The oncogenic driving force of CD30 signaling‐induced chromosomal instability in adult T‐cell leukemia/lymphoma

Abstract

Adult T‐cell leukemia/lymphoma (ATL) develops via stepwise accumulation of gene mutations and chromosome aberrations. However, the molecular mechanisms underlying this tumorigenic process are poorly understood. We previously reported the presence of a biological link between the expression of CD30, which serves as a marker for ATL progression, and the actively proliferating fraction of human T‐cell leukemia virus type 1 (HTLV‐1)‐infected cells that display polylobulation. Here, we demonstrated that CD30 signaling induced chromosomal instability with clonal expansion through DNA double‐strand breaks (DSBs) via an increase of intracellular reactive oxygen species. CD30⁺ATL cells were composed of subclones with additional genomic aberrations compared with CD30⁻ATL cells in ATL patients. Furthermore, we found an accumulation of copy number loss of DSB repair‐related genes as the disease progressed. Taken together, CD30 expression on ATL cells appears to be correlated with genomic instability, suggesting that CD30 signaling is one of the oncogenic factors of ATL progression with clonal evolution. This study provides new insight into the biological roles of CD30 signaling and could improve our understanding of tumorigenic processes of HTLV‐1‐infected cells.

CD30 signaling induced chromosomal instability through DSBs via an increase of intracellular reactive oxygen species in CD30‐expressing ATL cells. These results highlight that endogenous signaling by CD30 is an oncogenic signal that promotes ATL progression and that CD30 is a promising target for treating aggressive ATL and inhibiting disease progression in indolent ATL.

Abbreviations

ATL

adult T‐cell leukemia/lymphoma

CGH

comparative genomic hybridization

CNA

copy number alteration

CNG

copy number gain

CNL

copy number loss

DSB

DNA double‐strand break

GSH

glutathione

HR

homologous recombination

HTLV‐1

human T‐cell leukemia virus type 1

NAC

N‐acetyl‐L‐cysteine

NHEJ

non‐homologous end joining

ROS

reactive oxygen species

SSB

DNA single‐strand break

tBHP

tert‐butyl hydroperoxide

TPA

12‐O‐tetradecanoylphorbol 13‐acetate

1. INTRODUCTION

Chromosomal instability is a hallmark of human cancer, caused by incorrectly repaired DNA double‐strand breaks (DSBs). ¹ , ² The major pathways of DSB repair, non‐homologous end joining (NHEJ), and homologous recombination (HR) are crucial for maintaining genomic stability. Genetic disruption of these DSB repair pathways causes genomic instability in mammalian primary cells. ³ Single‐strand breaks (SSBs) in DNA, which can arise from oxidative damage, can be converted to DSBs. ⁴ Oxidative stress refers to elevated intracellular levels of reactive oxygen species (ROS) that cause damage to lipids, proteins, and DNA. ⁵ ROS is a general name for the highly reactive byproducts of aerobic metabolism, including hydrogen peroxide, hypochlorite ions, superoxide anions and hydroxyl radicals. Endogenous ROS are mainly generated as byproducts in mitochondria, which produce ATP efficiently through aerobic respiration using pyruvate and NADH generated by glycolysis.

CD30 is a member of the tumor necrosis factor receptor superfamily and activates pro‐survival signals by ligation of its ligand (CD30L) or by CD30 overexpression. ⁶ , ⁷ In various cell types, CD30 activates nuclear factor (NF)‐κB, ERK MAPK, p38 MAPK, JNK MAPK, and PI3K in a cell type‐dependent manner. ⁸ , ⁹

Adult T‐cell leukemia/lymphoma (ATL) is a T‐cell neoplasm with a poor prognosis caused by human T‐cell leukemia virus type 1 (HTLV‐1) infection. ATL develops after approximately 50 years of clinical latency in 2%–5% of HTLV‐1 carriers ¹⁰ ; it is estimated that 5–10 million carriers exist worldwide. ¹¹ Transformation of HTLV‐1‐infected T cells in vivo is a multistage process, which reflects the status of HTLV‐1 infection (i.e., asymptomatic state, smoldering, chronic, lymphoma, or acute type of ATL). ¹² HTLV‐1‐infected T cells transform via stepwise accumulation of genetic mutations and chromosome aberrations ¹³ , ¹⁴ , ¹⁵ However, the molecular mechanisms underlying this tumorigenic process are poorly understood.

We previously reported the presence of a biological link between the expression of CD30, which serves as a marker for ATL progression, and the actively proliferating fraction of HTLV‐1‐infected cells that display polylobulation. CD30 signaling induced by CD30L triggers cell‐cycle promotion and abnormal cell division of HTLV‐1‐infected T cells, which coincides with the generation of characteristic polylobulated cells. ⁹ These findings highlight the role of CD30⁺ cells as a potential reservoir for transformation and clonal expansion in HTLV‐1‐infected individuals.

In this report, we aimed to elucidate the mechanism of the tumorigenic process involving genomic aberrations in HTLV‐1‐infected T‐cells. We demonstrate that CD30 signaling induces chromosomal aberrations through DSBs via an increase in intracellular ROS in CD30‐expressing ATL cells. We present supportive evidence suggesting that CD30 expression on ATL cells is correlated with genomic instability. We further discuss the significance of CD30 signaling as an oncogenic signal and the therapeutic implications for the treatment of ATL patients.

2. MATERIALS AND METHODS

Additional materials and methods are available in Appendix S1.

2.1. Clinical samples

Primary cells were collected with informed consent in accordance with the Declaration of Helsinki protocol as part of a collaborative study with the Joint Study on Predisposing Factors of ATL Development. ¹⁶ The experiments and analyses using primary cells were performed at the University of Tokyo. Clinical data for ATL individuals used are presented in Table S1. All patients were categorized according to Shimoyama's criteria. ¹²

3. RESULTS

3.1. CD30 signaling induces an increase of intracellular reactive oxygen species without inducing apoptosis in adult T‐cell leukemia/lymphoma cell lines and CD30 ⁺ ATL cells from an adult T‐cell leukemia/lymphoma patient

We used an intracellular ROS indicator (CellRox) to determine whether CD30 signaling induces an increase of intracellular ROS. Treatment with the intracellular ROS inducing agent tBHP or TPA increased intracellular ROS in the ATL cell line HUT102. Moreover, tBHP or TPA treatment in the presence of the scavenging antioxidant N‐acetyl‐L‐cysteine (NAC) or glutathione (GSH), respectively, suppressed the increases in intracellular ROS (Figure 1A,B). The CD30 stimulation method was previously reported by us and other researchers. ⁹ , ¹⁹ , ²⁰ CD30 stimulation by CD30L significantly increased intracellular ROS in CD30‐expressing ATL cell lines (HUT102 and TL‐Om1) and CD30⁺ATL cells from the ATL patient (chronic#5), while CD30‐weak positive T‐cell lines (Jurkat and CEM) and CD30⁻ Ramos cells did not exhibit elevated ROS (Figure 1C,D; Figure S1A,B). We compared CD30 expression levels in primary CD30⁺ATL cells with those in cell lines. CD30 expression levels in primary CD30⁺ATL cells were stronger than in Jurkat and CEM cells and comparable to HUT102 cells but weaker than in TL‐Om1 (Figure S2). Jurkat cells and CEM cells appear to be insufficient for the induction of intracellular ROS by CD30 signaling due to low CD30 expression. We overexpressed CD30 in Jurkat cells (Jurkat‐CD30OE) and tested whether CD30 signaling induces an increase in intracellular ROS in Jurlat‐CD30OE cells. We found that overexpression of CD30 induced an increase in intracellular ROS upon CD30 stimulation by CD30L (Figure S3A,B). Because an increase in intracellular ROS might induce apoptosis, we evaluated the cell‐surface binding of annexin V and found that CD30 signaling did not induce apoptosis in ATL cell lines and CD30⁺ATL cells (Figure 1E; Figure S1D). These results indicate that CD30 signaling induces an increase of intracellular ROS without inducing apoptosis in ATL cell lines and CD30⁺ATL cells in a CD30 expression level‐dependent manner.

FIGURE 1.

The effect of CD30 stimulation on intracellular reactive oxygen species (ROS) in adult T‐cell leukemia/lymphoma (ATL) cell lines. (A) Histogram of fluorescence intensity of an intracellular ROS indicator, CellRox. HUT102 cells were treated with the ROS inducer tert‐Butyl hydroperoxide (tTBHP; 50 μM) (left middle) and the ROS scavenger N‐acetyl‐L‐cysteine (NAC) (1 mM) (left bottom). The median fluorescence intensity (MFI) of each treatment condition is shown as a bar graph (right). Data are shown as mean ± SD of five independent experiments. Asterisks indicate statistical significance (*p < 0.05, ***p < 0.001). (B) HUT102 cells were treated with the ROS inducer TPA (50 nM) and the ROS scavenger glutathione (GSH) (1 mM). The MFI of each treatment condition is shown as a bar graph. Data are shown as mean ± SD of three independent experiments. Asterisks indicate statistical significance (**p < 0.01, ***p < 0.001). (C) Histogram of fluorescence intensity of CD30 expression (red) in ATL cell lines (HUT102 and TL‐Om1), ALL T‐cell lines (Jurkat and CEM), and a Burkitt lymphoma cell line (Ramos). Control staining by an isotype control antibody is shown in gray. (D) The MFI of CellRox in stimulated cells is shown as a bar graph. Data are shown as mean ± SD of three independent experiments. Asterisks indicate statistical significance (**p < 0.01). (E) Effect of CD30 stimulation on annexin V binding on the plasma membrane. The proportion of annexin V⁺ HUT102 and TL‐Om1 cells is shown as a bar graph. Data are shown as mean ± SD of three independent experiments. n.s., not significant

3.2. CD30 signaling induces DNA double‐strand breaks via increased intracellular reactive oxygen species in adult T‐cell leukemia/lymphoma cell lines

An excess of ROS can cause SSBs and subsequent DSBs. ⁴ We used the comet assay to evaluate the type and quantity of DNA damage in ATL cells stimulated by CD30L. ¹⁷ The comet tail moment under alkaline conditions (alkaline comet tail moment) reflects both DSBs and SSBs, whereas the comet tail moment under neutral conditions (neutral comet tail moment) reflects only DSBs. We evaluated whether hydrogen peroxide, which is reported to be a DSB‐inducing agent, induces DSB in an ATL cell line. ²¹ Treatment with hydrogen peroxide produces an increase in both the alkaline and neutral comet tail moment in HUT102 cells, indicating that hydrogen peroxide induced DSBs in HUT102 cells in a dose‐dependent manner (Figure 2A). Therefore, we evaluated the relationship between CD30 signaling and DSBs by comet assay. Both alkaline and neutral comet tail moments increased in ATL cell lines and CD30⁺ATL cells stimulated by CD30L, while neither increased in Jurkat, CEM, and Ramos cells under the same experimental conditions (Figure 2B; Figure S1C). We tested whether CD30 signaling induces an increase in DSBs in Jurkat‐CD30OE cells. Both alkaline and neutral comet tail moments increased in Jurkat‐CD30OE cells stimulated by CD30L (Figure S3C). These results indicate that CD30 signaling promotes DSBs in ATL cell lines and CD30⁺ATL cells in a CD30 expression level‐dependent manner. Furthermore, we examined whether an increase of intracellular ROS by CD30 signaling induces DSBs. Treatment with the scavenging antioxidant GSH, which neutralized CD30 signaling‐induced intracellular ROS, significantly depressed both alkaline and neutral comet tail moments (Figure 2C,D). These results indicate that CD30 signaling triggers DSBs via an increase of intracellular ROS in ATL cell lines.

FIGURE 2.

Effect of CD30 stimulation on DNA double‐strand breaks (DSBs) through increased intracellular reactive oxygen species (ROS) in adult T‐cell leukemia/lymphoma (ATL) cell lines. (A) HUT102 cells were treated with hydrogen peroxide at the indicated concentration for 30 min, and DNA damage was analyzed by comet assay (alkaline comet: n = 31, 19, 26, 41; neutral comet: n = 27, 21, 23, 29 at each concentration, respectively). Representative comet assay images of treated cells are shown to the right of the box plots. Statistical significance was determined using the Kruskal–Wallis test. **p < 0.01, ***p < 0.001. (B) HUT102, TL‐Om1, Jurkat, CEM, and Ramos cells were stimulated by mock‐CHO or CD30L‐CHO cells for 24 h, and DNA damage was analyzed by comet assay (alkaline comet: n = 48, 27, 33, 41, 36, 41, 47, 41, 19, 29; neutral comet: n = 57, 40, 47, 38, 48, 56, 73, 62, 38, 46, respectively). Representative comet assay images of stimulated cells are shown below the boxplots plots. Statistical significance was determined by the Mann–Whitney U‐test. *p < 0.05, **p < 0.01, ***p < 0.001. (C) HUT102 cells were stimulated by CD30L‐CHO or mock‐CHO cells and treated with glutathione (GSH) (1 mM) for neutralization of intracellular ROS for 24 h. The MFI of each treatment condition is shown as a bar graph. Data are shown as mean ± SD of three independent experiments. Asterisks indicate statistical significance (*p < 0.05, **p < 0.01). (D) HUT102 cells were stimulated by mock‐CHO or CD30L‐CHO cells in the absence and presence of GSH (1 mM) for 24 h, and DNA damage was analyzed by comet assay (alkaline comet: n = 35, 56, 50; neutral comet: n = 46, 88, 55, respectively). Statistical significance was determined using the Kruskal–Wallis test. *p < 0.05, **p < 0.01. n.s., not significant

3.3. CD30 signaling triggers chromosomal instability in an adult T‐cell leukemia/lymphoma cell line

Double‐strand breaks increase the risk of chromosomal aberrations. To examine the relationship between CD30 signaling and chromosomal instability, we performed a comparative genomic hybridization (CGH) analysis with 180 K probes to detect copy number alterations (CNAs) throughout the entire genome region (Figure 3A). HUT102 cells were stimulated by CD30L for 72 h to induce DSBs via increased intracellular ROS and then cultured for an additional 17 days to evaluate subclonal expansions with a growth advantage caused by chromosomal aberrations. We compared the genomic DNA of HUT102 cells after expansion following CD30 stimulation with pre‐CD30 stimulation cells (day 0). HUT102 cells expanded following CD30 stimulation acquired significantly more CNAs than HUT102 cells expanded with mock stimulation (Figure 3B,C). Furthermore, we examined whether an increase in intracellular ROS induced by CD30 signaling promotes CNAs. Treatment with the scavenging antioxidant GSH, which neutralized CD30 signaling‐induced intracellular ROS and suppressed subsequent DSBs, significantly depressed CD30 signaling‐induced CNAs (Figure 3C). The average number of CNA regions and genes located in the CNA regions induced by CD30 signaling in three independent experiments was 35 and 89, respectively. The number of CNAs that overlapped in three independent experiments was six. The overlapping regions with copy number gain (CNG) were 3p22.1, 12q24.22, 20p13, and 21p11.2, while the overlapping regions with copy number loss (CNL) were 2q37.1, and 14q32.2 (Figure 3D). The number of genes located in CNAs that overlapped in three independent experiments was 10 (Figure 3E). The common genes located in CNG regions were PHETA1 (12q24.12), SIRPB2, SIRPD, SIRPB1, SIRPG, and SIRPG‐AS1 (20p13), while 3p22.1 and 21p11.2 were intergenic regions. The common genes located in the CNL regions were ALPP, ECEL1P2, ALPG (2q37.1), and EML1 (14q32.2). Although these 10 genes are not representative genes located in CNAs in in vivo ATL cells, amplification of the SIRPB1 gene is reported as a potential oncogenic event in prostate cancer. ²² The proto‐oncogenes or tumor suppressor genes with CNAs were detected in each experiment and included SKI, ARID1B, and RIN3, among others (Table S2). Thus, these genes with CNA might contribute to clonal expansion. Taken together, these results indicate that CD30 signaling induces chromosomal instability through DSBs via an increase of intracellular ROS and CNAs with clonal expansion in an ATL cell line.

FIGURE 3.

The effect of CD30 stimulation on induction of copy number alterations (CNAs) in an adult T‐cell leukemia/lymphoma (ATL) cell line. (A) The protocol for detection of CNAs induced by CD30 signaling using comparative genomic hybridization (CGH). (B) HUT102 cells were stimulated with mock‐CHO or CD30L‐CHO cells in the absence and presence of glutathione (GSH) (1 mM) for 72 h and then expanded for an additional 17 days. The representative images of CNAs on the genomes of HUT102 cells stimulated with mock‐CHO or CD30L‐CHO cells in the absence and presence of GSH. The genomes of HUT102 cells after stimulation with CD30L‐CHO or mock‐CHO cells were compared with the genome of HUT102 cells before stimulation with CD30L‐CHO/mock‐CHO cells. The horizontal numbers indicate chromosome numbers. The gain and loss regions are indicated as blue and red, respectively. (C) The number of CNAs induced by mock stimulation or CD30 stimulation in the absence and presence of GSH in HUT102 cells are shown as bar graphs. Data for HUT102 cells are shown as mean ± SD of three independent experiments. Asterisks indicate statistical significance (**p < 0.01). (D) The number of CNGs and CNLs detected by CGH in each experiment. The number of CNAs from three independent experiments is shown as Venn diagrams (CNG: left, CNL: right). The CNG and CNL regions common to the three independent experiments are shown as the overlapping region in the center. (E) The number of genes located in CNGs and CNLs detected by CGH in each experiment. The genes identified in 3 experiments are shown as Venn diagrams (CNG: left, CNL: right). The genes observed in all three independent experiments are indicated by the overlapping region in the center.

3.4. CD30 signaling triggers chromosomal instability in primary CD30 ⁺ ATL cells

Next, we evaluated whether CD30 signaling induces CNAs with clonal expansion using primary ATL cells (Figure 4A). Primary CD30⁺ATL cells were stimulated with CD30L‐TIG‐1 or mock‐TIG‐1 cells for 35 days. ⁹ Chronic CD30 stimulation significantly increased CNAs throughout the entire genome of primary CD30⁺ATL cells in three ATL patients compared with the mock condition (Figure 4B,C). The average number of CNA regions and genes located in these CNA regions induced by CD30 signaling in three patient samples was 5.3 and 205, respectively. Furthermore, the CNA regions induced by CD30 signaling overlapped with frequent focal CNA regions in ATL. ¹⁴ The regions with CNG were 3q26.33‐q29 in chronic#6 and 6p25.3 in chronic#7, while the regions with CNL were 1p21.3, 6q21, 7p21.3‐p14, and 10q26.13‐q26.2 in chronic#5 and 2q34 and 7q34 in chronic#7 (Figure 4C). In contrast, CNA regions induced by the mock condition did not overlap with frequent focal CNA regions. Thus, the CNAs induced by CD30 signaling suggest that they contributed to clonal expansion. Taken together, the above results indicate that CD30 signaling induces chromosomal instability and triggers CNAs with clonal expansion in primary CD30⁺ATL cells, suggesting that CD30 signaling is one of the oncogenic factors of ATL progression with CNAs.

FIGURE 4.

The effect of CD30 stimulation on the induction of copy number alterations (CNAs) in primary CD30⁺ATL cells. (A) The protocol for detection of CNAs induced by CD30 signaling using comparative genomic hybridization (CGH) analysis. Primary CD30⁺ATL cells were sorted by CD4, CD25, CADM1, and CD30 using FACSAria III. Representative images of CD30⁺ATL cells stimulated by CD30L (right) or mock (left) for 35 days are shown at the upper right. Scale bar, 100 μm. (B) Sorted primary CD30⁺ATL cells were stimulated by CD30L‐TIG‐1 or mock‐TIG‐1 cells for 35 days, as described in the Materials and Methods. Representative images of CNAs on the genomes of CD30⁺ATL cells stimulated by CD30L‐TIG‐1 or mock‐TIG‐1 cells are shown. The genomes of CD30⁺ATL cells after stimulation with CD30L‐TIG‐1 or mock‐TIG‐1 cells were compared with the genome of peripheral blood mononuclear cells (PBMCs) derived from the adult T‐cell leukemia/lymphoma (ATL) patient. The horizontal numbers indicate chromosome numbers. The gain and loss regions are indicated as blue and red, respectively. CNL of 6q21 is shown as an example CNA present in the chronic#5 sample. (C) The number of CNA regions induced by CD30 stimulation or mock stimulation in primary CD30⁺ATL cells is shown as a bar graph. The common CNAs detected in CD30L‐stimulated cells and mock‐stimulated cells were subtracted because they were the CNAs that occurred before stimulation with CD30L‐TIG‐1/mock‐TIG‐1 cells. The numbers at the bottom of the bar graph represent CNA regions induced by CD30L‐TIG‐1 or mock‐TIG‐1 cells in chronic#5, #6, and #7. The statistical significance of CNA number between CD30L‐stimulated cells and mock‐stimulated cells in each ATL individual was determined by a paired t‐test. The gain and loss regions are indicated as blue and red, respectively. Asterisks indicate frequent focal CNA regions in in vivo ATL.

Expanded CD30⁺ ATL cells from the ATL patient (chornic#5) sustained CD30 expression under the experimental conditions in Figure 4 (Figure S1A). Thus, the result indicates that CD30 expression in sorted CD30⁺ ATL cells is permanent or remains inducible.

3.5. Analysis of focal copy number alterations and gene mutations in CD30 ⁺ ATL and CD30⁻ATL cells of adult T‐cell leukemia/lymphoma patients

It is reported that CD30L is ubiquitously expressed by mast cells, eosinophils, neutrophils, macrophages, and activated lymphocytes. ²³ Thus, CD30‐expressing ATL cells are stimulated by CD30L‐expressing cells in the body. Therefore, we examined whether CD30 expression on ATL cells is correlated with genomic instability using the CGH + SNP method equipped with 400 K probes and whole exome sequencing (Figure 5A). CGH analysis revealed that the number of focal CNAs in CD30⁺ATL cells was significantly higher than in CD30⁻ATL cells in each ATL individual (Figure 5B). Arm‐level CNAs, defined by a CNA being longer than half the length of the chromosome arm, were not different between CD30⁺ATL cells and CD30⁻ATL cells (Figure S4). The frequency of arm‐level CNAs showed a tendency that was similar to a previous report. ¹³ The high‐frequency genes with focal CNG were DUSP22, IRF4, EXOC2, HUS1B (6p25‐p23), PLGRKT, CD274 (9p24.1), and Notch1 (9q34.3). In contrast, the high‐frequency genes with focal CNL were NRG1 (8p12), PTCH1 (9q22.32), NRXN3 (14q31.1), PUDP (Xp22.31), IMMP2L (7q31.1), and CBLB (3q13.11). Although NRG1 and PTCH1 are located in copy number variation regions, and PUDP is located in a fragile site, it has not been reported that these genes are high‐frequency genes with focal CNL in ATL (Figure 5D). Whole exome sequencing showed that the number of gene mutations in CD30⁺ATL cells was slightly higher than in CD30⁻ATL cells for each ATL individual (Figure 5C). High‐frequency mutated genes of CD30⁺ATL cells and CD30⁻ATL cells showed a similar tendency, including for PRKCB, PLCG1, TTN, TBL1XR1, COBL, and others (Figure S5). Next, we evaluated whether CD30⁺ATL cells were composed of subclones with additional gene mutations compared with CD30⁻ATL cells using PyClone‐VI. ¹⁸ Most mutations were shared between CD30⁺ATL and CD30⁻ATL cells, suggesting that CD30⁺ATL and CD30⁻ATL cells have a common ancestor (Figure S6). Clustering analyses by PyClone‐VI calculated several clusters based on gene mutations in CD30⁺ATL and CD30⁻ATL cells in individual ATL patients. These clusters represent clonal populations in CD30⁺ATL and CD30⁻ATL cells. In acute#3, CD30⁻ATL cells were composed of two populations, which are cluster1 and cluster2. The cellular prevalences, which are calculated by PyClone‐VI and represent each proportion of the cluster, were 0.996 for cluster1 and 0.703 for cluster2 when the maximum proportion was set to 1.000 (Figure 5E, left). These results indicate that cluster1 is the ancestor and cluster2 is derived from cluster1 in acute#3. In comparison, CD30⁺ATL cells in acute#3 were composed of three populations: cluster1, cluster2, and cluster3. The cellular prevalences of cluster1, cluster2, and cluster3 were 1.000, 0.791, and 0.675, respectively (Figure 5E, right). These results indicate that cluster3 occurs only in CD30⁺ATL cells and is derived from cluster2. Therefore, CD30⁺ATL cells were composed of subpopulations with additional gene mutations compared with CD30⁻ATL cells in acute#3. Clustering analyses by PyClone‐VI using CD30⁺ATL and CD30⁻ATL cells from eight ATL individuals, chronic#1–4, and acute#1–4, showed that the number of clusters in CD30⁺ATL cells was slightly higher than in CD30⁻ATL cells (Figure 5F). Taken together, CD30⁺ATL cells were composed of subclones with additional gene mutations compared with CD30⁻ATL cells, suggesting that CD30 expression on ATL cells is correlated with genomic instability.

FIGURE 5.

Analysis of focal copy number alterations (CNAs) and gene mutations in CD30⁺ATL and CD30⁻ATL cells of adult T‐cell leukemia/lymphoma (ATL) patients. (A) Representative images of sorted cell fractions in peripheral blood mononuclear cells (PBMCs) from an ATL patient (acute#3). CD30⁺ATL cells, CD30⁻ATL cells, and normal lymphocytes were sorted by CD4, CD25, CADM1, and CD30. The genomic DNA purified from each cell fraction was analyzed by CGH + SNP and whole exome sequencing. (B) The number of CNAs in CD30⁺ATL and CD30⁻ATL cells detected by CGH + SNP is shown as a bar graph (chronic#1–4, acute#1–4). Statistical significance of CNA number between CD30⁺ATL cells and CD30⁻ATL cells in each ATL individual was determined by a paired t‐test. (C) The number of gene mutations (indel, missense, nonsense and splice acceptor, and donor site) in CD30⁺ATL and CD30⁻ATL cells is shown as a bar graph (chronic#1–4, acute#1–4). The statistical significance of mutation number between CD30⁺ATL cells and CD30⁻ATL cells in each ATL individual was determined by a paired t‐test. (D) Frequency of genes located in focal CNAs in CD30⁺ATL and CD30⁻ATL cells (focal CNG: upper, focal CNL: lower). Genes detected in three or more of the eight cases are shown as bar graphs in order of observed frequency. (E) Representative images of clonal populations calculated by PyClone‐VI in CD30⁺ATL (right) and CD30⁻ATL (left) cells from acute patient #3. The values of each cluster indicate cellular prevalence. The trees connecting clusters (bottom) represent each clonal phylogeny. (F) Clusters based on gene mutations were calculated by PyClone‐VI in CD30⁺ATL and CD30⁻ATL cells from each of the 8 ATL patients, chronic#1–4 and acute#1–4. The number of clusters observed in each cell type is shown as a bar graph. Data are shown as mean ± SD. Asterisks indicate statistical significance (*p < 0.05).

3.6. Focal copy number alterations and mutations of DNA double‐strand break repair‐related genes in adult T‐cell leukemia/lymphoma cells

We compared the relationship between focal CNAs or gene mutations and 268 DSB repair‐related genes from the Molecular Signatures Database (GOBP DOUBLE STRAND BREAK REPAIR, GO: 0006302). This analysis revealed that most ATL cases have acquired focal CNAs or gene mutations in DSB repair‐related genes in the chronic and acute type states (Figure 6A,B). Although it was not significantly different, CD30⁺ATL cells acquired additional focal CNAs or additional gene mutations in DSB repair‐related genes compared with CD30⁻ATL cells in certain ATL patients (Figure 6A,B). In contrast, focal CNLs of DSB repair‐related genes in acute‐type patients were significantly enriched compared with those in chronic‐type patients (Figure 6A, upper panels). Therefore, ATL progression could be related to the enrichment of focal CNLs of DSB repair‐associated genes. Some ATL patients had serious gene mutations or focal CNLs (Figure 6A,B), including BARD1, FANCM, LIG4, Rad21, Rad51B, and XRCC3. The focal CNLs and mutations of these genes could lead to impaired DSB repair efficiency.

FIGURE 6.

Focal copy number alterations (CNAs) and gene mutations of DNA double‐strand break (DSB) repair‐related genes in CD30⁺ATL and CD30⁻ATL cells of ATL patients. (A) Focal copy number loss (CNL) (upper) and focal Copy number gain (CNG) (lower) of 268 DSB repair‐related genes extracted from the Molecular Signature Database (GOBP DOUBLE STRAND BREAK REPAIR, GO: 0006302) in CD30⁺ATL (right panel) and CD30⁻ATL (left panel) cells. Statistical significance in the enrichment of DSB‐related genes located in focal CNLs between chronic#1–4 and acute#1–4 was determined by the Fisher extract test. (B) Gene mutations (indel, missense, and nonsense) of 268 DSB repair‐related genes in CD30⁺ATL (right panel) and CD30⁻ATL (left panel) cells. n.s., not significant

4. DISCUSSION

In this study, we showed the induction of chromosomal aberrations by CD30 signaling in ATL cells. CD30L binding of CD30 recruits TRAF1‐3 and 5, which are adaptor molecules that bind the intracellular domain of CD30, and then directly stimulates the NF‐κB pathway to promote pro‐survival signals. We previously reported that CD30 signaling stimulates processing of p100 to p52 in the alternative pathway of NF‐κB and activation of the JNK, p38 MAPK, and PI3K pathways in ATL cells. These responses were undetected in Jurkat cells. ⁹ Here, we found that CD30 signaling induced an increase in intracellular ROS in ATL cells. It is reported that TNFR signaling via TNF and CD40 signaling via CD40 ligand induce increased intracellular ROS. TNFR and CD40 are TNFR superfamily genes along with CD30. ²⁴ , ²⁵ It has been reported that TNFR and CD40 both need to recruit TRAFs to increase intracellular ROS. We previously showed that CD30 binds CD30L and internalizes with CD30L in a trogocytic manner, and their complexes with TRAF2 form a CD30 signalosome. ¹⁹ Thus, it is thought that TRAF2 increase intracellular ROS in CD30 signaling.

DNA double‐strand breaks can lead to chromosomal instability if they are not repaired correctly. The risk of chromosomal instability is particularly high during mitosis. When cells encounter DSBs during interphase, they are able to arrest the cell cycle; during mitosis, however, cells no longer arrest but prioritize completion of cell division over repairing DNA damage. ² We previously reported that CD30 signaling activates cell cycling and cell proliferation in primary CD30⁺ATL cells. Taken together, the simultaneous induction of mitotic entry and DSBs can trigger chromosomal instability, supporting the role of CD30 signaling in triggering chromosomal instability. In fact, we demonstrated that CD30 signaling induced chromosomal instability in primary CD30⁺ATL cells from three ATL patients. Using an in vivo mouse model, Sperling et al. showed that chronic CD30 signaling induces lymphomagenesis, indicating that CD30 signaling functions as an oncogenic signal in vivo. ²⁶ The mechanism underlying the induction of chromosomal aberrations shown in this study supports the results of the in vivo lymphomagenesis model.

The CNA regions induced by CD30 signaling overlapped with frequent focal CNA regions in ATL. The CNL of 6q21 was induced by CD30 signaling in primary CD30⁺ATL cells in chronic#5. The loss of PRDM1, which is transcriptional repressor, located in 6q21 occurs not only in ATL but also in diffuse large B cell lymphoma, aggressive NK‐cell leukemia, and extranodal NK‐cell lymphoma of nasal type. ²⁷ , ²⁸ , ²⁹ A CNG of 3q29 and 6p25.3 was induced by CD30 signaling in primary CD30⁺ATL cells in chronic#6 and chronic#7, respectively. The gain of PAK2, located in 3q29, facilitates CADM1‐mediated stromal interactions and promotes the survival of ATL cells. ³⁰ PAK2 amplification is also a gene signature that predicts the prognosis of oral squamous cell carcinoma. ³¹ The gain of IRF4, located in 6p25.3, occurs with especially high frequency in ATL. IRF4 is regulated by a CD30/NF‐κB positive feedback loop in peripheral T‐cell lymphoma (PTCL). ³² These results support the hypothesis that CD30 signaling is an oncogenic signal with clonal evolution during ATL progression.

We evaluated the correlation between the number of CNAs and/or genetic mutations and CD30 expression levels in primary CD30⁺ATL cells and found no correlation in this study (Figure S7). However, the sample size may be too small to examine properly, and further analysis is needed in this regard.

We found that ATL cells accumulate focal CNLs of DSB repair‐related genes according to disease progression. The simultaneous promotion of DSBs by CD30 signaling and impaired DSB repair function due to CNAs or gene mutations in the DSB repair pathways are causative factors of chromosomal instability in ATL progression.

We elucidated new oncogenic roles of CD30 in ATL progression (Figure S8). The results strengthen the possibility that CD30 is a molecular target during ATL progression. Brentuximab vedotin (BV), an anti‐CD30 monoclonal antibody conjugated with auristatin E, has been highly effective in a broad range of CD30⁺ lymphomas, especially classical Hodgkin's lymphoma and anaplastic large cell lymphoma. We previously reported that BV depletes CD30⁺ cells in peripheral blood mononuclear cells from HTLV‐1‐infected individuals, including ATL individuals. ⁹ The in vivo effectiveness of BV in mice bearing HTLV‐1‐infected cell lines was also reported. ³³ BV plus the CHP (cyclophosphamide + doxorubicin + prednisone) regimen has achieved superior overall survival compared to the CHOP (cyclophosphamide + doxorubicin + vincristine + prednisone) regimen in CD30⁺ PTCL including ATL. ³⁴ However, due to the small number of ATL patients recruited in this clinical trial, the efficacy of BV in CD30‐expressing ATL remains unknown. Although further study is necessary to optimize the use of BV in ATL, Oka et al. and Baba et al. recently reported successful treatment with BV of a relapsed and refractory ATL patient and an ATL patient with strong CD30‐expressing ATL cells, respectively. ³⁵ , ³⁶ Taken together, CD30 represents a promising therapeutic target for ATL.

In conclusion, we demonstrated that CD30 signaling induced chromosomal instability through DSBs via an increase of intracellular ROS. These results strongly highlight that endogenous signaling by CD30 is an oncogenic signal that promotes ATL progression and that CD30 is a promising target for treating aggressive ATL (i.e., acute and lymphoma types) and for inhibiting disease progression in indolent ATL (i.e., smoldering and chronic types).

FUNDING INFORMATION

JSPS KAKENHI Grant Numbers 19K16580, 22K15420, 20K07379, and JP16H06277. Japan Agency for Medical Research and Development Grant Numbers 20fk0108126h0001 and JP22fk0108124. A research grant from Takeda Pharmaceutical.

CONFLICT OF INTEREST

M. Nakashima and K. Uchimaru received a research grant from Takeda Pharmaceutical. The other authors do not have any conflicts of interest to declare.

ETHICS STATEMENT

Approval of the research protocol by an Institutional Reviewer Board: This study was approved by the Research Ethics Committee of The University of Tokyo (Permission number 19‐220) and was performed in accordance with the Declaration of Helsinki.

Informed Consent: Written informed consent was obtained from all donors who provided samples.

Registry and the Registration No. of the study/trial: N/A.

Animal Studies: N/A.

Supporting information

Figures S1‐S8

Tables S1‐S2

Appendix S1

Nakashima M, Utsunomiya A, Watanabe T, Horie R, Uchimaru K. The oncogenic driving force of CD30 signaling‐induced chromosomal instability in adult T‐cell leukemia/lymphoma. Cancer Sci. 2023;114:1556‐1568. doi: 10.1111/cas.15706