SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

The morphological discrimination of leukemic from non-leukemic T cells is often difficult in adult T-cell leukemia (ATL) as ATL cells show morphological diversity, with the exception of typical “flower cells.” Because defects in the expression of CD3 as well as CD7 are common in ATL cells, we applied multi-color flow cytometry to detect a putative leukemia-specific cell population in the peripheral blood from ATL patients. CD4+CD14 cells subjected to two-color analysis based on a CD3 vs CD7 plot clearly demonstrated the presence of a CD3dimCD7low subpopulation in each of nine patients with acute-type ATL. The majority of sorted cells from this fraction showed a flower cell-like morphology and carried a high proviral load for the human T-cell leukemia virus type 1 (HTLV-I). Genomic integration site analysis (inverse long-range PCR) and analysis of the T cell receptor Vβ repertoire by flow cytometry indicated that the majority of leukemia cells were included in the CD3dimCD7low subpopulation. These results suggest that leukemic T cells are specifically enriched in a unique CD3dimCD7low subpopulation of CD4+ T cells in acute-type ATL. (Cancer Sci 2011; 102: 569–577)

Adult T-cell leukemia (ATL) is a malignant disorder caused by human T-cell leukemia virus type 1 (HTLV-I)(1) and is characterized clinically by generalized lymphadenopathy, hepatosplenomegaly, skin lesions, hypercalcemia and a characteristic morphology termed “flower cells.” Importantly, ATL is one of the most incurable lymphoid malignancies. This disease is endemic to several regions in the world, including sub-Saharan Africa, the Caribbean basin, South America and Japan, and 10–20 million people are estimated to be infected by this virus worldwide.(2,3)

Evaluation of the response after chemotherapy for ATL partly depends on the proportion of ATL cells in the peripheral blood. However, the morphological diversity of ATL cells may lead to inaccurate estimations. Accurate estimation of the chemotherapeutic effect is pivotal in clinical practice because ATL cells often become chemoresistant, even during chemotherapy. Methods to detect ATL cells with greater precision than morphological examination are therefore required.

Aberrant expression of cell-surface antigens in myeloid/lymphoid leukemia cells has been studied extensively.(4–6) Using fluorescence-activated cell sorting (FACS) analysis, gating cells with diminished CD45 expression in acute myeloid/lymphoid leukemia is widely used for purifying leukemia cells. However, in ATL there are only limited data regarding the identification of transformed leukemia cells by similar methods. Previous studies indicated that most ATL cells lack CD7 and exhibit diminished CD3 expression.(7–10) Although a study using CD3 gating by FACS analysis has indicated that ATL cells were distinguishable from normal lymphocytes as a CD3low population,(10) these cells were not well characterized as ATL cells.

In the present study, we focused on the enrichment of ATL cells by constructing CD3 vs CD7 plots from multi-color FACS. CD3dimCD7dim and CD3dimCD7low cells were extensively studied and compared with normal control samples. Taken together, our data suggest that ATL cells are purified in CD3dimCD7low subpopulations. The purification of ATL cells by FACS may therefore allow monitoring of disease activity and yield insight into the biology of this disease.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Cell lines and patient samples.  TL-Om1, a HTLV-I-infected cell line, was provided by Dr. Toshiki Watanabe (The University of Tokyo), and was cultured in RPMI 1640 medium containing 10% fetal bovine serum. Peripheral blood samples were collected from patients admitted to our hospital (Research Hospital, Institute of Medical Science, The University of Tokyo, Tokyo, Japan) during the period from August 2009 to April 2010 with written informed consent. All patients were diagnosed with acute-type ATL according to Shimoyama’s criteria.(8) Blood samples were collected before treatment using the LSG15 protocol(11) or during the recovery phase between chemotherapy sessions. Samples collected from five healthy volunteers (median age, 45 years) were used as normal controls. The present study was approved by the institutional review board of our hospital.

Flow cytometry and cell sorting.  Peripheral blood mononuclear cells (PBMC) were isolated from heparin-treated whole blood by density gradient centrifugation using Lymphoprep (Axis-Shield, Dundee, UK) and subsequently suspended in phosphate-buffered saline (PBS) containing 5% mouse serum (DAKO, Glostrup, Denmark) for prevention of nonspecific antibody binding. Cells were stained using a combination of phycoerythrin (PE)-CD7, PE-Cy7-CCR4, allophycocyanin (APC)-CD25, APC-Cy7-CD3, Pacific Blue-CD4 and Pacific Orange-CD14. Pacific Orange-CD14 was purchased from Caltag-Invitrogen (Carlsbad, CA, USA). All other antibodies were obtained from BD BioSciences (San Jose, CA, USA). Propidium iodide (PI; Sigma, St Louis, MO, USA) was added to the samples to stain dead cells immediately prior to FACS analysis. Cells were also stained with APC-FoxP3 (eBioscience, San Diego, CA, USA) using intracellular staining methods as previously described.(12) A TCR-Vβ repertoire kit (Beckman Coulter, Miami, FL, USA) was used for T-cell receptor (TCR) Vβ repertoire analysis according to the manufacturer’s instructions. A BD FACS Aria (BD Immunocytometry Systems, San Jose, CA, USA) was used for all multi-color FACS analysis and cell sorting. Data were analyzed using FlowJo software (Treestar, San Carlos, CA, USA).

Quantification of HTLV-I proviral load by real-time quantitative polymerase chain reaction (PCR).  The HTLV-I proviral load in PBMC was quantified by real-time quantitative polymerase chain reaction (PCR; TaqMan method) using the ABI Prism 7000 sequence detection system (Applied Biosystems, Foster City, CA, USA) as previously described.(13) Briefly, a total of 50 ng of genomic DNA was extracted from human PBMC using a QIAamp DNA blood Micro kit (Qiagen, Hilden, Germany). Triplicate samples of the DNA were amplified. Each PCR mixture containing a HTLV-I pX region-specific primer pair at 0.1 μM (forward primer 5′-CGGATACCCAGTCTACGTGTT-3′ and reverse primer 5′-CAGTAGGGCGTGACGATGTA-3′), FAM-labeled probe at 0.1 μM (5′- CTGTGTACAAGGCGACTGGTGCC-3′) and 1× TaqMan Universal PCR master mix (Applied Biosystems) were subjected to 50 cycles of denaturation (95°C, 15 s) and annealing to extension (60°C, 1 min), following an initial Taq polymerase activation step (95°C, 10 min). The RNase P control reagent (Applied Biosystems) was used as an internal control for calculation of the input cell number (using VIC reporter dye). DNA extracted from TL-Om1 and normal human PBMC were used as positive and negative controls, respectively. The HTLV-I proviral load (%) was calculated as the copy number of the pX region per input cell number. To correct the deviation of acquired data in each experiment, data from TL-Om1 (positive control) were adjusted to 100% and the sample data was corrected by proportional calculation accordingly.

Inverse long PCR.  For clonality analysis, inverse long PCR was performed. First, 1 μg of genomic DNA extracted from the FACS-sorted cells was digested with EcoRI, HindIII and PstI at 37°C overnight. Purification of DNA fragments was performed using a QIAEX2 gel extraction kit (Qiagen). The purified DNA was self-ligated with T4 DNA ligase (Takara Bio, Otsu, Japan) at 16°C overnight. The circular DNA obtained from the EcoRI digestion fragment was then digested with MluI, which cuts the pX region of the HTLV-I genome and prevents amplification with the viral genome. Inverse long PCR was performed using Takara LA Taq polymerase (Takara Bio). The primer pairs for the EcoRI-treated template were: forward primer 5′-TGCCTGACCCTGCTTGCTCAACTCTACGTCTTTG-3′ and reverse primer 5′-AGTCTGGGCCCTGACCTTTTCAGACTTCTGTTTC-3′. For the HindIII-treated group, forward primer 5′-TAGCAGGAGTCTATAAAAGCGTGGAGACAG-3′ and reverse primer 5′-TGGGCAGGATTGCAGGGTTTAGAGTGG-3′ were used. For the PstI-treated group, forward primer 5′-CAGCCCATTCTATAGCACTCTCCAGGAGAG-3′ and reverse primer 5′-CAGTCTCCAAACACGTAGACTGGGTATCCG-3 were used. Each 50-μL reaction mixture contained 0.4 mM of each dNTP, 25 mM MgCl2, 10× LA PCR buffer II containing 20 mM Tris–HCl and 100 mM KCl, 0.5 mM primer, 2.5 U LA Taq polymerase and 50 ng of the processed genomic DNA. The reaction mixture of the EcoRI- or PstI-treated group was subjected to 35 cycles of denaturation (94°C, 30 s) and annealing to extension (68°C, 8 min). For the HindIII group, the PCR conditions were denaturation (98°C, 30 s), annealing to extension (64°C, 10 min) for 5 cycles, followed by 30 cycles of denaturation (94°C, 30 s), annealing (64°C, 3 min) and extension (72°C, 15 min). Following PCR, the products were subjected to electrophoresis in 0.8% agarose gels. In the CD3dimCD7low subpopulation from which a sufficient amount of DNA was extracted, PCR were performed in duplicate.

Cytospin and May–Giemsa staining.  Cells enriched by cell sorting were washed twice with PBS. Aliquots of 100 μL of the cell suspension were mixed with 20 μL of 10% bovine serum albumin. The mixtures were centrifuged at 20g for 5 min onto glass slides. The fixed cells were air-dried and then subjected to May–Giemsa staining.

Statistical analyses.  Data are expressed as the means ± standard deviation (SD). One-way analysis of variance (anova) was used for statistical analyses, and P < 0.05 was taken to indicate statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Multi-color FACS, including CD3 vs CD7 plots, in patients with acute-type ATL.  We constructed a gating procedure for flow cytometric analysis of acute-type ATL cells using a combination of CD3 and CD7. Figure 1A shows the representative flow cytometric data of an ATL sample (from patient no. 2 in Table 1). Dead cells (PI positive) were initially excluded on the forward scatter (FSC) vs PI plot. Next, monocytes (CD4dim CD14+) were excluded on the CD4 vs CD14 plot. After CD4+ T lymphocytes were gated on the CD3 vs CD4 plot, a CD3 vs CD7 plot was constructed. Based on the cell density and fluorescence intensity of CD3 and CD7, we designated three subpopulations on this plot: CD3highCD7high, CD3dimCD7dim and CD3dimCD7low (Fig. 1A). Using the same gating procedure, we analyzed nine patients with acute-type ATL and five normal controls (Fig. 1B). The patient characteristics analyzed in the present study are shown in Table 1. In normal controls, the expression pattern of CD3 vs CD7 was similar. The highest cell density was observed in the CD3highCD7high subpopulation, and the CD3dimCD7dim subpopulation was observed adjacent to it. The CD3dimCD7low subpopulation was a minor but distinct subpopulation. In contrast, the highest cell density was observed in the CD3dimCD7low subpopulation in all acute-type ATL samples except for patient no. 4, from whom the sample was obtained under conditions of well-controlled ATL during chemotherapy. These subpopulations were distinct but the expression pattern of the CD3 vs CD7 plot, such as the degree of downregulation of CD3 and CD7, was variable among patients. The proportion of the CD3dimCD7low subpopulation was significantly higher in acute-type ATL CD4+ lymphocytes than in normal controls (Fig. 1C).

image

Figure 1.  CD3 vs CD7 plots from FACS analysis of patients with acute-type adult T-cell leukemia (ATL) and normal controls. (A) Representative flow cytometric analysis of a patient with acute-type ATL (patient no. 2). The CD3 vs CD7 plot in CD4+ cells was constructed according to the gating procedure shown in this figure. In the plot, we designated three subpopulations: CD3highCD7high, CD3dimCD7dim and CD3dimCD7low. (B) Flow cytometric profile of the CD3 vs CD7 plot in patients with acute-type ATL and normal controls. (C) Percentages of CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations in CD4+ T cells in patients with acute-type ATL and normal controls. Each line represents an individual sample. ATL group, n = 9; control group, n = 5; FSC, forward scatter; PI, Propidium iodide.

Download figure to PowerPoint

Table 1.   Clinical profile of nine acute-type ATL patients in the present study
No.AgeSexWBC (/μL)Lymph (%)ATL cells† (%)Organ involvement
  1. †Proportion of ATL cells in the peripheral blood WBC evaluated by morphological examination. ATL, adult T-cell leukemia; LN, lymph nodes; Lymph, lymphocytes; WBC, white blood cells (normal range, 3500–9100/μL).

160M520015.011.0Skin
269F160043.59.0Liver, LN, pleural effusion
361M18 62024.743.7Liver, uvea
459F64208.50.0Liver, LN, skin
570F29056.02.0Liver, spleen, LN
660F457019.073.0Skin
753F12 21011.052.0LN
874F648016.525.5Liver, spleen, LN
963F34 81021.533.5Liver, spleen, LN, lung

Analysis of the HTLV-I proviral load in CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations.  We next estimated the HTLV-I proviral load by quantitative real-time PCR in each FACS-sorted subpopulation. Representative results from three patients with acute-type ATL (patients no. 6, 7 and 8) are shown in Figure 2. In all patient samples, HTLV-I proviral integration, analyzed by real-time PCR, was detected in all subpopulations. However, the proviral load (%) was significantly higher in CD3dimCD7dim and CD3dimCD7low subpopulations compared with the CD3highCD7high subpopulation. The proviral load of the CD3dimCD7low subpopulation in patients no. 7 and 8 was nearly 200%, indicating integration of two copies of the HTLV-I viral genome and that almost all of the cells were infected with HTLV-I. Similarly, in patient no. 6, the majority of the CD3dimCD7low subpopulation was infected with HTLV-I. A substantial proportion of the CD3dimCD7dim subpopulation was infected with HTLV-I in patients no. 7 and 8, and nearly all the cells in the same subpopulation in patient no. 6 were infected with HTLV-I.

image

Figure 2.  Quantification of the human T-cell leukemia virus type 1 (HTLV-I) proviral load in CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations. Genomic DNA was extracted from the FACS-sorted cells of each subpopulation and subjected to real-time quantitative PCR Representative data of three cases (patients no. 6, 7 and 8) are shown.

Download figure to PowerPoint

Differences in the immunophenotype of CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations in patients with acute-type ATL.  To further characterize these three subpopulations, we next examined CCR4 and CD25 expression in each subpopulation. Representative results of a normal control and a patient with acute-type ATL are shown in Figure 3A. The mean fluorescence intensities (MFI) of CD25 and CCR4 of each subpopulation in all patients with ATL and normal controls are shown in Figure 3B. Both CCR4 and CD25 expression levels were very low and maintained at similar levels throughout all subpopulations in normal control cells and in the CD3highCD7high subpopulation of patients with ATL. In contrast, CCR4 expression was significantly upregulated in the CD3dimCD7dim and CD3dimCD7low subpopulations of patients with ATL compared with the CD3highCD7high subpopulation. The expression of CD25 was also upregulated in these subpopulations but this difference was not significant (P = 0.36). The expression of Forkhead box P3 (FoxP3), a master regulator in the development and function of regulatory T (Treg) cells,(14) was also analyzed in some patients. As shown in Figure 3C, FoxP3 expression in the CD3dimCD7low subpopulations was variably upregulated among patients. In addition, in patient no. 9, FoxP3 was upregulated in the CD3highCD7high and CD3dimCD7dim subpopulations.

image

Figure 3.  Immunophenotypic analysis in CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations. (A) Expression of CCR4 and CD25 in each subpopulation. Representative FACS data of a normal control (no. 1) and a patient with adult T-cell leukemia (ATL) (no. 6) are shown Gray dots, isotype antibody-stained cells; black dots, specific antibody-stained cells. (B) Mean fluorescence intensity (MFI) of CD25 and CCR4 in each subpopulation from all normal controls and patients with ATL. The MFI is shown in arbitrary units defined as follows: MFI of specific antibody/MFI of isotype antibody. Each dot represents a sample *P < 005 by anova. (C) Expression of FoxP3 in each subpopulation. ND, analysis could not be performed in the CD3highCD7high and CD3dimCD7dim subpopulations in patients no. 6 and 7 due to an insufficient number of cells.

Download figure to PowerPoint

Analysis of clonality in the CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations by inverse long PCR.  To further analyze the enrichment of ATL cells in the CD3dimCD7low subpopulation, we estimated clonality in each FACS-sorted subpopulation by inverse long PCR in four patients with acute-type ATL (Fig. 4). An intense band, suggesting a major clone, was detected in the CD3dimCD7low subpopulations in all patients. In the same subpopulation, multiple bands with weak intensity were also observed. As the levels of DNA extracted from the CD3dimCD7low subpopulation were sufficient, we performed duplicate PCR in three patient samples (Fig. 4B–D). Detection of the major bands was consistent, but the presence of the minor bands was variable. In the CD3dimCD7dim subpopulations, bands of the same size as those of the CD3dimCD7low subpopulations were observed, indicating that a distinct population in the CD3dimCD7dim subpopulations belonged to identical clones.

image

Figure 4.  Analysis of clonality in the CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations using inverse long PCR. (A–D) Genomic DNA was extracted from FACS-sorted cells of each subpopulation and subjected to inverse long PCR. Representative data of four cases (patients no. 3, 6, 7 and 8) are shown For the CD3dimCD7low subpopulations of patients no. 3, 6 and 7, PCR was performed in duplicate (black bars). ATL, adult T-cell leukemia.

Download figure to PowerPoint

Clonality in the CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations by flow cytometry-based TCR-Vβ repertoire analysis.  To further confirm clonality and to evaluate the degree of enrichment in each subpopulation, we performed TCR-Vβ repertoire analysis by flow cytometry(15) in three ATL cases. The representative results are shown in Figure 5. In patient no. 3, over 95% of the CD3dimCD7low subpopulation used specific TCR-Vβ (Vβ9) and their proportion was quite low in the CD3highCD7high and CD3dimCD7dim subpopulations. In addition, in the two other cases, over 90% of cells in the CD3dimCD7low subpopulation used the same TCR-Vβ (data not shown). These results indicate that ATL cells are highly purified in the CD3dimCD7low subpopulation.

image

Figure 5.  Clonality in the CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations using flow cytometry-based T-cell receptor (TCR)-Vβ repertoire analysis. (A) Representative data are shown A monoclonal pattern of TCR-Vβ9 expression was evident in the CD3dimCD7low subpopulation of the adult T-cell leukemia (ATL) sample. Representative dot plots of 3 of the 24 TCR-Vβ repertoire (Vβ9, 16 and 17) are shown. (B) Bar graph representation of the data from Figure 5A. The percentages of cells positive for each TCR-Vβ repertoire in the CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations. White bar, CD3highCD7high; gray bar, CD3dimCD7dim; black bar, CD3dimCD7low.

Download figure to PowerPoint

Differences in morphology of the CD3highCD7high, CD3dimCD7dim, and CD3dimCD7low subpopulations in patients with acute-type ATL.  We reviewed the glass-slide specimens of FACS-sorted samples to evaluate the morphology of each subpopulation on the CD3 vs CD7 plots. Representative results for two patients (no. 6 and 7) are shown in Figure 6A. In both patients, atypical lymphocytes with notched nuclei and/or basophilic cytoplasm were observed in all three subpopulations. In contrast, abnormal lymphocytes, including cells with multilobulated nuclei (flower cells) were mainly observed in the CD3dimCD7low subpopulation in patient no. 6 (Fig. 6, left) and in the CD3dimCD7dim and CD3dimCD7low subpopulations in patient no. 7 (Fig. 6, right panel).

image

Figure 6.  Morphology of the CD3highCD7high, CD3dimCD7dim and CD3dimCD7low subpopulations in two representative adult T-cell leukemia (ATL) samples (patients no. 6 and 7). (A) May–Giemsa staining of FACS-sorted cells from each subpopulation from two patients with acute-type ATL. Top, CD3highCD7high subpopulation; middle, CD3dimCD7dim subpopulation; bottom, CD3dimCD7low subpopulation. (B) Percentages of cells with different morphology in each subpopulation. Normal, lymphocytes with normal morphology; atypical, lymphocytes with notched nuclei and basophilic cytoplasm; abnormal, lymphocytes with convoluted, deeply indented or multilobulated flower cells.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

To investigate the characteristics of ATL cells, the purification of tumor cells is essential. In the present study, we successfully discriminated the CD3dimCD7low subpopulation in CD4+ T cells in the peripheral blood of patients with acute-type ATL by constructing a CD3 vs CD7 plot of CD4+ T cells from multi-color FACS (Fig. 1). Previously, Yokote et al.(10) reported that CD3low gating facilitated the discrimination of ATL cells by flow cytometry. If we constructed a CD4 vs either CD3 or CD7 plot, in which the downregulated cell subpopulation was not clearly separated, then we could not define distinct cell subpopulations using the CD3 or CD7 marker alone, because the degrees of downregulation of CD3 and CD7 are variable. It should be noted that the combination of CD3 downregulation and diminished expression or absence of CD7 clearly indicates this subpopulation. In addition, gating-out monocytes in the CD4 vs CD14 plot is important for the CD3 vs CD7 plot because monocytes were CD3/CD7 dull-positive based on the nonspecific binding of the antibody.

A substantial subpopulation of T cells has been reported to be CD7-deficient under physiological(16,17) and certain pathological conditions, including autoimmune disorders and viral infection.(18–22) Consistent with these reports, the present study indicated that the proportion of CD4+CD7- T cells in the peripheral blood of healthy adults is up to 10% (Fig. 1B,C). In ATL samples, the CD3 vs CD7 plot revealed various patterns, which may reflect the differences in clinical characteristics of each patient; however, the CD3dimCD7low subpopulation, which was a minor population in the normal controls, was prominent in all ATL samples (Fig. 1B,C). These results prompted us to study this subpopulation in detail. Estimation of the HTLV-I proviral load by quantitative real-time PCR showed that the majority of cells in the CD3dimCD7low subpopulation were infected with HTLV-I (Fig. 2). Immunophenotypic analysis revealed that the expression of CD25, a common ATL marker,(9,23) and CCR4, reported to be expressed in around 90% of cases of ATL,(24,25) were upregulated in the CD3dimCD7low subpopulations of ATL samples in contrast to normal controls in which both markers were weakly expressed in the equivalent subpopulation (Fig. 3A,B). As several studies indicated that ATL cells originate from CD4+CD25+FoxP3+ Treg cells,(26) we next analyzed FoxP3 expression in each subpopulation. In the CD3dimCD7low subpopulation, the FoxP3 expression levels were variable, consistent with previous reports.(27) In one case, FoxP3 expression was upregulated in the CD3highCD7high and CD3dimCD7dim subpopulations suggesting that they were normal Treg cells.

The analysis of clonality is extremely important for determining whether cells are transformed and Southern blot analysis is usually used to confirm clonality. However, in the present study, the cell number following cell sorting was not sufficient for Southern blotting, and thus inverse long PCR for clonality analysis of HTLV-I-infected cells was used.(25) Studies of four ATL samples revealed clonal expansion of ATL cells in the CD3dimCD7low subpopulations, although minor clones may exist in the population (Fig. 4). When PCR was performed in duplicate, we found that the major bands were consistently detected in all cases. However, the detection of multiple minor bands was not consistent. As reported previously, the inverse long PCR method stochastically amplifies the template originating from small clones.(28,29) The minor bands observed in the present study will contain small clones. However, the presence of nonspecific bands cannot be eliminated.

The inverse long PCR method is commonly used for clonality analysis; however, it cannot quantify the size of major/minor clones and the degree of enrichment in each subpopulation. Therefore, we tested the FACS-based TCR-Vβ repertoire analysis combined with our multi-color FACS system (Fig. 5). In ATL patient no. 3, almost all cells in the CD3dimCD7low subpopulations were clonal cells with TCR-Vβ9. Inverse long PCR analysis in the same patient showed multiple minor bands in the CD3dimCD7low subpopulations (Fig. 4D). These results did not conflict with those of the TCR-Vβ repertoire analysis, as the inverse long PCR method is a more sensitive method for detecting small clones compared with flow cytometry. Taken together, the series of analyses in the present study indicated that the CD3dimCD7low subpopulations consist of highly purified ATL cells in patients with acute-type ATL.

A substantial proportion of cells in the CD3dimCD7dim subpopulation consisted of morphologically abnormal lymphocytes (Fig. 6) that exhibited upregulation of CD25 and CCR4 expression (Fig. 3A,B). Using the inverse long PCR method, a similar band pattern between CD3dimCD7dim and CD3dimCD7low subpopulations was observed in patients no. 6 and 8, suggesting that these cells belonged to the same clone (Fig. 4A,B). However, not all of the cells in this subpopulation were infected with HTLV-I because the HTLV-I proviral load was less than that of the CD3dimCD7low subpopulation (Fig. 2). Thus, at least a small number of the CD3dimCD7dim cells were expected to be ATL cells. Those cells observed in the CD3dimCD7dim subpopulation that were phenotypically different from the CD3dimCD7low subpopulations were of particular interest. We detected a band of the same size on inverse long PCR in the CD3dimCD7dim subpopulations as in the CD3dimCD7low subpopulation. This may have been because the two subpopulations originated from the same clone that evolved from a CD3dimCD7dim to a CD3dimCD7low phenotype. Further studies are required to determine the characteristics of the CD3dimCD7dim subpopulation in greater detail.

The results of the present study indicated that HTLV-I-infected cells distribute from a CD7high to a CD7low subpopulation, although the proportion of HTLV-I-infected cells was remarkably low in the CD3highCD7high subpopulation (Fig. 2). A considerable proportion of cells in the CD3highCD7high subpopulation consisted of morphologically atypical lymphocytes (Fig. 6), but the CD25 and CCR4 levels were not upregulated (Fig. 3A,B). When analyzing the pattern of the inverse long PCR of the CD3highCD7high subpopulations, we observed a difference from those of the CD3dimCD7dim and CD3dimCD7low subpopulations (Fig. 4). In patients no. 6 (Fig. 4B) and 7 (Fig. 4C), the band detected in the CD3highCD7high subpopulation may represent an expanded clone that was not transformed. Most likely, these cells do not represent ATL cells, but oligoclonal HTLV-I-infected lymphocytes. Previous studies indicated that HTLV-I-infected cells undergo transformation through multi-step oncogenesis.(30) A detailed analysis of these three subpopulations may therefore provide some insight into the oncogenesis of HTLV-I-infected cells.

Accurate determination of ATL cells in peripheral blood is critical for estimating the response to chemotherapy. However, as discussed above, morphological studies (Fig. 6) have limitations in their ability to discriminate ATL from non-ATL cells.(31,32). Recently, hematopoietic stem cell transplantation has been explored as a promising treatment to overcome the poor prognosis of this most incurable lymphoid malignancy,(33,34) and monitoring minimal residual disease following hematopoietic stem cell transplantation is more important. Our method of analyzing ATL cells may be particularly useful for monitoring minimal residual disease. Although the CD3dimCD7dim subpopulation in our analysis may have included some ATL cells, this is a minor population in the peripheral blood of patients with acute-type ATL, and it is sufficient for practical use to monitor the CD3dimCD7low subpopulation. Another possible use of our procedure is for the definitive classification of ATL subtypes according to Shimoyama’s criteria.(8) A proportion of abnormal lymphocytes in peripheral blood comprise part of the criteria for ATL-subtype classification but it is sometimes confusing. Our multi-color FACS system may clearly quantify this proportion.

In conclusion, we have constructed a multi-color FACS system to purify ATL cells in the peripheral blood of patients with acute-type ATL. This system may be useful for precisely monitoring the disease during chemotherapy, detecting minimal residual disease and analyzing ATL cells. This system may be of great benefit in investigating oncogenesis in HTLV-I-infected cells.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

The authors thank Dr. Toshiki Watanabe, Dr. Kazumi Nakano and Dr. Tadanori Yamochi (The University of Tokyo) for providing the TL-Om1 cell line and plasmid containing the HTLV-I genome, which were used as standards for quantification of the proviral load. We are also grateful for their substantial technical assistance. We thank Dr. Naofumi Mastsuno (The University of Tokyo) for providing technical assistance regarding multi-color flow cytometry. We are grateful to the hospital staff that have made a commitment to providing high quality care for all of our patients. This study was supported by the Ministry of Education, Culture, Sports, Science and Technology and the Ministry of Health, Labor, and Welfare of Japan.

Disclosure Statement

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

The authors declare no financial conflicts of interest.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References