Lenalidomide abrogates the survival effect of bone marrow stromal cells in chronic lymphocytic leukemia

In the pathogenesis of chronic lymphocytic leukemia (CLL) the microenvironment plays an important role, as it produces survival signals and mediates drug resistance. Lenalidomide, which has immunomodulatory effect, can enhance the activation of T‐, NK‐cells and endothelial cells, however there are no data available whether it can modulate bone marrow stromal cells (BMSCs). In our study, we investigated the effects of lenalidomide on BMSCs and CLL cells. CLL cells were cultured alone or with BMSCs and were treated with lenalidomide. Apoptosis, immunophenotype, and cytokine secretion of BMSCs and CLL cells were determined by flow cytometry. Lenalidomide slightly increased the apoptosis of CLL cells and abrogated the anti‐apoptotic effect of BMSCs on CLL cells. Lenalidomide treatment decreased the expression of antigens on CLL cells, which mediate the interactions with the microenvironment. Interestingly, lenalidomide enhanced the expression of IRF4 and the co‐stimulatory molecule CD86. The secretion of several cytokines was not changed significantly by lenalidomide. CD49d‐negative CLL cases were more sensitive to lenalidomide treatment. Our results suggest that lenalidomide has a limited effect on BMSCs, but it renders CLL cells more immunogenic and unresponsive to survival signals provided by BMSCs.


-513
Besides genetic factors, the microenvironment mediates resistance to conventional chemo-immunotherapy (e.g. fludarabine, cyclophosphamide, rituximab) or new targeted therapies (e.g. venetoclax, ibrutinib). 3,4 Lenalidomide (Revlimid, CC-5013), a secondgeneration immunomodulatory drug (IMiD) -which is now recommended as maintenance therapy for CLL 5 according to the 2019 NCCN guideline-exerts immunomodulatory and antiproliferative effect. 6 The known direct molecular target of lenalidomide is Cereblon. Lenalidomide causes selective ubiquitination and degradation of two lymphoid transcription factors, IKZF1 and IKZF3, and decreases the expression of interferon regulatory factor 4 (IRF4). 7 Lenalidomide enhances the activation of T-and NK-cells 8 and regulates the endothelium. 9 However, there are no data whether it can modulate bone marrow stromal cells (BMSCs), which provide several survival factors for CLL cells. 1 In our study, we investigated whether lenalidomide influence the bone marrow stromal cells regarding the protection from apoptosis and the cytokine production. We showed that lenalidomide abrogated the anti-apoptotic and immunophenotype alteration effect of BMSCs on CLL cells. Furthermore, lenalidomide treatment increased the level of IRF4 in CLL cells. Interestingly, the secretion of several cytokines was not changed significantly by lenalidomide. Our results suggest that lenalidomide has a limited effect on BMSCs, but renders CLL cells unresponsive to survival signals provided by BMSCs.

| Patients, samples and drugs
Samples of 21 CLL patients (mean age 73.1. years (54-85), female/ male ratio (8/13)) were used in the study. The diagnoses were established according to the World Health Organization (WHO) classification of tumours of lymphoid tissues. 10 The patients were not previously treated or had not received treatment in the preceding six months. From the 21 studied cases only three cases showed 11q deletion, one case with 11q and 13p deletion and one case had TP53 mutation. Written informed consent was obtained from all participants, and the study was conducted in accordance with the Declaration of Helsinki and has been approved by the local Ethics Committee in Semmelweis University (199/2015). Lenalidomide (provided by Celgene) was used in 10 μM concentrations as used before in an in vitro study. 11

| Cell isolation and cell culture
CLL cells were isolated from peripheral blood by Ficoll density gradient centrifugation (n = 13) or by EasySep immunomagnetic positive selection (n = 8). After Ficoll (Histopaque-1077, Sigma Aldrich, USA) density gradient centrifugation the CLL cell ratio was above 90% among peripheral blood mononuclear cells (PBMCs). We performed B-cell isolation by EasySep CD19 Positive Selection Kit II (Stemcell TM technologies, Canada) in order to achieve the maximal CLL cell ratio in our samples. Consequently, the CLL cell ratio was proved to be above 99% (Suppl. Figure S1). We performed the positive selection according to the manufacturer's instructions.
The cells were cultured as previously described. 12 Briefly, 2 � 10 6 /ml CLL cells were cultured in RPMI-1640 (Sigma) supplemented with 10% FBS (Biosera, Philippines) medium alone or in co-culture with BMSCs for 7 days. Lenalidomide was added on the first day. The BMSC cultures, isolated from bone marrow aspirates of healthy individuals or ITP patients without abnormal cells (as determined by flow cytometry), were prepared as described previously. 13 BMSCs were cultured in 24-well plates maintained at a concentration of 2 � 10 4 cells/ml in Dulbecco's modified Eagle's medium (DMEM; Sigma) supplemented with 20% FBS (Biosera). In co-culture experiments, the DMEM medium of the BMSC monolayers was replaced with CLL cell suspension (1 ml of RPMI medium containing 2 � 10 6 CLL cells were added to each well).

| Immunophenotyping of CLL cells by flow cytometry
CLL cells were stained with fluorochrome-conjugated monoclonal antibodies (see in Suppl. Table S1). The measurements were carried presented. In case of Mcl-1 staining, results of the isotype control were controversial, therefore we used the simple MFI values. The antibody was validated with an Mcl-1 positive cell line.

| Determination of apoptosis by subG1 measurement
SubG1 apoptosis measurements were performed as described by Gong et al. 15

| Cytokine detection by CBA assay
The concentration of cytokines was determined by cytokine bead assays (CBA immunoassay) (Suppl Table S2). We cultured CLL cells with or without stromal cells for 7 days and treated them with lenalidomide. We cultured BMSCs without CLL cells to check the cytokine production of them and the effect of lenalidomide on them.
We collected supernatant from all of the samples. The assay was carried out according to the manufacturer's instructions. The samples were acquired by a FACSVerse (BD) flow cytometer and analysed by FCAP Array software (BD). The results were presented as relative cytokine levels. Cytokine levels were normalized to the value measured in the medium of the untreated CLL cells cultured alone.

| Statistical analysis
All variables were tested for normal distribution to select the appropriate parametric or non-parametric statistical procedure.
Paired t-test, Wilcoxon paired test and Mann-Whitney U-test were used for statistical evaluation using the SPSS statistics software, version 25.0 (SPSS, USA). Statistical significance was set at p < 0.05.

| Lenalidomide inhibits the anti-apoptotic effect of BMSCs on CLL cells
In our study, we investigated the effect of lenalidomide on CLL cells cultured alone or in co-culture with BMSCs for 7 days. We found that  Figure S2).
Since CD49d is an important factor in the communication of CLL cells and microenvironment, we compared the survival of CD49dpositive (n = 10) and CD49d-negative (n = 11) samples ( Figure 1B).
We showed earlier that the in vitro apoptosis rate was lower in CD49d-negative cases. 12 Here we found that the CD49d-negative

| The effect of lenalidomide on the cytokine production of stromal cells
It was shown that lenalidomide alters the cytokine production of T-cells and NK-cells in the microenvironment of CLL cells. 8,9 Here, we investigated the effect of lenalidomide on the cytokine production of BMSCs. We measured the level of IL-6, IL-8, IL-1β, human bFGF,

| Lenalidomide alters the immunophenotype of CLL cells
The immunophenotype of CLL cells changed during co-culturing with BMSCs as showed in our previous work. 12

| DISCUSSION
Lenalidomide, a second-generation derivative of thalidomide, is approved for the treatment of multiple myeloma, 18 myelodysplastic syndromes and mantle cell lymphoma. 19,20 In these diseases, it has direct cytotoxic and immunomodulatory effects 21 on pathologic cells.
The drug also has a significant effect in CLL, which is mediated by the inhibition of the proliferation of CLL cells, enhanced immune responses, and the reduction of survival factors from the microenvironment. 6,8 In the treatment of CLL, lenalidomide was optimally applied in combination with other drugs. [22][23][24] The direct cytotoxic effects of lenalidomide in CLL have been investigated in a few studies, 6,25 and no notable apoptosis rate could be detected after 48 hours. However, Acebes-Huerta A. et al. observed a significant apoptosis rate after 7 days. 8 We cultured isolated CLL cells in the presence or absence of BMSCs and treated them with lenalidomide for 7 days and we also detected significant but not very high level apoptosis induced by lenalidomide. Furthermore, it inhibited the anti-apoptotic effect of BMSCs. Similarly, the group of Maffei R. . The difference in IRF4 expression was investigated between CD49d-negative (n = 9) and CD49dpositive (n = 7) samples. Relative IRF4 expression was calculated by the difference between the median fluorescence of isotype controls and IRF4-labeled samples. The bar diagram represents the mean ± SEM of relative IRF4 expression values. (B). The CLL cells were cultured alone or in the presence of bone marrow stromal cells (str) for 7 days (n = 5) and were treated with lenalidomide (Len). IRF4 and Mcl-1 expression were investigated after 7 days. The bar diagram represents the mean ± SEM of IRF4 and Mcl-1 expression values. Mann-Whitney and Wilcoxon tests were performed for statistical evaluation. Significant differences from CLL samples were marked with an asterisk (*), significant differences from CLL + stroma co-culture samples were marked with a dagger ( †). Statistical significance was set at p < 0.05

F I G U R E 3
The effect of lenalidomide on the cytokine production of BMSCs. We investigated the effect of lenalidomide on the cytokine production of BMSCs. CLL cells were cultured alone or in the presence of bone marrow stromal cells (str) for 7 days and were treated with lenalidomide in 10 μM concentration (Len). The supernatant was collected after 7 days treatment and cytokine level (IL-6, IL-8, and human bFGF) was analysed using a CBA assay. We normalised the level of cytokines with the MFI of CLL samples. The bar diagram represents the mean ± SEM of relative cytokine level in the supernatant of CLL cases (n = 10). Significant differences from CLL samples were marked with an asterisk (*). Statistical significance was set at p < 0.05 KRISTON ET AL. Patients who responded to lenalidomide therapy showed a more pronounced decrease of IL-6, IL-8, VEGF and bFGF than nonresponders. 31 Lenalidomide decreased the level of CCL2, CXCL12, and HGF1 produced by NLCs. We supposed that lenalidomide also decreases the cytokine production of BMSCs. We confirmed the former results that IL-6, IL-8, and bFGF levels increased in the coculture of CLL cells and BMSCs. 13,32 Interestingly, although the secretion of IL-6, IL-8 28 was slightly decreased by lenalidomide in accordance with a previous work, 31 the levels of other cytokines (human bFGF, sCD54, sCD154, and TNF) were not altered by lenalidomide treatment.
The clinical effects of lenalidomide treatment were associated with the upregulation of CD40, CD80, CD86, CD154 on CLL cells. 25,33,34 Lenalidomide also influenced the expression of CD20, however, the results are controversial: Lapalombella R. et al.
detected the upregulation of CD20 after lenalidomide treatment. 8 We investigated the effect of lenalidomide on the immunophenotype of CLL cells. Here we show that lenalidomide decreased the expression of markers mediating microenvironmental interactions, such as CD5, CD49d and CD19, and increased the expression of CD86 on CLL cells. Currently, a correlation between the expression of IRF4 and CD86 was shown in CLL and the poor prognostic role of low IRF4 expression was demonstrated due to enhanced tumour immune evasion. 35 Lenalidomide also decreased the level of CD44, which is responsible for cell adhesion-mediated drug resistance in lenalidomide-refractory myeloma cells. 36 Lenalidomide counteracted the stimulating effect of BMSCs in case of CD5, CD19, CD49d and enhanced further the expression of CD86. We compared the two prognostic groups-CD49d-positive and CD49dnegative CLL samples-in terms of how they response to lenalidomide treatment. We found differences in CD5 and CD49d expression: lenalidomide decreased the CD5 expression on CD49dnegative samples and inhibited the stimulating effect of BMSCs on these samples. This process may be due to anti-proliferative effect of lenalidomide, 6 because CLL cells with high CD5 expression have a higher proliferation rate as shown in Calissano's study. 37 Furthermore, in CD49d-expressing CLL cells, lenalidomide was able to decrease the CD49d level, and BMSCs had no effect on its expression.
Our results suggest that lenalidomide gives rise to apoptosis of CLL cells, but exert an only minimal effect on BMSCs. Lenalidomide alters the immunophenotype of CLL cells rendering them more immunogenic and inhibiting the protective effect of the bone marrow stroma. Here, we prove for the first time that CD49dnegative cells are more sensitive to the pro-apoptotic and immunomodulatory effects of lenalidomide. Thus, such cases may respond better to lenalidomide treatment and are the optimal candidate for this therapy.