1 ADT 0 0 0 (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) (−)(+) CD274 SV CD274 SV CD274 SV CD274 SV Fraction of cells in nonmalignant Fraction of cells in nonmalignant

E F G Single cell Cd274 3′-UTR cKO mice ex4 ex5 ex6 ex7 Cd274 3′-UTR cKO mice B Myeloid 0.2 kb

l Floxed 15 CDS UTR WT allele (0.9 kb) leve ** ** ex7 CDS 10 Targeting vector DT PolyA WT Targeted allele (0.2 kb) 5 Deleted (0.9 kb) Floxed allele Cd4-Cre 0.9 kb (0.3 kb)

Average PD-L1 ADT 0 Deleted allele Tail B CD4+ T myeloid Tail BCD4+ T myeloid (−)(+) (−)(+) 0.9 kb CD274 SV PCR primer PGK neo loxPFRT Cd4-Cre(−) Cd4-Cre(+)

H I J K Cd274 3′-UTR cKO mice In vitro HEK293T In vitro Jurkat In vitro CD4+ T Myeloid B Myeloid B Myeloid Myeloid 5 *** *** 0.2 ** ** 0.3 *** 6 * 4 0.2 4 3 0.1 2 0.1 2

1 (n = 4) (n = 4) (n = 4) Pd-l1 protein expression 0 0 0 Fraction of GFP-positive cells Fraction of GFP-positive cells Cd274 3′-UTR fl/WT fl/WT fl/WT fl/WT Fraction of GFP-positive cells

Cd4-Cre (–)(+) (–)(+) Mock PD-L1 Mock PD-L1 Mock PD-L1 Mock PD-L1 Mock PD-L1

L M Activated In vitro CD45+ cells CD8+ T cells * 4 ×10 * + * CD8 T-cell– 24 h All depleted Sorting 4 PBMCs

− T-cell number Mock–GFP GFP + HEK293T 2

GFP+ PD-L1–GFP 0 (n = 4) HEK293T Absolute CD8 Second (–) All GFP−GFP+ First coculture Second coculture First (–) Mock PD-L1

Figure 7. Effects of genetically overexpressed PD-L1 on immune cells. A–E, Average PD-L1 ADT level in malignant cells (A), the fraction of each cell type (B) and each CD8+ T-cell subcluster (C) in nonmalignant cells (NM01–08), clonal expansion levels of CD8+ T cells (D), and average PD-L1 ADT level in each cell type (E) in ATL samples with or without CD274 SV. Subclusters with ≥30 cells are shown. A and E, Welch t test on log-transformed values. F, Scheme of Cd274 3′-UTR cKO mice. CDS, coding sequence; DT, diphtheria toxin. G and H, Genotyping PCR of tail and B (B220+), CD4+ T, and myeloid (Gr-1+Mac1+) cells (G) and relative Pd-l1 expression of CD4+ T and myeloid cells (H) in Cd274 3′-UTRfl/WT mice with (n = 9) or without (n = 14) Cd4-Cre. I and J, The fraction of GFP+ cells in normal B (CD19+) and myeloid (CD14+) cells cocultured with HEK293T cells (n = 4; I) and Jurkat T cells (n = 4; J) transduced with indicated plasmids. K, The fraction of GFP+ cells in myeloid cells cultured in supernatant from GFP- or PD-L1–GFP fusion–overexpressing HEK293T cells (n = 4). L, Experimental design for coculture assay. M, Absolute number of CD8+ T cells after 48 hours of coculture described in L (n = 4). B–D, H–K, and M, Welch t test. ***, P < 0.0005; **, P < 0.005; *, P < 0.05. See also Supplementary Fig. S8.

OF15 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

Custom code used to analyze the single-cell data set is available at samples were rapidly thawed and transferred into 10% fetal bovine https://github.com/nccmo/ATL_scRNA_seq. serum (FBS; Biosera) containing 20 mL prewarmed RPMI-1640 (Nacalai Tesque). After centrifugation at room temperature, pellets Human Samples were resuspended in 3 mL RPMI-1640 with 10% FBS, filtered using Primary PBMCs from 11 ACs (11 samples) and 30 ATL patients a 70-μm nylon mesh (Thermo Fisher Scientific), and stained with (34 samples including 4 sequential ones, which consisted of 19 acute, 0.5 μg/mL 4′,6-diamidino-2-phenylindole (DAPI; DOJINDO) for live/ 12 chronic, and 3 smoldering subtypes) were collected at University of dead staining. For cell sorting, all samples were gated based on forward Miyazaki Hospital and Imamura General Hospital. These studies were and side scatter, followed by exclusion of doublets, and then gated low 6 approved by the Institutional Review Board of the National Cancer on DAPI viable cells. After sorting 1 × 10 live cells, these cells were Center and other participating institutes. Written informed consent centrifuged at 300 × g at 4°C for 7 minutes, and resuspended in 50 μL was obtained from all participants in accordance with the Declara- Cell Staining Buffer (BioLegend) with Human TruStain FcX Blocking tion of Helsinki protocol. All patients were previously untreated with Solution (BioLegend) and incubated on ice for 10 minutes. The PBMCs systemic chemotherapy at the time of peripheral blood (PB) collec- were stained with a panel of TotalSeq-C antibodies (BioLegend), con- tion, except for one patient with sequential sampling who received sisting of 99 antibodies (ADT Set 1) for 18 samples and 105 antibodies chemotherapy between the first (ATL01) and second (ATL01-2) PB (ADT Set 2) for 31 samples (see Supplementary Table S2), according collections. HD samples were purchased from Zen-Bio, Precision to the manufacturer’s instructions except that each antibody was Bioservices, iQ Biosciences, and HemaCareCorporation (see also used at the amount of 0.2 μg. The scRNA-seq, scADT-seq, and scTCR/ Supplementary Table S1). The available clinical characteristics and BCR-seq libraries were prepared using the Chromium Single-Cell V(D) sequencing information are summarized in Supplementary Table S1. J Reagent Kits (v1 and v1.1): Single-Cell 5′ Library and Gel Bead Kit (PN-1000006) or Chromium Next GEM Single-Cell 5′ Library and Gel Mice Bead Kit v1.1 (PN-1000165), Chromium Single-Cell 5′ Feature Bar- code Library Kit (PN-1000080), Chromium Single-Cell V(D)J Enrich- Female C57BL/6 mice (6–10 weeks old) were obtained from CLEA ment Kit, Human T-cell (PN-1000005), Chromium Single-Cell V(D)J Japan and maintained under pathogen-free conditions. Cd4-Cre [B6. Enrichment Kit, Human B Cell (PN-1000016), Chromium Single-Cell Cg-Tg(Cd4-cre)1Cwi/BfluJ] mice were purchased from The Jackson 5′ Library Construction Kit (PN-1000020), Chromium Single-Cell A Laboratory. Cd274 3′-UTR cKO mice were generated by TransGenic Chip Kit (PN-120236) or Chromium Next GEM Chip G Single-Cell Inc. in this study. Schematic representation of the targeting vector Kit (PN-100120), and i7 Multiplex Kit (PN-120262), Single Index Kit and the resultant allele is shown in Fig. 7F, in which 5′ homologous T Set A (PN-1000213), or Chromium i7 Multiplex Kit N, Set A (PN- arm (4.7 kb), exon 7 [coding sequence (CDS) and 3′-UTR] flanked 1000084; 10x Genomics), according to the instructions in Chromium with loxP (3.3 kb), the FLP recognition target (FRT)-neomycin resist- Next GEM Single-Cell V(D)J Reagent Kits v1.1 with Feature Barcoding ance (Neo)-FRT cassette (1.8 kb), exon 7 CDS with growth hormone Technology for Cell-Surface Protein (manual part no. CG000208 Rev poly A signal (0.3 kb), and 3′ homologous arm (3.2 kb) were arranged A). These libraries were sequenced using the NovaSeq 6000 system in order. To identify embryonic stem (ES) cells with appropriate with the NovaSeq 6000 S4 Reagent Kit (Illumina) at Macrogen as 150 homologous recombination, we also constructed a control vector base pair paired-end reads except for three scBCR-seq libraries (ATL12, with elongated 5′ homologous arm (5.9 kb) and 3′ homologous arm ATL14, and ATL22) sequenced in-house using the NextSeq 500 system (3.8 kb) as a positive control in PCR selection. The linearized target- with the NextSeq 500/550 High Output Kit (Illumina). ing vector was electroporated into RENKA (C57BL/6) ES cells. After selection with G418 antibiotic, surviving clones were expanded for Statistical Analysis PCR analysis using 5′ flanking–, ′3 flanking–, and ′5 loxP–specific primers to identify recombinant ES clones. Secondary confirmation Statistical analyses were performed with R3.6.1 software (The R of positive clones identified by PCR was performed by Southern Foundation for Statistical Computing). Comparison of categorical blotting using hybridization technique with probes targeted against and continuous data was performed using the Fisher exact test and the Neo region. Positive ES clones were expanded and microinjected Welch t test, respectively, unless otherwise specified. All tests were into ICR eight-cell stage embryos through the aggregation technique two-sided, unless otherwise specified. Box plots show median (lines), and then transplanted into C57BL/6 mice (Sankyo-Laboratory Ser- interquartile range (IQR; boxes), and ± 1.5 × IQR (whiskers). vice). Resulting chimeras with a high percentage of black coat color were crossed to a germline Flp-deleter mouse line [B6;D2-Tg(CAG- Authors’ Disclosures Flp)18Imeg] to eliminate the FRT-Neo-FRT cassette. Cd274 3′-UTR cKO N. Nakano reports personal fees from Novartis Pharma, Takeda Phar- mice were genotyped using a PCR-based assay to detect both the WT and maceutical, Kyowa Kirin, Celgene, Eisai, Nippon Shinyaku, Astellas floxed alleles. Genomic DNA was isolated from tail biopsies or PBMCs Pharma, Bristol Myers Squibb, MSD, Otsuka Pharmaceutical, Asahi and subjected to PCR. The PCR samples were denatured at 94°C for 2 Kasei Medical, Chugai Pharmaceutical, and JIMRO outside the sub- minutes, subjected to 35 cycles of amplification (98°C for 10 seconds, mitted work. A. Utsunomiya reports personal fees from Novartis 62.5°C for 30 seconds, and 68°C for 3 minutes), and followed by a final Pharma, Kyowa Kirin, Daiichi Sankyo, Bristol Myers Squibb, Celgene, extension step at 68°C for 5 minutes. PCR products were resolved by Pfizer, Minophagen Pharmaceutical, Janssen Pharmaceutical, HUYA agarose gel electrophoresis. PCR primers are listed in the Key Resources Japan, JIMRO, Meiji Seika Pharma, and Otsuka Medical Devices Table in the Supplementary Data. Male and female 12- to 16-week-old outside the submitted work. Y. Togashi reports grants and personal mice were used for each animal experiment. Littermates were used as fees from Ono Pharmaceutical and Bristol Myers Squibb; personal controls in all experiments. All mouse experiments were approved by fees from Chugai Pharmaceutical and MSD; and grants from KOTAI the Animals Committee for Animal Experimentation of the National Biotechnologies and Daiichi Sankyo outside the submitted work. Cancer Center and met the Guidelines for Proper Conduct of Animal S. Ogawa reports a patent for P6810396 licensed and with royal- Experiments established by the Science Council of Japan. ties paid from Asahi Genomics. K. Kataoka reports grants from AMED, Japan Society for the Promotion of Science, National Cancer Sample Preparation, Library Preparation, and Sequencing Center Research and Development Funds, Takeda Science Founda- PBMCs were isolated from whole blood using Ficoll-Paque PLUS tion, Japan Leukemia Research Fund, and The Naito Foundation (cytiva) and cryopreserved in CELLBANKER 1plus (TaKaRa Bio) at during the conduct of the study; grants, personal fees, and non- −150°C according to the manufacturer’s instructions. Frozen PB financial support from Otsuka Pharmaceutical, grants from Chordia

SEPTEMBER 2021 blood CANCER DISCOVERY | OF16

Therapeutics, Bristol Myers Squibb, Ono Pharmaceutical, Shionogi, 4. Rowan AG, Dillon R, Witkover A, Melamed A, Demontis MA, Teijin Pharma, Japan Blood Products Organization, Mochida Phar- Gillet NA, et al. Evolution of retrovirus-infected premalignant T-cell maceutical, and JCR Pharmaceuticals, grants and personal fees from clones prior to adult T-cell leukemia/lymphoma diagnosis. Blood 2020; Chugai Pharmaceutical, Takeda Pharmaceutical, Eisai, Kyowa Kirin, 135:2023–32. and Sumitomo Dainippon Pharma, personal fees from Celgene, 5. Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman PV, Mar BG, Astellas Pharma, Novartis, AstraZeneca, and Janssen Pharmaceuti- et al. Age-related clonal hematopoiesis associated with adverse out- cal, and other support from Asahi Genomics outside the submitted comes. N Engl J Med 2014;371:2488–98. work; and a patent for genetic alterations as a biomarker in T-cell 6. Yokoyama A, Kakiuchi N, Yoshizato T, Nannya Y, Suzuki H, Takeuchi Y, et al. Age-related remodelling of oesophageal epithelia by mutated lymphomas licensed to Kyoto University and a patent for PD-L1 cancer drivers. Nature 2019;565:312–7. abnormalities as a predictive biomarker for immune-checkpoint 7. Ishitsuka K, Tamura K. Human T-cell leukaemia virus type I and blockade therapy licensed to Kyoto University. No disclosures were adult T-cell leukaemia-lymphoma. Lancet Oncol 2014;15:e517–e26. reported by the other authors. 8. Iwanaga M, Watanabe T, Utsunomiya A, Okayama A, Uchimaru K, Koh KR, et al. Human T-cell leukemia virus type I (HTLV-1) proviral Authors’ Contributions load and disease progression in asymptomatic HTLV-1 carriers: a J. Koya: Conceptualization, resources, data curation, formal analysis, nationwide prospective study in Japan. Blood 2010;116:1211–9. validation, investigation, visualization, methodology, writing–original 9. Bangham CR, Osame M. Cellular immune response to HTLV-1. draft, writing–review and editing. Y. Saito: Data curation, software, Oncogene 2005;24:6035–46. formal analysis, investigation, visualization, methodology, writing– 10. Cook LB, Fuji S, Hermine O, Bazarbachi A, Ramos JC, Ratner L, et al. original draft, writing–review and editing. T. Kameda: Resources, Revised adult T-cell leukemia-lymphoma international consensus writing–review and editing. Y. Kogure: Investigation, writing–review and meeting report. J Clin Oncol 2019;37:677–87. editing. M. Yuasa: Data curation, formal analysis, investigation, visu- 11. Kataoka K, Nagata Y, Kitanaka A, Shiraishi Y, Shimamura T, alization, writing–original draft, writing–review and editing. J. Nagasaki: Yasunaga J, et al. Integrated molecular analysis of adult T cell leukemia/ Formal analysis, validation, visualization. M.B. McClure: Investigation, lymphoma. Nat Genet 2015;47:1304–15. 12. Kataoka K, Shiraishi Y, Takeda Y, Sakata S, Matsumoto M, Nagano S, writing–review and editing. S. Shingaki: Investigation, writing–review et al. Aberrant PD-L1 expression through 3’-UTR disruption in mul- and editing. M. Tabata: Investigation, writing–review and editing. tiple cancers. Nature 2016;534:402–6. Y. Tahira: Resources. K. Akizuki: Resources. A. Kamiunten: Resources. 13. Baslan T, Hicks J. Unravelling biology and shifting paradigms in can- M. Sekine: Resources. K. Shide: Resources. Y. Kubuki: Resources. cer with single-cell sequencing. Nat Rev Cancer 2017;17:557–69. T. Hidaka: Resources. A. Kitanaka: Resources. N. Nakano: 14. Stoeckius M, Hafemeister C, Stephenson W, Houck-Loomis B, Resources. A. Utsunomiya: Resources. Y. Togashi: Formal analy- Chattopadhyay PK, Swerdlow H, et al. Simultaneous epitope and tran- sis, validation, visualization. S. Ogawa: Resources. K. Shimoda: scriptome measurement in single cells. Nat Methods 2017;14:865–8. Resources, funding acquisition. K. Kataoka: Conceptualization, for- 15. Zhang L, Yu X, Zheng L, Zhang Y, Li Y, Fang Q, et al. Lineage tracking mal analysis, supervision, funding acquisition, investigation, writing– reveals dynamic relationships of T cells in colorectal cancer. Nature original draft, project administration, writing–review and editing. 2018;564:268–72. 16. Liu Y, Beyer A, Aebersold R. On the dependency of cellular protein Acknowledgments levels on mRNA abundance. Cell 2016;165:535–50. This work was supported by AMED [JP20wm0325013 (K. Kataoka), 17. Imhof BA, Aurrand-Lions M. Adhesion mechanisms regulating the JP19ck0106261 (K. Kataoka), JP19ck0106254 (K. Shimoda), migration of monocytes. Nat Rev Immunol 2004;4:432–44. JP20ck0106542 (K. Kataoka), JP20ck0106538 (K. Shimoda), and 18. Watanabe T. Adult T-cell leukemia: molecular basis for clonal expansion and transformation of HTLV-1–infected T cells. Blood 2017;129: JP20cm0106575 (K. Kataoka)], Japan Society for the Promotion 1071–81. of Science (JSPS) KAKENHI (19K22573 to K. Kataoka), National 19. Stubbington MJT, Lonnberg T, Proserpio V, Clare S, Speak AO, Cancer Center Research and Development Funds (31-A-4 to Dougan G, et al. T cell fate and clonality inference from single-cell K. Kataoka), Takeda Science Foundation (K. Kataoka), Japan Leu- transcriptomes. Nat Methods 2016;13:329–32. kemia Research Fund (K. Kataoka), and The Naito Foundation 20. Carter JA, Preall JB, Atwal GS. Bayesian inference of allelic inclusion (K. Kataoka). The authors thank Y. Hokama, F. Ueki, and Y. Ito for rates in the human T cell receptor repertoire. Cell Syst 2019;9:475–82. technical assistance. The sequencing resources for scTCR/BCR-seq 21. La Manno G, Soldatov R, Zeisel A, Braun E, Hochgerner H, Petukhov V, were provided by Fundamental Innovative Oncology Core at the et al. RNA velocity of single cells. Nature 2018;560:494–8. National Cancer Center. The supercomputing resources were pro- 22. Holling TM, Schooten E, van Den Elsen PJ. Function and regulation vided by the Human Genome Center, the Institute of Medical Sci- of MHC class II molecules in T-lymphocytes: of mice and men. Hum ence, The University of Tokyo. Immunol 2004;65:282–90. 23. Pasello M, Manara MC, Scotlandi K. CD99 at the crossroads of physi- Received March 5, 2021; revised June 17, 2021; accepted July 9, ology and pathology. J Cell Commun Signal 2018;12:55–68. 2021; published first July 13, 2021. 24. Cedeno-Laurent F, Dimitroff CJ. Galectin-1 research in T cell immunity: past, present and future. Clin Immunol 2012;142:107–16. 25. Zheng C, Zheng L, Yoo JK, Guo H, Zhang Y, Guo X, et al. Landscape of infiltrating T cells in liver cancer revealed by single-cell sequencing. References Cell 2017;169:1342–56. 1. de Martel C, Ferlay J, Franceschi S, Vignat J, Bray F, Forman D, et al. 26. Tirosh I, Venteicher AS, Hebert C, Escalante LE, Patel AP, Yizhak K, Global burden of cancers attributable to infections in 2008: a review et al. Single-cell RNA-seq supports a developmental hierarchy in and synthetic analysis. Lancet Oncol 2012;13:607–15. human oligodendroglioma. Nature 2016;539:309–13. 2. Butel JS. Viral carcinogenesis: revelation of molecular mechanisms 27. Brady BL, Steinel NC, Bassing CH. Antigen receptor allelic exclusion: and etiology of human disease. Carcinogenesis 2000;21:405–26. an update and reappraisal. J Immunol 2010;185:3801–8. 3. Mason WS, Gill US, Litwin S, Zhou Y, Peri S, Pop O, et al. HBV 28. Brunetta E, Fogli M, Varchetta S, Bozzo L, Hudspeth KL, Marcenaro E, DNA integration and clonal hepatocyte expansion in chronic hepa- et al. The decreased expression of Siglec-7 represents an early marker titis B patients considered immune tolerant. Gastroenterology 2016; of dysfunctional natural killer-cell subsets associated with high levels 151:986–98. of HIV-1 viremia. Blood 2009;114:3822–30.

OF17 | blood CANCER DISCOVERY SEPTEMBER 2021 AACRJournals.org

29. Sciumè G, De Angelis G, Benigni G, Ponzetta A, Morrone S, Santoni A, 32. Al-Nedawi K, Meehan B, Micallef J, Lhotak V, May L, Guha A, et al. CX3CR1 expression defines 2 KLRG1+ mouse NK-cell subsets et al. Intercellular transfer of the oncogenic receptor EGFRvIII by with distinct functional properties and positioning in the bone mar- microvesicles derived from tumour cells. Nat Cell Biol 2008;10: row. Blood 2011;117:4467–75. 619–24. 30. Zhang L, Li Z, Skrzypczynska KM, Fang Q, Zhang W, O’Brien SA, 33. Peinado H, Aleckovic M, Lavotshkin S, Matei I, Costa-Silva B, et al. Single-cell analyses inform mechanisms of myeloid-targeted Moreno-Bueno G, et al. Melanoma exosomes educate bone marrow therapies in colon cancer. Cell 2020;181:442–59. progenitor cells toward a pro-metastatic phenotype through MET. 31. Topalian SL, Taube JM, Anders RA, Pardoll DM. Mechanism-driven Nat Med 2012;18:883–91. biomarkers to guide immune checkpoint blockade in cancer therapy. Nat Rev Cancer 2016;16:275–87.

SEPTEMBER 2021 blood CANCER DISCOVERY | OF18

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on October 1, 2021. Copyright 2021 American Association for Cancer Research. Single-Cell Analysis of the Multicellular Ecosystem in Viral Carcinogenesis by HTLV-1

Junji Koya, Yuki Saito, Takuro Kameda, et al.

Blood Cancer Discov Published OnlineFirst July 13, 2021.

Updated version Access the most recent version of this article at: doi: 10.1158/2643-3230.BCD-21-0044

Supplementary Access the most recent supplemental material at: Material http://bloodcancerdiscov.aacrjournals.org/content/suppl/2021/07/10/2643-3230.BCD-21-0044.DC 1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://bloodcancerdiscov.aacrjournals.org/content/early/2021/08/20/2643-3230.BCD-21-0044. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.