The STAT5 transcription factor in B-cells of patients with chronic lymphocytic leukemia

A. S. Matvieieva, L. M. Kovalevska, T. S. Ivanivska © 2019 A. S. Matvieieva et al.; Published by the Institute of Molecular Biology and Genetics, NAS of Ukraine on behalf of Biopolymers and Cell. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited ISSN 0233-7657 Biopolymers and Cell. 2019. Vol. 35. N 1. P 30–38 doi: http://dx.doi.org/10.7124/bc.000993


Introduction
Chronic lymphocytic leukemia (CLL) [1] is one of the most common forms of leukemia in Europe and the United States. The incidence rate is approximately 3.5 per 100,000 population (5.0 for men and 2.5 for women) [2,3]. Prior to the onset of CLL, 5-10% of patients develop monoclonal B cell lymphocytosis (MBL), when the number of B-lymphocytes in the peripheral blood counts 5·10 4 -2·10 6 /ml. MBL occurs in individuals over 40. With a frequency of about 1% per year, MBL can progress to CLL, when the number of B cells is 5-10·10 6 /ml [4].
CLL develops due to the slow accumulation of the long-lived, but immunologically incompetent B-lymphocytes. Such cells are often referred to as "immuno-senescent", that describes their inability to differentiate into plasma cells, producing antibodies [4]. CLL cells do not proliferate, can not be activated by ligands and do not undergo apoptosis. Importantly, the transforming Epstein-Barr virus (EBV) could infect CLL cells, but infected cells do not proliferate even in vitro. It was shown, that one of the important viral proteins, i.e. LMP1, is not expressed in CLL cells [5,6]. Obviously, that several transcription factors, ATF-2/c-Jun, for example [5], are missing or not activated in CLL cells.
The activation of the canonical TGFB pathway in B cells usually leads to induction of pro-apoptotic BMF, BIM, BAX and, and as a consequence, to apoptosis [7]. In CLL cells, the level of BCL2 does not differ from the expression of this gene in the peripheral blood B-cells of healthy individuals. TGFB receptors (TGFBRs) are expressed approximately equally in B-CLL and peripheral blood B-lymphocytes [8,9]. However, most of the genes that are usually induced by activation of the TGFB-SMAD2/3 pathway, namely BCL2L1 (BCL-XL), CCND2 (cyclin D2), ID1, MYC, ATF3, TGIF1 andKLF10 (TIEG) are basically not expressed in CLL cells [9].
In the present paper, the expression levels of the STAT1-6 genes, the phosphorylation status of the STAT5 protein and STAT5 cellular localization were studied in CLL cells and in the peripheral blood B-cells of healthy donors, with the aim to find a cause of inhibition of the IL-2-STAT5 pathways upon CLL.

Materials and Methods
The samples of the peripheral blood of 9 patients with CLL and one patient with B-cell prolymphocytic leukemia (BCPL) were obtained from the staff of the Department of Onco-hematology (headed by Professor D.F.Gluzman) at the R.E. Kavetsky Institute of Experimental Pathology, Oncology and Radiobiology (IEPOR) of National Academy of Sciences of Ukraine. In order to verify the CLL diagnosis, the immunocytochemical methods were used utilizing anti-alkaline phosphatase (APAAP) labeled streptavidin-biotin and alkaline phosphatase (LSAB-AP) and a wide panel of monoclonal antibodies. As a control, B-cells were isolated from peripheral blood of two healthy donors in the ficoll-verografin gradient. T cells were removed by forming rosettes with erythrocytes of sheep, followed by centrifugation. The experimental protocol was approved by a Committee on Bioethics at R.E. Kavetsky IEPOR of National Academy of Sciences of Ukraine.
The CLL cells were isolated from peripheral blood in the ficoll-verografin gradient. 50000 cells were attached to a glass slide for the further immune-fluorescent analysis.
Double staining of cells was performed, according to the following scheme: rabbit anti-phosphorylated STAT5 (recognizing both, STAT5A and STAT5B) (Cell Signaling, USA); secondary anti-rabbit antibodies produced in swine and conjugated with fluorescein-5-iso-  [14], in which 8603 genes were analyzed. For other STAT genes, the initial analysis of 19574 genes is described in a work [15]. Conditions of the analysis: change of expression two-folds at least, only 5% of the best genes was chosen, p = 0.0067.
The STAT5 transcription factor in B-cells of patients with chronic lymphocytic leukemia thiocyanate (FITC, DAKO, Denmark); then mouse monoclonal antibody against STAT5A (Cell Signaling, USA); and the secondary anti-mouse antibodies produced in horse and conjugated with Texas red (TR, DAKO, Denmark). The DNA was stained with Hoechst 3321 (Sigma-Aldrich, USA). The images were captured by a CCD camera (Hamamatsu, Japan), assembled and analyzed in Photoshop.
To analyze the STAT gene expression at the mRNA level, a public database Oncomine was used. Oncomine contains the published data that was collected, standardized, annotated and statistically analyzed by Compendia Bioscience (www.oncomine.com, October 2018, Thermo Fisher Scientific, Ann-Arbor, MI, USA).

Results and Discussion
The bioinformatics analysis of the Oncomine database showed that the STAT genes at the mRNA level were expressed at approximately the same level in CLL cells, as in peripheral blood mononuclear cells (Fig. 1). Of note, the STAT2, STAT5А and STAT5В genes are ex-pressed at lower levels in peripheral blood mononuclear cells and in CLL cells, compared with the STAT1, 3, 4 and 6 genes. The lowest relative values of expression in CLL were shown for STAT5B. The obtained results confirm our previously published data [10,12,13].
Subsequently, the levels of phosphorylation of the STAT5 proteins (A and B isoforms) were assessed as well as the STAT5 cellular localization, using immunostaining. The control double staining (excluding one of the primary antibodies) did not show any background signals.
Noteworthy, the STAT5 protein (isoforms A and B) showed basal levels of phosphorylation in the control samples, i.e. B cells of healthy donors (Fig. 2, green signal). The phosphorylated protein was observed almost exclusively in the nucleus. Moreover, STAT5 formed large nuclear inclusions (indicated by green arrows in Fig. 2). The STAT5A protein was also localized mainly in the nucleus in B-cells of healthy donors (red signal in Figure 2), but a proportion of protein was observed in the cytoplasm as well (the red arrows in Fig. 2). Of note, the signals of phosphorylated STAT5 and STAT5A were partially co-localized in the nucleus (Fig.  2, marked with asterisks), indicating activation of the IL-2-STAT5 (JAK-STAT5) pathway in B cells of healthy donors. Previously, the constitutively active STAT5 was found in the nucleus of rapidly proliferating malignant hematopoetic cells [16].
In contrast to the pattern observed in B-cells of healthy individuals, in CLL cells a very low signal of the phosphorylated STAT5 proteins was observed (Fig. 3, in green). The expression levels of the STAT5A protein were quite low as well (Fig. 3, red signal). Noteworthy, when the STAT5A signal was rather high, phosphorylation was not detected (Fig. 3, samples  102884 and 97570, in green). Moreover, the phosphorylated form was localized almost exclusively in the cytoplasm. In several patients the STAT5A protein was practically absent (Fig. 3, samples 103963 and  89819). Also, the expression levels of both isoforms could be very low (Fig. 3, sample  103063).
We have to emphasize that the STAT5A protein and the STAT5 phosphorylated isoforms were localized exclusively in the cytoplasm of peripheral blood mononuclear cells of CLL patients, in contrast to the pattern observed in B cells of healthy individual.
In B cell of the BCPL patient, the phosphorylated form of the STAT5 proteins was not detected. However, a high expression le vel of STAT5A was found, and a large proportion of the protein was localized in the nucleus (Fig. 4, red signal).
The obtained results allow us to speculate that the inhibition of the IL-2-STAT5 (JAK-STAT5) pathway in CLL cells might be due to the low levels of STAT5 phosphorylation, or the complete lack of the STAT5 phosphorylation. Noteworthy also, the STAT5A protein is found mainly in cytoplasm, which suggests that the protein complexes, activating transcription of the STAT5-depending genes are absent in the nucleus. It was demonstrated earlier that the STAT5 functions as transcription factor exclusively in the nucleus [20].
Actually, the reason for such cellular localization of the STAT5 proteins I CLL cells remains an open question. No doubt, that localization of proteins depends on their phosphorylation status. Moreover, activation of the JAK-STAT pathway is also regulated by the formation of homo(hetero)dimers of the various phosphorylated STAT proteins [21]. It is important, to study the phosphorylation status of other STAT proteins, namely, STAT2, 3 and 6, and their cellular localization before and after the interaction of surface IL2R with the corresponding ligand (IL2) in CLL cells.

Conclusions
As we have shown previously, the IL-2-STAT5 (JAK-STAT5) cellular signaling pathway is inhibited in CLL cells. In the present research we found a low level of the STAT5 phosphorylation, or even the complete absence of the phosphorylated protein in leukemic cells. The STAT5A protein is localized mainly in cytoplasm, indicating the absence of active transcriptional complexes in the nucleus, i.e. the STAT5 dependent genes are not induced.