- Email: [email protected]
Relative expression of different Ikaros isoforms in childhood acute leukemia
Blood Cells, Molecules, and Diseases 41 (2008) 278–283
Contents lists available at ScienceDirect
Blood Cells, Molecules, and Diseases j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y b c m d
Relative expression of different Ikaros isoforms in childhood acute leukemia Alexander N. Meleshko ⁎, Ludmila V. Movchan, Michael V. Belevtsev, Tatjana V. Savitskaja Belarusian Research Center for Pediatric Oncology and Hematology, Minsk, Belarus
a r t i c l e
i n f o
Article history: Submitted 9 June 2008 (Communicated by M. Lichtman, M.D., 24 June 2008) Keywords: Ikaros Dominant-negative Isoform Acute lymphoblastic leukemia
a b s t r a c t Ikaros is a zinc-finger transcriptional factor playing an essential role in lymphoid lineage commitment and differentiation. Animal models and analysis of human Ikaros in leukemic cells demonstrate deregulation of Ikaros expression. Short isoforms with a truncated DNA-binding domain suppress functions of Ikaros in a dominant-negative manner. Previous studies demonstrated that human leukemias are heterogeneous for Ikaros expression. We estimate the relative level of Ikaros mRNA transcripts in 80 childhood ALL cases in comparison with AML and healthy donor groups. We detected eight major isoforms and several minor mutant isoforms in most patients with acute lymphoblastic and myeloid leukemia and in healthy donors, but the relative level of expression varied. The relatively high level of Ik4A isoform, rarely mentioned in previous reports, was detected in all analyzed groups. The ratio between functional and all isoforms was used to determine functional activity of Ikaros. The ratio was significantly less in AML (p = 0.027) and BCR-ABL positive ALL (p = 0.0028) than in healthy bone marrow. We found a negative association between the Ikaros ratio and myeloid coexpression in B-cell ALL, the most prominent was for CD15. The Ikaros ratio positively correlates with CD5 and negatively with CD7 expression in T-ALL. We suggest that an anti-proliferation and anti-activation effect of full-length Ikaros may be mediated through regulation of CD5 and CD7. © 2008 Elsevier Inc. All rights reserved.
Introduction Ikaros is a transcription factor, which plays an important role in controlling hematopoietic, particularly lymphoid cell differentiation, proliferation and function [1,2]. The Ikaros protein, like other proteins of the Ikaros family, contains two separate regions of zinc-finger domains: 4 DNA-binding zinc fingers near the N-terminus and 2 zinc fingers for protein–protein interactions near the C-terminus. The Ikaros gene (ZNFN1A1) contains 7 exons and is transcribed as a number of isoforms due to alternative splicing with alternative use of exons 2–6. At least 11 isoforms of human Ikaros has been described [3] (Fig. 1). Long isoforms (Ik1 to 3) have at least three zinc fingers which are able to bind DNA and considered to be functional. Short isoforms (Ik4 to 8) lack two or more zinc-finger domains, so they cannot bind DNA and impair the function of Ikaros proteins in a dominant-negative (DN) manner. The essential role of Ikaros in the regulation of development from stem cell to mature lymphocytes was established by knock-out experiments in mice. Mutant Ikaros −/− mice have severe lymphoid cell defects [4,5] but heterozygous Ikaros DN +/− mice invariably developed T-cell malignances [6]. Before the onset of transformation mutant mice had a hyper-responsive phenotype to IL-2R and TCR stimulation, and the degree of proliferation correlates with the level of
⁎ Corresponding author. E-mail address: [email protected] (A.N. Meleshko). 1079-9796/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.bcmd.2008.06.006
Ikaros activity [7]. Ikaros was also shown to be involved in apoptosis and cell cycle regulation in lymphocytes [8–10]. These evidence support the idea that Ikaros may function as a hematological tumor suppressor. Analysis of Ikaros isoform expression was performed in several clinical studies in leukemia patients. Overexpression of short DN Ikaros isoforms (Ik4, 7, 8 and/or 6) were detected in infant ALL, childhood, adult B and T-lineage ALL [11–16]. Data about Ikaros expression profile in AML are less frequent. Yagi et al. has detected Ik6 in 7 of 10 cases of M4 and M5 in a group of 24 different AML [9]. In another study, Ik1, 2 and short isoforms were detected in 3 cases (two M2 and one M4) out of 7 AML analyzed [17]. All these and some other studies use reverse-transcription PCR and qualitative analysis in agarose gel or immunobloting to discriminate different isoforms of Ikaros on RNA or protein level. The only quantitative analysis using real-time PCR was performed by Olivero et al. [18]. The authors used two sets of primers and a probe to detect separately Ik1 + 2 and Ik1 + 2 + 4 + 7 + 8 so that they can calculate the ratio between large and short isoforms. Unfortunately, this elegant method does not allow quantifying each isoform and loses sight of Ikx and Ik6, which are not mentioned in that issue. In this study we use RT-PCR and subsequent analysis by polyacrylamide gel electrophoresis to separate and characterize expression of each Ikaros isoform in childhood ALL in comparison with some AML samples and healthy control. We also performed extensive association between Ikaros expression profile and immunophenotype and fusion oncogene expression.
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
279
Materials Patients, samples and cell lines A total of 80 children with ALL were evaluated in the study. 66 patients had B-precursor ALL (60 at diagnosis, 6 at relapse), and 14 patients had T-ALL (13 at diagnosis, 1 at relapse). Diagnosis was based on morphology, cytogenetics, immunophenotype and molecular biology analyses. Age distribution was from 1 month to 18.5 years (median 6.6 years). Leukemic blast count in bone marrow ranged from 53.5% to 99%, median 93%. Nine patients with AML were analyzed. Control groups include 9 healthy donors: 5 bone marrow donors, 1 blood donor, 1 bone marrow iron-deficiency anemia patient and two bone marrow samples from leukemic patients in stable remission. Five cell lines were analyzed for Ikaros isoform expression: B-lineage, IM-9, Raji, T-lineage, Molt-3, Jurkat and myeloid, К562. RNA extraction and cDNA preparation Mononuclear cells were isolated from 2–5 ml of BM sample by Ficoll–Paque centrifugation. Total RNA from fresh cells was extracted using Gen Elute Mammalian Total RNA Miniprep Kit (Sigma-Aldrich, St Louis, MO, USA). Former patient samples were lyzed in TRI Reagent® RNA Isolation Reagent (Sigma) and stored at minus 80 °C before the extraction. Quantity and quality of obtained total RNA were defined by spectrophotometry using Gene Quant RNA/DNA Calculator (Amersham Biosciences/GE Healthcare, UK) and gel electrophoresis. 1 µg of RNA was taken for cDNA synthesis in a 20 µl reaction mix with 5 mM MgCl2, 1 mM dNTP, 25 mM random hexamers (GE Healthcare, UK), 20 U RNAsin (Promega), and 200 U Reverta (Promega) according to standard protocol [19]. RT-PCR analysis and quantification of Ikaros isoforms PCR amplification for Ikaros was first performed with the primer pair of F1 ATGGATGCTGACGAGGGTCAAGAC and R1 TTAGCTCATGTGGAAGCGGTGCTC under the following conditions: 35 cycles of 30 s at 94 °C, 30 s at 62 °C and 1 min at 72 °C, with a final extension step of 7 min at 72 °C. 2 µg of the PCR products from the first step were amplified further by nested PCR, with primers of F2 CCCCTGTAAGCGATACTCCAGATG and R2 GATGGCTTGGTCCATCACGTGGGA. The second PCR step was performed under the following conditions: 40 cycles of 30 s at 94 °C, 90 s at 68 °C, and 2 min at 72 °C, with a final extension
Fig. 2. Polyacrylamide gel electrophoresis analysis of Ikaros gene showing different isoform expression in leukemic samples and healthy controls. MNC — mononuclear cells from normal bone marrow. Isoforms with a 30 bp deletion and a 60 bp insertion are indicated. Mutant isoform denoted with asterisk is represented in details in Fig. 3. In the left part of the figure the profile of the first line, plotted by KODAK 1D v3.6.1, is shown.
step of 7 min at 72 °C. The primer sequences were selected in a previous study [14]. PCR products were primarily examined by 1.5% agarose gel electrophoresis. To identify different Ikaros isoforms we ran PCR products in 8% non-denaturating polyacrylamide gel using Hoefer vertical electrophoresis system (Amersham Biosciences/GE Healthcare, UK) with maximum gel length (24 sm) for better bands discrimination. The gel was stained with ethidium bromide and fixed to a graphic file by Gel Doc 2000 (Bio-Rad, USA). Analysis of the image was performed with KODAK 1D v3.6.1 Scientific Imaging system. This software automatically or manually detects lines and bands, and constructs a graph profile for a line. After subtracting the background, several parameters for each exact band were evaluated, including molecular weight relative to DNA ladder standard and relative intensity — the percent intensity of a band within a specified range of a line. In that way we were able to define the relative level of all Ikaros isoforms. An example of a gel was shown in Fig. 2. Sequence analysis Individual bands of each size for some patients, as well as all unusual bands were cut from the gel and subjected to PCR using the second step primers. Specific PCR products of the right size were purified by polyacrylamide gel electrophoresis and sequenced. Sequencing was performed with BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, USA) on ABI PRISM 3130 automatic genetic analyzer (Applied Biosystems, USA). The obtained sequence was aligned with germline sequences of Ikaros gene exons by BioEdit and MEGA-3 software to identify an isoform. Immunophenotypic analysis
Fig. 1. Schematic representation of human Ikaros isoforms. Zinc fingers are represented as black vertical bars.
Freshly obtained bone marrow samples were subjected to detailed three-color immunophenotyping analysis. The following markers were analyzed: CD45, CD1a, CD2, CD3, CD4, CD8, CD5, CD7, CD10, CD19, CD20, CD22, CD79, CyIgM, sIgM, CD13, CD33, CD15, CD11c,
280
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
MPO, CD34, CD117, TdT, and HLA-DR with a panel of monoclonal antibodies (Becton Dickinson, USA) using FACScan flow cytometer (Becton Dickinson, USA). Cases expressing a given marker on more than 20% of blast cells were considered positive. The mean fluorescent intensity (MFI) for all markers was determined to evaluate the level of marker expression. Results Incidence and relative level of Ikaros isoform expression We performed detailed analysis of different Ikaros isoform expression by nested RT-PCR and polyacrylamide gel electrophoresis. The following PCR product types of isoforms were usually distinguished: Ik1 (893 bp), Ikx (758 bp), Ik2/3 (632 bp), Ik4 (505 bp), Ik5/7 (464–463 bp), Ik4A (370 bp), Ik8 (337 bp) and Ik6 (202 bp) with the primers used in this study. Sequencing analysis, performed for several patients and all band types, confirmed correct isoform definition. In all cases we sequenced a band of 632 bp that was Ik2 but not Ik3. Analysis of a sequence of 464 bp in some cases revealed Ik5, in the other — Ik7. We detect also a 60 bp insertion in exon 2 and a 30 bp deletion in exon 6 for all isoforms, most often for Ik2, Ik8 and Ik6. Mutant isoforms were detected by an incomplete sequence of one of the exons. For statistical analysis we refereed all isoforms to the types listed above on the basis of exon order and the number of zinc fingers. Expression of different Ikaros isoforms was compared between four groups: B-lineage ALL, T-lineage ALL, AML and healthy donors. The presence of predominant isoforms accounted if the band relative intensity was more than 20% of the whole electrophoresis line density. The 20% threshold was chosen, because slack bands with less than 20% intensity was not reliably reproducible in repeated experiments. Incidence of predominant isoforms are presented in the Table 1. We also estimated an average level of each isoform expression for all four groups (Table 2). All Ikaros isoforms were expressed in most patients with acute lymphoblastic and myeloid leukemia and healthy donors, but the relative level of expression varied. The most predominant isoform was Ik2, then Ik4 and Ik1. Some differences are found in isoform distribution between lineages of leukemia and health controls. In normal bone marrow and blood expression of Ik1, Ikx and Ik2 are relatively higher, and there is only a weak or absent expression of Ik6 and Ik5/6 compared to leukemic cells. Ikx is a novel isoform, found and denoted this way by K.J. Payne et al. [20], but in uniform classification Ikx is referred to as Ik3A [19]. This isoform was first described by K.J. Payne et al. as typical for normal hemopoiesis [20] and later they as well show predominant expression in myeloid lineage [21] what corresponds with our data: Ikx found at a higher level in acute myeloid leukemias and healthy controls. In these groups Ikx is more expressed than Ik1, while for ALL this feature is reversed. Surprisingly, we found a widespread isoform of 370 bp lacking exons 3, 5 and 6. The isoform appeared to be that which was described previously [3,20] as Ik4A, so we used this name. In most of the previous studies it was not detected, maybe because the PCR product (protein) of Ik4A is close by length to Ik8 and they are not distinguishable in agarose gel. Ik6, the shortest isoform, was found in 7/68 B-lineage ALL samples, and in these patients only Ik6 was detected as predominant, together with Ik8. In one patient with T-ALL
Table 2 Average level of different Ikaros isoforms expression in leukemia patients and healthy donors
B-ALL T-ALL AML Healthy donors
Ik 1
Ik x
Ik 2/3
Ik4
Ik 5/7
Ik4A
Ik 8
Ik 6
11.1% 7.1% 3.7% 10.3%
6.5% 5.7% 10.8% 18.6%
27.4% 31.2% 26.3% 38.5%
16.3% 17.9% 19.9% 15%
2.3% 2.3% 0.2% 1.3%
14.6% 15.3% 23.4% 5.4%
11.2% 11.5% 10.7% 7.5%
10.8% 9.1% 5.1% 3.5%
only Ik6 expression was found, and one of the AML patients showed a predominantly Ik6 and Ik4A expression. The rarest isoform was Ik5/7 — its expression was not detected at all or at a low level; only in one patient with B-precursor ALL it was 21.4% of the Ikaros transcripts. We never detected the Ik2A described by A. Rebollo [3] in that study. Mutant or unusual Ikaros isoforms Sequencing analysis of PCR products showed modifications of splicing variants with a 30 bp deletion at the 3′ end of exon 6 and a 60 bp insertion at the 3′ end of exon 2, as described previously [11,12,20]. There is evidence that such kind of Ikaros transcript modifications play a role in normal hemopoiesis and T-cell regulation [20,22]. We found the insertion and deletion variants of all isoforms Ik1–Ik6. Well-defined bands were observed for deletion and insertion variants of Ik6, Ik2, and deletion variants of Ik8, Ikx and Ik1. In some cases a prominent band of Ik8del was observed in complete absence of Ik8 without deletion. In all cases these isoforms counted as respective isoforms without insertions and deletions. We revealed some unusual isoforms, resulting most likely from splicing inaccuracy. Deletion mutant isoform 394 bp was found in a patient in clinical remission after treatment of AML, M3 subtype. The isoform consists of exon 1, part of exon 2, the 5′ end of exon 4 and exons 5, 6, and 7 (Fig. 3). The structure of that isoform is similar to Ik7 with exon 5 and one zinc finger in it. Its intensity was 24.1% of the Ikaros transcripts in the sample. The shortest isoform 151 bp was found in two patients with common ALL. It was similar to Ik6, missing exons 3, 4, 5, and 6, but with a 25 bp deletion in the 3′-end of exon 2 and a 37 bp deletion in the 5′end of exon 7 (Fig. 4). This truncated isoform was expressed in both patients at levels 10–14% of the total of Ikaros transcripts in the sample. We were able to detect a PCR product of that size in some other patients at a very low level, but this detection was not reproducible. Ikaros expression profile in patient's and healthy donor's groups To estimate the level of functional Ikaros expression and compare groups of patients we calculated the Ikaros ratio between a functional isoform (Ik1 + Ikx + Ik2) intensity and the sum of all isoform intensities, as it was proposed by Olivero et al. [18]. In a general group of patients and healthy controls the ratio ranged from 0 (only short isoforms are detected) to 1 (only long isoforms are detected). We compared the
Table 1 Incidence of predominant Ikaros isoforms expressed in leukemia patients and healthy donors
B-ALL T-ALL AML Healthy donors
Ik 1
Ik x
Ik 2/3
Ik4
Ik 5/7
Ik4A
Ik 8
Ik 6
20.6% 6% 0% 11%
8.8% 0% 30% 11%
63.2% 75% 70% 77%
36.8% 50% 40% 33.3%
2.9% 0% 0% 0%
22.1% 25% 50% 22.2%
26.5% 25% 10% 11%
11.8% 6% 10% 0%
Fig. 3. Fragment of sequence and schematic structure of mutant Ikaros isoform. Underlined sequence corresponds to forward PCR primer.
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
281
Fig. 4. Fragment of sequence and schematic structure of mutant short Ikaros isoform.
Ikaros ratio in healthy donors and groups of patients with B-lineage ALL and different fusion oncogene expression, T-ALL and AML (Fig. 5). The highest ratio (0.48 to 1, median 0.61) was determined for healthy donors with a prevalence of the long isoforms; B-lineage ALL, T-ALL and AML show a tendency of gradual ratio reduction, a significant difference for AML. In B-lineage ALL the Ikaros ratio varies in a wide range (0 to 1, median 0.51). In T-ALL the Ikaros ratio was slightly reduced to the short isoforms expression (0 to 0.84, median 0.52) and the lowest ratio was determined in AML patients (0.03 to 0.75, median 0.34) that is significantly less than in the control group (p = 0.027, Mann–Whitney U Test). In B-precursor ALL patients with fusion oncogenes TEL/AML1 (N = 14) and E2A/PBX (N =12) the Ikaros ratio was reduced, the most evident reduction was found in patients with BCR/ABL (N =4) compared to healthy control (p = 0.0028, Mann– Whitney U Test). Ikaros expression profile according to immunophenotype of ALL We analyzed the Ikaros ratio in immunophenotypic subgroups of B-lineage ALL. The ratio showed a downtrend with the maturation of B-cell leukemia. Median of the Ikaros ratio was 0.72 in pro-B ALL (N = 5), 0.51 in common ALL (N = 44) and 0.24 in pre-B ALL (N = 16), but
differences between groups do not reach significance. This reduction in Ikaros profile during maturation stages is due to a decrease in Ik2 and an increase in Ik8 expression (data nor shown). Other isoforms do not change significantly. The Ikaros isoform profile was analyzed according to coexpression of myeloid marker for B-lineage leukemic cells. Negative association was found between the Ikaros ratio and myeloid immunophenotypic markers CD13, CD33, CD15, and MPO by both percent of positive cells and mean intensity of fluorescence (MFI). The correlation was highly significant for CD15 (Fig. 6). The same correlation was shown also for Ik1 and Ik2. On the contrary, Ik DN isoforms tend to positive correlation with myelocoexpression, in part with Ik8 and CD33 expression (R = 0.53; p = 0.03). Negative correlation was found between the Ikaros ratio and percent of Tdt positive cells (p = 0.045, Spearman correlations). The Ikaros ratio is significantly higher in Tdt negative B-cell leukemias (median 0.75) than in Tdt positive cases (median 0.47) (p = 0.01, Mann– Whitney U Test). Analysis of the associations of Ikaros profile and immunophenotype of T-ALL, despite a low case number (14 T-ALL patients), revealed that a correlation of the Ikaros ratio with CD5 and CD7 expression was found. CD5 shows a positive correlation with the Ikaros ratio by both percent of positive cells (R = 0.5, p = 0.005) and mean fluorescence intensity (R = 0.53, p = 0.047). CD7 negatively correlates with the Ikaros ratio, significantly by MFI (R = −0.65, p = 0.01). Association between Ikaros and MPO expression was noted. A reduced level of MPO coexpression correlates with a high Ikaros ratio in T-ALL cells, significantly for MPO MFI (p = 0.037). Taking into account a low case number (4 MPO positive T-ALL) this association requires conformation. No other correlation of Ikaros profile and myelo- and B-cell markers was found. Discussion
Fig. 5. Ratio of isoforms (Ik1 + x + 2)/(Ik4 + 5/7 + 8 + 4A + 6) in groups of samples. The first box indicates healthy bone marrow donors. FG — means patients with B-lineage ALL without fusion oncogene; TEL/AML1, E2A/PBX, BCR/ABL — B-lineage ALL patients with respective fusion oncogene. Statistic significance is indicated in comparison to donors control group.
The Ikaros transcription factor has been shown to play an essential role in lymphoid and myeloid differentiation and is supposed to serve as a tumor suppressing gene for these cell lineages [3,23]. Alternative splicing of Ikaros gene results in a number of mRNA and protein isoforms with distinct activity and capability of DNA binding. In this study, we use RT-PCR analysis and polyacrylamide gel electrophoresis to estimate profile of Ikaros isoforms expression in childhood ALL in comparison with AML and healthy bone marrow samples. Several previous researches perform detection of long and short Ikaros isoforms in leukemia samples [9,11–17]. Real-time PCR assay is a suitable method for quantitative analysis of Ikaros gene and isoform expression, and Olivero et al. have been followed this way [18]. Using nested RT-PCR and analysis of band intensity in a gel we lost the advantage of quantitative real-time assay. Nevertheless, we chose semi-quantitative analysis for the following reasons. First, we did not intend to measure the levels of Ikaros expression itself, but a comparative proportion of different Ikaros isoforms in leukemia and control samples. Second, we took advantage of analysis with one primer set for each sample instead of multiple primer and probe sets. Third, we were able to detect all possible splicing and mutant isoform
282
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
Fig. 6. Negative correlation between the Ikaros ratio and CD15 expression on leukemia cells.
in a gel line, distinguish them by size and cut bands of interests for sequencing without special purification. We detected 8 different splicing isoforms totally, described by the length and number of zinc-finger domains in DNA-binding domain into two groups: functional Ikaros isoforms with at least 3 zinc fingers (Ik1, Ikx, Ik2) and short, dominant-negative isoforms (Ik4A, Ik5/7, Ik4A, Ik8, Ik6). Frequent modifications of Ikaros transcripts, 30-base deletions and 60-base insertions, are described to appear in normal human hemopoietic cells as well as in leukemic cells [20], so we counted them as respective isoforms without deletion or insertion. All isoforms were found in all patient and control groups, but the relative level of their incidence and intensity was appreciably different. We did not found any new splicing isoforms of Ikaros. Ikx was described as a novel isoform before the beginning of the current study by K.J. Payne et al. [20], and named as Ik3A by A. Rebollo [3]. It was previously shown, that Ikx is selectively expressed in normal myeloid lineage, while Ik1 predominates in erythroid and most lymphoid lineage cells [21]. This tendency corresponded to our results on leukemia samples. Ikx was found as the predominant isoform in 30% of the AML samples, while no AML cases were positive for Ik1. On the contrary a slight band of Ikx was detected in ALL only in a few samples. A pattern of Ikaros expression in AML is under our further investigation, but we can conclude that a high relation of Ik1 to Ikx in lymphoid vs. myeloid lineage differentiation concerns lymphoid and myeloid leukemia. Further studies are required to determine the role of Ikx isoform in hematopoiesis and leukemogenesis. Some tendencies were found for reduction of the Ikaros ratio during differentiation of leukemia through pro-B, common and pre-B stages. This observation may reflect some changes of Ikaros function in the process of normal B-lymphocyte differentiation. We did not find a restriction of Ik6 to a subset of common ALL, as it was published previously [15]. Association of short DN Ikaros isoforms with coexpression of myeloid markers proves the idea that functional Ikaros is also playing a role in keeping lymphoid identity. Interestingly, among all myeloid markers the clearest correlation for all isoforms appeared with CD15. CD15 is a carbohydrate adhesion molecule (not a protein) bound to glycolipid or glycoprotein, so Ikaros cannot regulate expression of CD15 directly as a transcription factor. Regulation of CD15 expression by Ikaros is possible through a more complex mechanism: repression of fucosyltransferase 4 gene, which codes an enzyme that synthesizes CD15. Association of Ikaros function with Tdt expression is certain because Ikaros was discovered for the first time as a protein that could
regulate Tdt and CD3 expression. Of special interest there is a positive correlation between Ikaros ratio and CD5 expression and a negative correlation between Ikaros and CD7. CD5 is a protein receptor for CD72, expressed on the surface of B1 and T cell. CD5 serves to mitigate activating signals from the BCR and TCR to keep activation and proliferation of T cell under control. It is known, that Ikaros and its homolog Aiolos set a threshold for activation of T and B-cell upon BCR or TCR engagement. Ikaros mutant T-cells show a hyper-responsive phenotype and are prone to transformation [7]. CD7, an early T-cell marker, is a receptor that activates PI3K, increases signaling from TCR and IL2R, and contributes to apoptosis induction. We can propose that the suppressive effect of Ikaros partially relies on up-regulation of CD5 and down-regulation of CD7. Loss of Ikaros function by expression of short DN Ikaros isoforms leads to hyper-proliferation, resistance to apoptosis and probably development of leukemia. The results we obtained are in good correspondence with previous issues showing that expression of short non-DNA-binding Ikaros isoforms is associated with leukemias. In this research, we aimed to perform semi-quantitative analysis of a comparative proportion of different Ikaros isoforms in leukemia. To our knowledge, this is the first report to document the relative level of Ikaros isoform expression in childhood ALL in comparison to normal bone marrow. Our data confirm that all detected Ikaros isoforms can be expressed in both B- and T-lineage ALL as well as in AML and normal bone marrow cells, but the relative level of their expression differs essentially. In rare instances, so-called ‘mutant’ deletion and insertion Ikaros isoforms were detected in this and previous studies [12,13]. On the other hand, a number of studies have demonstrated that in most cases genomic analysis of leukemic cells did not identify any mutation at the DNA level responsible for the high rate of short isoforms expression [3,11,24]. Thus the alternative splicing regulation is crucial in Ikaros transcription alterations in leukemic cells. In the absence of genomic mutations, in-frame deletions might come from alternative splice sites in an exon sequence. It has not yet proved for certain if dominant-negative Ikaros overexpression occurs as an initial event in leukemiogenesis, or if it occurs in the process of transformation caused by some other genetic aberrations. Some evidence supports the latter idea, such as induction of aberrant Ikaros splicing by BCR-ABL1 [25] and MLLAF4 [16] fusion oncogenes. Further investigations of mechanisms of spicing regulation in normal cell development and cancerogenesis are supposed to clarify the role of Ikaros and other Ikaros family members in leukemia.
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
Acknowledgments The authors thank V. Fedosenko for help in image preparation and Dr. Uladzislau Ivaniutin for reviewing the manuscript. This project was funded by the grant #B06M-107 of the Foundation for Fundamental Research of the Republic of Belarus. References [1] M. Cortes, E. Wong, J. Koipally, K. Georgopoulos, Control of lymphocyte development by the Ikaros gene family, Curr. Opin. Immunol 11 (2) (1999 Apr) 167–171. [2] B.S. Cobb, S.T. Smale, Ikaros-family proteins: in search of molecular functions during lymphocyte development, Curr Top Microbiol Immunol 290 (2005) 29–47. [3] A. Rebollo, C. Schmitt, Ikaros, Aiolos and Helios: transcription regulators and lymphoid malignancies, Immunol Cell Biol 81 (3) (2003 Jun) 171–175. [4] J.H. Wang, A. Nichogiannopoulou, L. Wu, L. Sun, A.H. Sharpe, M. Bigby, K. Georgopoulos, Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation, Immunity 5 (6) (1996 Dec) 537–549. [5] K. Georgopoulos, M. Bigby, J.H. Wang, A. Molnar, P. Wu, S. Winandy, A. Sharpe, The Ikaros gene is required for the development of all lymphoid lineages, Cell 79 (1) (1994 Oct 7) 143–156. [6] S. Winandy, P. Wu, K. Georgopoulos, A dominant mutation in the Ikaros gene leads to rapid development of leukemia and lymphoma, Cell 83 (2) (1995 Oct 20) 289–299. [7] N. Avitahl, S. Winandy, C. Friedrich, B. Jones, Y. Ge, K. Georgopoulos, Ikaros sets thresholds for T cell activation and regulates chromosome propagation, Immunity 10 (3) (1999 Mar) 333–343. [8] D. Pulte, R.A. Lopez, S.T. Baker, M. Ward, E. Ritchie, C.A. Richardson, D.W. O'Neill, A. Bank, Ikaros increases normal apoptosis in adult erythroid cells, Am J Hematol 81 (1) (2006 Jan) 12–18. [9] T. Yagi, S. Hibi, M. Takanashi, et al., High frequency of Ikaros isoform 6 expression in acute myelomonocytic and monocytic leukemias: implications for upregulation of the antiapoptotic protein Bcl-XL in leukemogenesis, Blood 99 (4) (2002 Feb 15) 1350–1355. [10] K.L. Kathrein, R. Lorenz, A.M. Innes, E. Griffiths, S. Winandy, Ikaros induces quiescence and T-cell differentiation in a leukemia cell line, Mol. Cell Biol 25 (5) (2005 Mar) 1645–1654. [11] L. Sun, N. Heerema, L. Crotty, Expression of dominant-negative and mutant isoforms of the antileukemic transcription factor Ikaros in infant acute lymphoblastic leukemia, Proc. Natl. Acad. Sci 96 (1999) 680–685.
283
[12] L. Sun, P.A. Goodman, C.M. Wood, et al., Expression of aberrantly spliced oncogenic ikaros isoforms in childhood acute lymphoblastic leukemia, J Clin Oncol 17 (12) (1999 Dec) 3753–3766. [13] L. Sun, M.L. Crotty, M. Sensel, H. Sather, C. Navara, J. Nachman, P.G. Steinherz, P.S. Gaynon, N. Seibel, C. Mao, A. Vassilev, G.H. Reaman, F.M. Uckun, Expression of dominant-negative Ikaros isoforms in T-cell acute lymphoblastic leukemia, Clin Cancer Res 5 (8) (1999 Aug) 2112–2120. [14] M. Takanashi, T. Yagi, T. Imamura, Y. Tabata, A. Morimoto, S. Hibi, E. Ishii, S. Imashuku, Expression of the Ikaros gene family in childhood acute lymphoblastic leukaemia, Br J Haematol 117 (3) (2002 Jun) 525–530. [15] C. Tonnelle, M.C. Imbert, D. Sainty, S. Granjeaud, C. N'Guyen, C. Chabannon, Overexpression of dominant-negative Ikaros 6 protein is restricted to a subset of B common adult acute lymphoblastic leukemias that express high levels of the CD34 antigen, Hematol. J 4 (2) (2003) 104–109. [16] A. Ruiz, J. Jiang, H. Kempski, H.J. Brady, Overexpression of the Ikaros 6 isoform is restricted to t(4;11) acute lymphoblastic leukaemia in children and infants and has a role in B-cell survival, Br. J. Haematol 125 (1) (2004 Apr) 31–37. [17] C. Tonnelle, B. Calmels, C. Maroc, J. Gabert, C. Chabannon, Ikaros gene expression and leukemia, Leuk. Lymphoma 43 (1) (2002 Jan) 29–35. [18] S. Olivero, C. Maroc, E. Beillard, J. Gabert, W. Nietfeld, C. Chabannon, C. Tonnelle, Detection of different Ikaros isoforms in human leukaemias using real-time quantitative polymerase chain reaction, Br. J. Haematol 110 (4) (2000 Sep) 826–830. [19] J. Gabert, E. Beillard, V.H. van der Velden, et al., Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia — a Europe Against Cancer program, Leukemia 17 (12) (2003 Dec) 2318–2357. [20] K.J. Payne, J.H. Nicolas, J.Y. Zhu, L.W. Barsky, G.M. Crooks, Cutting edge: predominant expression of a novel Ikaros isoform in normal human hemopoiesis, J. Immunol 167 (4) (2001 Aug) 1867–1870. [21] K.J. Payne, G. Huang, E. Sahakian, J.Y. Zhu, N.S. Barteneva, L.W. Barsky, M.A. Payne, G.M. Crooks, Ikaros isoform x is selectively expressed in myeloid differentiation, J. Immunol 170 (6) (2003 Mar 15) 3091–3098. [22] T. Ronni, K.J. Payne, S. Ho, M.N. Bradley, G. Dorsam, S. Dovat, Human Ikaros function in activated T cells is regulated by coordinated expression of its largest isoforms, J. Biol. Chem 282 (4) (2007 Jan 26) 2538–2547. [23] C. Schmitt, C. Tonnelle, A. Dalloul, C. Chabannon, P. Debré, A. Rebollo, Aiolos and Ikaros: regulators of lymphocyte development, homeostasis and lymphoproliferation, Apoptosis 7 (3) (2002 Jun) 277–284. [24] C.G. Mullighan, S. Goorha, I. Radtke, et al., Genome-wide analysis of genetic alterations in acute lymphoblastic leukaemia, Nature 446 (7137) (2007 Apr 12) 758–764. [25] F. Klein, N. Feldhahn, S. Herzog, M. Sprangers, J.L. Mooster, H. Jumaa, M. Müschen, BCR-ABL1 induces aberrant splicing of IKAROS and lineage infidelity in pre-B lymphoblastic leukemia cells, Oncogene 25 (7) (2006 Feb 16) 1118–1124.
Contents lists available at ScienceDirect
Blood Cells, Molecules, and Diseases j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y b c m d
Relative expression of different Ikaros isoforms in childhood acute leukemia Alexander N. Meleshko ⁎, Ludmila V. Movchan, Michael V. Belevtsev, Tatjana V. Savitskaja Belarusian Research Center for Pediatric Oncology and Hematology, Minsk, Belarus
a r t i c l e
i n f o
Article history: Submitted 9 June 2008 (Communicated by M. Lichtman, M.D., 24 June 2008) Keywords: Ikaros Dominant-negative Isoform Acute lymphoblastic leukemia
a b s t r a c t Ikaros is a zinc-finger transcriptional factor playing an essential role in lymphoid lineage commitment and differentiation. Animal models and analysis of human Ikaros in leukemic cells demonstrate deregulation of Ikaros expression. Short isoforms with a truncated DNA-binding domain suppress functions of Ikaros in a dominant-negative manner. Previous studies demonstrated that human leukemias are heterogeneous for Ikaros expression. We estimate the relative level of Ikaros mRNA transcripts in 80 childhood ALL cases in comparison with AML and healthy donor groups. We detected eight major isoforms and several minor mutant isoforms in most patients with acute lymphoblastic and myeloid leukemia and in healthy donors, but the relative level of expression varied. The relatively high level of Ik4A isoform, rarely mentioned in previous reports, was detected in all analyzed groups. The ratio between functional and all isoforms was used to determine functional activity of Ikaros. The ratio was significantly less in AML (p = 0.027) and BCR-ABL positive ALL (p = 0.0028) than in healthy bone marrow. We found a negative association between the Ikaros ratio and myeloid coexpression in B-cell ALL, the most prominent was for CD15. The Ikaros ratio positively correlates with CD5 and negatively with CD7 expression in T-ALL. We suggest that an anti-proliferation and anti-activation effect of full-length Ikaros may be mediated through regulation of CD5 and CD7. © 2008 Elsevier Inc. All rights reserved.
Introduction Ikaros is a transcription factor, which plays an important role in controlling hematopoietic, particularly lymphoid cell differentiation, proliferation and function [1,2]. The Ikaros protein, like other proteins of the Ikaros family, contains two separate regions of zinc-finger domains: 4 DNA-binding zinc fingers near the N-terminus and 2 zinc fingers for protein–protein interactions near the C-terminus. The Ikaros gene (ZNFN1A1) contains 7 exons and is transcribed as a number of isoforms due to alternative splicing with alternative use of exons 2–6. At least 11 isoforms of human Ikaros has been described [3] (Fig. 1). Long isoforms (Ik1 to 3) have at least three zinc fingers which are able to bind DNA and considered to be functional. Short isoforms (Ik4 to 8) lack two or more zinc-finger domains, so they cannot bind DNA and impair the function of Ikaros proteins in a dominant-negative (DN) manner. The essential role of Ikaros in the regulation of development from stem cell to mature lymphocytes was established by knock-out experiments in mice. Mutant Ikaros −/− mice have severe lymphoid cell defects [4,5] but heterozygous Ikaros DN +/− mice invariably developed T-cell malignances [6]. Before the onset of transformation mutant mice had a hyper-responsive phenotype to IL-2R and TCR stimulation, and the degree of proliferation correlates with the level of
⁎ Corresponding author. E-mail address: [email protected] (A.N. Meleshko). 1079-9796/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.bcmd.2008.06.006
Ikaros activity [7]. Ikaros was also shown to be involved in apoptosis and cell cycle regulation in lymphocytes [8–10]. These evidence support the idea that Ikaros may function as a hematological tumor suppressor. Analysis of Ikaros isoform expression was performed in several clinical studies in leukemia patients. Overexpression of short DN Ikaros isoforms (Ik4, 7, 8 and/or 6) were detected in infant ALL, childhood, adult B and T-lineage ALL [11–16]. Data about Ikaros expression profile in AML are less frequent. Yagi et al. has detected Ik6 in 7 of 10 cases of M4 and M5 in a group of 24 different AML [9]. In another study, Ik1, 2 and short isoforms were detected in 3 cases (two M2 and one M4) out of 7 AML analyzed [17]. All these and some other studies use reverse-transcription PCR and qualitative analysis in agarose gel or immunobloting to discriminate different isoforms of Ikaros on RNA or protein level. The only quantitative analysis using real-time PCR was performed by Olivero et al. [18]. The authors used two sets of primers and a probe to detect separately Ik1 + 2 and Ik1 + 2 + 4 + 7 + 8 so that they can calculate the ratio between large and short isoforms. Unfortunately, this elegant method does not allow quantifying each isoform and loses sight of Ikx and Ik6, which are not mentioned in that issue. In this study we use RT-PCR and subsequent analysis by polyacrylamide gel electrophoresis to separate and characterize expression of each Ikaros isoform in childhood ALL in comparison with some AML samples and healthy control. We also performed extensive association between Ikaros expression profile and immunophenotype and fusion oncogene expression.
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
279
Materials Patients, samples and cell lines A total of 80 children with ALL were evaluated in the study. 66 patients had B-precursor ALL (60 at diagnosis, 6 at relapse), and 14 patients had T-ALL (13 at diagnosis, 1 at relapse). Diagnosis was based on morphology, cytogenetics, immunophenotype and molecular biology analyses. Age distribution was from 1 month to 18.5 years (median 6.6 years). Leukemic blast count in bone marrow ranged from 53.5% to 99%, median 93%. Nine patients with AML were analyzed. Control groups include 9 healthy donors: 5 bone marrow donors, 1 blood donor, 1 bone marrow iron-deficiency anemia patient and two bone marrow samples from leukemic patients in stable remission. Five cell lines were analyzed for Ikaros isoform expression: B-lineage, IM-9, Raji, T-lineage, Molt-3, Jurkat and myeloid, К562. RNA extraction and cDNA preparation Mononuclear cells were isolated from 2–5 ml of BM sample by Ficoll–Paque centrifugation. Total RNA from fresh cells was extracted using Gen Elute Mammalian Total RNA Miniprep Kit (Sigma-Aldrich, St Louis, MO, USA). Former patient samples were lyzed in TRI Reagent® RNA Isolation Reagent (Sigma) and stored at minus 80 °C before the extraction. Quantity and quality of obtained total RNA were defined by spectrophotometry using Gene Quant RNA/DNA Calculator (Amersham Biosciences/GE Healthcare, UK) and gel electrophoresis. 1 µg of RNA was taken for cDNA synthesis in a 20 µl reaction mix with 5 mM MgCl2, 1 mM dNTP, 25 mM random hexamers (GE Healthcare, UK), 20 U RNAsin (Promega), and 200 U Reverta (Promega) according to standard protocol [19]. RT-PCR analysis and quantification of Ikaros isoforms PCR amplification for Ikaros was first performed with the primer pair of F1 ATGGATGCTGACGAGGGTCAAGAC and R1 TTAGCTCATGTGGAAGCGGTGCTC under the following conditions: 35 cycles of 30 s at 94 °C, 30 s at 62 °C and 1 min at 72 °C, with a final extension step of 7 min at 72 °C. 2 µg of the PCR products from the first step were amplified further by nested PCR, with primers of F2 CCCCTGTAAGCGATACTCCAGATG and R2 GATGGCTTGGTCCATCACGTGGGA. The second PCR step was performed under the following conditions: 40 cycles of 30 s at 94 °C, 90 s at 68 °C, and 2 min at 72 °C, with a final extension
Fig. 2. Polyacrylamide gel electrophoresis analysis of Ikaros gene showing different isoform expression in leukemic samples and healthy controls. MNC — mononuclear cells from normal bone marrow. Isoforms with a 30 bp deletion and a 60 bp insertion are indicated. Mutant isoform denoted with asterisk is represented in details in Fig. 3. In the left part of the figure the profile of the first line, plotted by KODAK 1D v3.6.1, is shown.
step of 7 min at 72 °C. The primer sequences were selected in a previous study [14]. PCR products were primarily examined by 1.5% agarose gel electrophoresis. To identify different Ikaros isoforms we ran PCR products in 8% non-denaturating polyacrylamide gel using Hoefer vertical electrophoresis system (Amersham Biosciences/GE Healthcare, UK) with maximum gel length (24 sm) for better bands discrimination. The gel was stained with ethidium bromide and fixed to a graphic file by Gel Doc 2000 (Bio-Rad, USA). Analysis of the image was performed with KODAK 1D v3.6.1 Scientific Imaging system. This software automatically or manually detects lines and bands, and constructs a graph profile for a line. After subtracting the background, several parameters for each exact band were evaluated, including molecular weight relative to DNA ladder standard and relative intensity — the percent intensity of a band within a specified range of a line. In that way we were able to define the relative level of all Ikaros isoforms. An example of a gel was shown in Fig. 2. Sequence analysis Individual bands of each size for some patients, as well as all unusual bands were cut from the gel and subjected to PCR using the second step primers. Specific PCR products of the right size were purified by polyacrylamide gel electrophoresis and sequenced. Sequencing was performed with BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, USA) on ABI PRISM 3130 automatic genetic analyzer (Applied Biosystems, USA). The obtained sequence was aligned with germline sequences of Ikaros gene exons by BioEdit and MEGA-3 software to identify an isoform. Immunophenotypic analysis
Fig. 1. Schematic representation of human Ikaros isoforms. Zinc fingers are represented as black vertical bars.
Freshly obtained bone marrow samples were subjected to detailed three-color immunophenotyping analysis. The following markers were analyzed: CD45, CD1a, CD2, CD3, CD4, CD8, CD5, CD7, CD10, CD19, CD20, CD22, CD79, CyIgM, sIgM, CD13, CD33, CD15, CD11c,
280
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
MPO, CD34, CD117, TdT, and HLA-DR with a panel of monoclonal antibodies (Becton Dickinson, USA) using FACScan flow cytometer (Becton Dickinson, USA). Cases expressing a given marker on more than 20% of blast cells were considered positive. The mean fluorescent intensity (MFI) for all markers was determined to evaluate the level of marker expression. Results Incidence and relative level of Ikaros isoform expression We performed detailed analysis of different Ikaros isoform expression by nested RT-PCR and polyacrylamide gel electrophoresis. The following PCR product types of isoforms were usually distinguished: Ik1 (893 bp), Ikx (758 bp), Ik2/3 (632 bp), Ik4 (505 bp), Ik5/7 (464–463 bp), Ik4A (370 bp), Ik8 (337 bp) and Ik6 (202 bp) with the primers used in this study. Sequencing analysis, performed for several patients and all band types, confirmed correct isoform definition. In all cases we sequenced a band of 632 bp that was Ik2 but not Ik3. Analysis of a sequence of 464 bp in some cases revealed Ik5, in the other — Ik7. We detect also a 60 bp insertion in exon 2 and a 30 bp deletion in exon 6 for all isoforms, most often for Ik2, Ik8 and Ik6. Mutant isoforms were detected by an incomplete sequence of one of the exons. For statistical analysis we refereed all isoforms to the types listed above on the basis of exon order and the number of zinc fingers. Expression of different Ikaros isoforms was compared between four groups: B-lineage ALL, T-lineage ALL, AML and healthy donors. The presence of predominant isoforms accounted if the band relative intensity was more than 20% of the whole electrophoresis line density. The 20% threshold was chosen, because slack bands with less than 20% intensity was not reliably reproducible in repeated experiments. Incidence of predominant isoforms are presented in the Table 1. We also estimated an average level of each isoform expression for all four groups (Table 2). All Ikaros isoforms were expressed in most patients with acute lymphoblastic and myeloid leukemia and healthy donors, but the relative level of expression varied. The most predominant isoform was Ik2, then Ik4 and Ik1. Some differences are found in isoform distribution between lineages of leukemia and health controls. In normal bone marrow and blood expression of Ik1, Ikx and Ik2 are relatively higher, and there is only a weak or absent expression of Ik6 and Ik5/6 compared to leukemic cells. Ikx is a novel isoform, found and denoted this way by K.J. Payne et al. [20], but in uniform classification Ikx is referred to as Ik3A [19]. This isoform was first described by K.J. Payne et al. as typical for normal hemopoiesis [20] and later they as well show predominant expression in myeloid lineage [21] what corresponds with our data: Ikx found at a higher level in acute myeloid leukemias and healthy controls. In these groups Ikx is more expressed than Ik1, while for ALL this feature is reversed. Surprisingly, we found a widespread isoform of 370 bp lacking exons 3, 5 and 6. The isoform appeared to be that which was described previously [3,20] as Ik4A, so we used this name. In most of the previous studies it was not detected, maybe because the PCR product (protein) of Ik4A is close by length to Ik8 and they are not distinguishable in agarose gel. Ik6, the shortest isoform, was found in 7/68 B-lineage ALL samples, and in these patients only Ik6 was detected as predominant, together with Ik8. In one patient with T-ALL
Table 2 Average level of different Ikaros isoforms expression in leukemia patients and healthy donors
B-ALL T-ALL AML Healthy donors
Ik 1
Ik x
Ik 2/3
Ik4
Ik 5/7
Ik4A
Ik 8
Ik 6
11.1% 7.1% 3.7% 10.3%
6.5% 5.7% 10.8% 18.6%
27.4% 31.2% 26.3% 38.5%
16.3% 17.9% 19.9% 15%
2.3% 2.3% 0.2% 1.3%
14.6% 15.3% 23.4% 5.4%
11.2% 11.5% 10.7% 7.5%
10.8% 9.1% 5.1% 3.5%
only Ik6 expression was found, and one of the AML patients showed a predominantly Ik6 and Ik4A expression. The rarest isoform was Ik5/7 — its expression was not detected at all or at a low level; only in one patient with B-precursor ALL it was 21.4% of the Ikaros transcripts. We never detected the Ik2A described by A. Rebollo [3] in that study. Mutant or unusual Ikaros isoforms Sequencing analysis of PCR products showed modifications of splicing variants with a 30 bp deletion at the 3′ end of exon 6 and a 60 bp insertion at the 3′ end of exon 2, as described previously [11,12,20]. There is evidence that such kind of Ikaros transcript modifications play a role in normal hemopoiesis and T-cell regulation [20,22]. We found the insertion and deletion variants of all isoforms Ik1–Ik6. Well-defined bands were observed for deletion and insertion variants of Ik6, Ik2, and deletion variants of Ik8, Ikx and Ik1. In some cases a prominent band of Ik8del was observed in complete absence of Ik8 without deletion. In all cases these isoforms counted as respective isoforms without insertions and deletions. We revealed some unusual isoforms, resulting most likely from splicing inaccuracy. Deletion mutant isoform 394 bp was found in a patient in clinical remission after treatment of AML, M3 subtype. The isoform consists of exon 1, part of exon 2, the 5′ end of exon 4 and exons 5, 6, and 7 (Fig. 3). The structure of that isoform is similar to Ik7 with exon 5 and one zinc finger in it. Its intensity was 24.1% of the Ikaros transcripts in the sample. The shortest isoform 151 bp was found in two patients with common ALL. It was similar to Ik6, missing exons 3, 4, 5, and 6, but with a 25 bp deletion in the 3′-end of exon 2 and a 37 bp deletion in the 5′end of exon 7 (Fig. 4). This truncated isoform was expressed in both patients at levels 10–14% of the total of Ikaros transcripts in the sample. We were able to detect a PCR product of that size in some other patients at a very low level, but this detection was not reproducible. Ikaros expression profile in patient's and healthy donor's groups To estimate the level of functional Ikaros expression and compare groups of patients we calculated the Ikaros ratio between a functional isoform (Ik1 + Ikx + Ik2) intensity and the sum of all isoform intensities, as it was proposed by Olivero et al. [18]. In a general group of patients and healthy controls the ratio ranged from 0 (only short isoforms are detected) to 1 (only long isoforms are detected). We compared the
Table 1 Incidence of predominant Ikaros isoforms expressed in leukemia patients and healthy donors
B-ALL T-ALL AML Healthy donors
Ik 1
Ik x
Ik 2/3
Ik4
Ik 5/7
Ik4A
Ik 8
Ik 6
20.6% 6% 0% 11%
8.8% 0% 30% 11%
63.2% 75% 70% 77%
36.8% 50% 40% 33.3%
2.9% 0% 0% 0%
22.1% 25% 50% 22.2%
26.5% 25% 10% 11%
11.8% 6% 10% 0%
Fig. 3. Fragment of sequence and schematic structure of mutant Ikaros isoform. Underlined sequence corresponds to forward PCR primer.
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
281
Fig. 4. Fragment of sequence and schematic structure of mutant short Ikaros isoform.
Ikaros ratio in healthy donors and groups of patients with B-lineage ALL and different fusion oncogene expression, T-ALL and AML (Fig. 5). The highest ratio (0.48 to 1, median 0.61) was determined for healthy donors with a prevalence of the long isoforms; B-lineage ALL, T-ALL and AML show a tendency of gradual ratio reduction, a significant difference for AML. In B-lineage ALL the Ikaros ratio varies in a wide range (0 to 1, median 0.51). In T-ALL the Ikaros ratio was slightly reduced to the short isoforms expression (0 to 0.84, median 0.52) and the lowest ratio was determined in AML patients (0.03 to 0.75, median 0.34) that is significantly less than in the control group (p = 0.027, Mann–Whitney U Test). In B-precursor ALL patients with fusion oncogenes TEL/AML1 (N = 14) and E2A/PBX (N =12) the Ikaros ratio was reduced, the most evident reduction was found in patients with BCR/ABL (N =4) compared to healthy control (p = 0.0028, Mann– Whitney U Test). Ikaros expression profile according to immunophenotype of ALL We analyzed the Ikaros ratio in immunophenotypic subgroups of B-lineage ALL. The ratio showed a downtrend with the maturation of B-cell leukemia. Median of the Ikaros ratio was 0.72 in pro-B ALL (N = 5), 0.51 in common ALL (N = 44) and 0.24 in pre-B ALL (N = 16), but
differences between groups do not reach significance. This reduction in Ikaros profile during maturation stages is due to a decrease in Ik2 and an increase in Ik8 expression (data nor shown). Other isoforms do not change significantly. The Ikaros isoform profile was analyzed according to coexpression of myeloid marker for B-lineage leukemic cells. Negative association was found between the Ikaros ratio and myeloid immunophenotypic markers CD13, CD33, CD15, and MPO by both percent of positive cells and mean intensity of fluorescence (MFI). The correlation was highly significant for CD15 (Fig. 6). The same correlation was shown also for Ik1 and Ik2. On the contrary, Ik DN isoforms tend to positive correlation with myelocoexpression, in part with Ik8 and CD33 expression (R = 0.53; p = 0.03). Negative correlation was found between the Ikaros ratio and percent of Tdt positive cells (p = 0.045, Spearman correlations). The Ikaros ratio is significantly higher in Tdt negative B-cell leukemias (median 0.75) than in Tdt positive cases (median 0.47) (p = 0.01, Mann– Whitney U Test). Analysis of the associations of Ikaros profile and immunophenotype of T-ALL, despite a low case number (14 T-ALL patients), revealed that a correlation of the Ikaros ratio with CD5 and CD7 expression was found. CD5 shows a positive correlation with the Ikaros ratio by both percent of positive cells (R = 0.5, p = 0.005) and mean fluorescence intensity (R = 0.53, p = 0.047). CD7 negatively correlates with the Ikaros ratio, significantly by MFI (R = −0.65, p = 0.01). Association between Ikaros and MPO expression was noted. A reduced level of MPO coexpression correlates with a high Ikaros ratio in T-ALL cells, significantly for MPO MFI (p = 0.037). Taking into account a low case number (4 MPO positive T-ALL) this association requires conformation. No other correlation of Ikaros profile and myelo- and B-cell markers was found. Discussion
Fig. 5. Ratio of isoforms (Ik1 + x + 2)/(Ik4 + 5/7 + 8 + 4A + 6) in groups of samples. The first box indicates healthy bone marrow donors. FG — means patients with B-lineage ALL without fusion oncogene; TEL/AML1, E2A/PBX, BCR/ABL — B-lineage ALL patients with respective fusion oncogene. Statistic significance is indicated in comparison to donors control group.
The Ikaros transcription factor has been shown to play an essential role in lymphoid and myeloid differentiation and is supposed to serve as a tumor suppressing gene for these cell lineages [3,23]. Alternative splicing of Ikaros gene results in a number of mRNA and protein isoforms with distinct activity and capability of DNA binding. In this study, we use RT-PCR analysis and polyacrylamide gel electrophoresis to estimate profile of Ikaros isoforms expression in childhood ALL in comparison with AML and healthy bone marrow samples. Several previous researches perform detection of long and short Ikaros isoforms in leukemia samples [9,11–17]. Real-time PCR assay is a suitable method for quantitative analysis of Ikaros gene and isoform expression, and Olivero et al. have been followed this way [18]. Using nested RT-PCR and analysis of band intensity in a gel we lost the advantage of quantitative real-time assay. Nevertheless, we chose semi-quantitative analysis for the following reasons. First, we did not intend to measure the levels of Ikaros expression itself, but a comparative proportion of different Ikaros isoforms in leukemia and control samples. Second, we took advantage of analysis with one primer set for each sample instead of multiple primer and probe sets. Third, we were able to detect all possible splicing and mutant isoform
282
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
Fig. 6. Negative correlation between the Ikaros ratio and CD15 expression on leukemia cells.
in a gel line, distinguish them by size and cut bands of interests for sequencing without special purification. We detected 8 different splicing isoforms totally, described by the length and number of zinc-finger domains in DNA-binding domain into two groups: functional Ikaros isoforms with at least 3 zinc fingers (Ik1, Ikx, Ik2) and short, dominant-negative isoforms (Ik4A, Ik5/7, Ik4A, Ik8, Ik6). Frequent modifications of Ikaros transcripts, 30-base deletions and 60-base insertions, are described to appear in normal human hemopoietic cells as well as in leukemic cells [20], so we counted them as respective isoforms without deletion or insertion. All isoforms were found in all patient and control groups, but the relative level of their incidence and intensity was appreciably different. We did not found any new splicing isoforms of Ikaros. Ikx was described as a novel isoform before the beginning of the current study by K.J. Payne et al. [20], and named as Ik3A by A. Rebollo [3]. It was previously shown, that Ikx is selectively expressed in normal myeloid lineage, while Ik1 predominates in erythroid and most lymphoid lineage cells [21]. This tendency corresponded to our results on leukemia samples. Ikx was found as the predominant isoform in 30% of the AML samples, while no AML cases were positive for Ik1. On the contrary a slight band of Ikx was detected in ALL only in a few samples. A pattern of Ikaros expression in AML is under our further investigation, but we can conclude that a high relation of Ik1 to Ikx in lymphoid vs. myeloid lineage differentiation concerns lymphoid and myeloid leukemia. Further studies are required to determine the role of Ikx isoform in hematopoiesis and leukemogenesis. Some tendencies were found for reduction of the Ikaros ratio during differentiation of leukemia through pro-B, common and pre-B stages. This observation may reflect some changes of Ikaros function in the process of normal B-lymphocyte differentiation. We did not find a restriction of Ik6 to a subset of common ALL, as it was published previously [15]. Association of short DN Ikaros isoforms with coexpression of myeloid markers proves the idea that functional Ikaros is also playing a role in keeping lymphoid identity. Interestingly, among all myeloid markers the clearest correlation for all isoforms appeared with CD15. CD15 is a carbohydrate adhesion molecule (not a protein) bound to glycolipid or glycoprotein, so Ikaros cannot regulate expression of CD15 directly as a transcription factor. Regulation of CD15 expression by Ikaros is possible through a more complex mechanism: repression of fucosyltransferase 4 gene, which codes an enzyme that synthesizes CD15. Association of Ikaros function with Tdt expression is certain because Ikaros was discovered for the first time as a protein that could
regulate Tdt and CD3 expression. Of special interest there is a positive correlation between Ikaros ratio and CD5 expression and a negative correlation between Ikaros and CD7. CD5 is a protein receptor for CD72, expressed on the surface of B1 and T cell. CD5 serves to mitigate activating signals from the BCR and TCR to keep activation and proliferation of T cell under control. It is known, that Ikaros and its homolog Aiolos set a threshold for activation of T and B-cell upon BCR or TCR engagement. Ikaros mutant T-cells show a hyper-responsive phenotype and are prone to transformation [7]. CD7, an early T-cell marker, is a receptor that activates PI3K, increases signaling from TCR and IL2R, and contributes to apoptosis induction. We can propose that the suppressive effect of Ikaros partially relies on up-regulation of CD5 and down-regulation of CD7. Loss of Ikaros function by expression of short DN Ikaros isoforms leads to hyper-proliferation, resistance to apoptosis and probably development of leukemia. The results we obtained are in good correspondence with previous issues showing that expression of short non-DNA-binding Ikaros isoforms is associated with leukemias. In this research, we aimed to perform semi-quantitative analysis of a comparative proportion of different Ikaros isoforms in leukemia. To our knowledge, this is the first report to document the relative level of Ikaros isoform expression in childhood ALL in comparison to normal bone marrow. Our data confirm that all detected Ikaros isoforms can be expressed in both B- and T-lineage ALL as well as in AML and normal bone marrow cells, but the relative level of their expression differs essentially. In rare instances, so-called ‘mutant’ deletion and insertion Ikaros isoforms were detected in this and previous studies [12,13]. On the other hand, a number of studies have demonstrated that in most cases genomic analysis of leukemic cells did not identify any mutation at the DNA level responsible for the high rate of short isoforms expression [3,11,24]. Thus the alternative splicing regulation is crucial in Ikaros transcription alterations in leukemic cells. In the absence of genomic mutations, in-frame deletions might come from alternative splice sites in an exon sequence. It has not yet proved for certain if dominant-negative Ikaros overexpression occurs as an initial event in leukemiogenesis, or if it occurs in the process of transformation caused by some other genetic aberrations. Some evidence supports the latter idea, such as induction of aberrant Ikaros splicing by BCR-ABL1 [25] and MLLAF4 [16] fusion oncogenes. Further investigations of mechanisms of spicing regulation in normal cell development and cancerogenesis are supposed to clarify the role of Ikaros and other Ikaros family members in leukemia.
A.N. Meleshko et al. / Blood Cells, Molecules, and Diseases 41 (2008) 278–283
Acknowledgments The authors thank V. Fedosenko for help in image preparation and Dr. Uladzislau Ivaniutin for reviewing the manuscript. This project was funded by the grant #B06M-107 of the Foundation for Fundamental Research of the Republic of Belarus. References [1] M. Cortes, E. Wong, J. Koipally, K. Georgopoulos, Control of lymphocyte development by the Ikaros gene family, Curr. Opin. Immunol 11 (2) (1999 Apr) 167–171. [2] B.S. Cobb, S.T. Smale, Ikaros-family proteins: in search of molecular functions during lymphocyte development, Curr Top Microbiol Immunol 290 (2005) 29–47. [3] A. Rebollo, C. Schmitt, Ikaros, Aiolos and Helios: transcription regulators and lymphoid malignancies, Immunol Cell Biol 81 (3) (2003 Jun) 171–175. [4] J.H. Wang, A. Nichogiannopoulou, L. Wu, L. Sun, A.H. Sharpe, M. Bigby, K. Georgopoulos, Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation, Immunity 5 (6) (1996 Dec) 537–549. [5] K. Georgopoulos, M. Bigby, J.H. Wang, A. Molnar, P. Wu, S. Winandy, A. Sharpe, The Ikaros gene is required for the development of all lymphoid lineages, Cell 79 (1) (1994 Oct 7) 143–156. [6] S. Winandy, P. Wu, K. Georgopoulos, A dominant mutation in the Ikaros gene leads to rapid development of leukemia and lymphoma, Cell 83 (2) (1995 Oct 20) 289–299. [7] N. Avitahl, S. Winandy, C. Friedrich, B. Jones, Y. Ge, K. Georgopoulos, Ikaros sets thresholds for T cell activation and regulates chromosome propagation, Immunity 10 (3) (1999 Mar) 333–343. [8] D. Pulte, R.A. Lopez, S.T. Baker, M. Ward, E. Ritchie, C.A. Richardson, D.W. O'Neill, A. Bank, Ikaros increases normal apoptosis in adult erythroid cells, Am J Hematol 81 (1) (2006 Jan) 12–18. [9] T. Yagi, S. Hibi, M. Takanashi, et al., High frequency of Ikaros isoform 6 expression in acute myelomonocytic and monocytic leukemias: implications for upregulation of the antiapoptotic protein Bcl-XL in leukemogenesis, Blood 99 (4) (2002 Feb 15) 1350–1355. [10] K.L. Kathrein, R. Lorenz, A.M. Innes, E. Griffiths, S. Winandy, Ikaros induces quiescence and T-cell differentiation in a leukemia cell line, Mol. Cell Biol 25 (5) (2005 Mar) 1645–1654. [11] L. Sun, N. Heerema, L. Crotty, Expression of dominant-negative and mutant isoforms of the antileukemic transcription factor Ikaros in infant acute lymphoblastic leukemia, Proc. Natl. Acad. Sci 96 (1999) 680–685.
283
[12] L. Sun, P.A. Goodman, C.M. Wood, et al., Expression of aberrantly spliced oncogenic ikaros isoforms in childhood acute lymphoblastic leukemia, J Clin Oncol 17 (12) (1999 Dec) 3753–3766. [13] L. Sun, M.L. Crotty, M. Sensel, H. Sather, C. Navara, J. Nachman, P.G. Steinherz, P.S. Gaynon, N. Seibel, C. Mao, A. Vassilev, G.H. Reaman, F.M. Uckun, Expression of dominant-negative Ikaros isoforms in T-cell acute lymphoblastic leukemia, Clin Cancer Res 5 (8) (1999 Aug) 2112–2120. [14] M. Takanashi, T. Yagi, T. Imamura, Y. Tabata, A. Morimoto, S. Hibi, E. Ishii, S. Imashuku, Expression of the Ikaros gene family in childhood acute lymphoblastic leukaemia, Br J Haematol 117 (3) (2002 Jun) 525–530. [15] C. Tonnelle, M.C. Imbert, D. Sainty, S. Granjeaud, C. N'Guyen, C. Chabannon, Overexpression of dominant-negative Ikaros 6 protein is restricted to a subset of B common adult acute lymphoblastic leukemias that express high levels of the CD34 antigen, Hematol. J 4 (2) (2003) 104–109. [16] A. Ruiz, J. Jiang, H. Kempski, H.J. Brady, Overexpression of the Ikaros 6 isoform is restricted to t(4;11) acute lymphoblastic leukaemia in children and infants and has a role in B-cell survival, Br. J. Haematol 125 (1) (2004 Apr) 31–37. [17] C. Tonnelle, B. Calmels, C. Maroc, J. Gabert, C. Chabannon, Ikaros gene expression and leukemia, Leuk. Lymphoma 43 (1) (2002 Jan) 29–35. [18] S. Olivero, C. Maroc, E. Beillard, J. Gabert, W. Nietfeld, C. Chabannon, C. Tonnelle, Detection of different Ikaros isoforms in human leukaemias using real-time quantitative polymerase chain reaction, Br. J. Haematol 110 (4) (2000 Sep) 826–830. [19] J. Gabert, E. Beillard, V.H. van der Velden, et al., Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia — a Europe Against Cancer program, Leukemia 17 (12) (2003 Dec) 2318–2357. [20] K.J. Payne, J.H. Nicolas, J.Y. Zhu, L.W. Barsky, G.M. Crooks, Cutting edge: predominant expression of a novel Ikaros isoform in normal human hemopoiesis, J. Immunol 167 (4) (2001 Aug) 1867–1870. [21] K.J. Payne, G. Huang, E. Sahakian, J.Y. Zhu, N.S. Barteneva, L.W. Barsky, M.A. Payne, G.M. Crooks, Ikaros isoform x is selectively expressed in myeloid differentiation, J. Immunol 170 (6) (2003 Mar 15) 3091–3098. [22] T. Ronni, K.J. Payne, S. Ho, M.N. Bradley, G. Dorsam, S. Dovat, Human Ikaros function in activated T cells is regulated by coordinated expression of its largest isoforms, J. Biol. Chem 282 (4) (2007 Jan 26) 2538–2547. [23] C. Schmitt, C. Tonnelle, A. Dalloul, C. Chabannon, P. Debré, A. Rebollo, Aiolos and Ikaros: regulators of lymphocyte development, homeostasis and lymphoproliferation, Apoptosis 7 (3) (2002 Jun) 277–284. [24] C.G. Mullighan, S. Goorha, I. Radtke, et al., Genome-wide analysis of genetic alterations in acute lymphoblastic leukaemia, Nature 446 (7137) (2007 Apr 12) 758–764. [25] F. Klein, N. Feldhahn, S. Herzog, M. Sprangers, J.L. Mooster, H. Jumaa, M. Müschen, BCR-ABL1 induces aberrant splicing of IKAROS and lineage infidelity in pre-B lymphoblastic leukemia cells, Oncogene 25 (7) (2006 Feb 16) 1118–1124.