To the Editor:
With the advent of small molecule inhibitors, sustained responses in chronic lymphocytic leukemia (CLL) patients have been achieved. Ibrutinib, an inhibitor of Bruton’s tyrosine kinase (BTK), is a particularly potent treatment for CLL, with overall response rates of 82% in treatment-naive and 43% in relapsed-refractory patients . These responses are accompanied by transient lymphocytosis in 50–70% of patients and a concomitant reduction of tumor burden in lymphoid organs . We have recently demonstrated that the high-risk, CD49d-high CLL subgroup  displays reduced lymphocytosis and minor lymph node shrinkage upon ibrutinib treatment . CD49d is the alpha subunit of the integrin VLA-4, which plays an essential role in the tissue retention of cells, orchestrating their adhesion to stromal cells, e.g., follicular dendritic cells (fDCs), which in turn express the VLA-4 ligand VCAM-1. To facilitate this strong adhesion, VLA-4 needs to be activated . We have demonstrated that in CLL the extended, high-affinity conformation of VLA-4 can be triggered by B cell receptor (BCR) stimulation . Notably, upon treatment with ibrutinib, CD49d-high CLL cells maintain residual BCR-mediated inside-out activation of VLA-4, thus explaining the reduced clinical response of CD49d-expressing CLL .
Eµ-TCL1-transgenic (TCL1-tg) mice represent a widely used mouse model for CLL . By using this model, here we found that: (i) murine leukemic cells express high levels of surface CD49d; (ii) the BCR-VLA-4 axis in TCL1-tg mice, as investigated upon BCR stimulation and/or exposure to BCR inhibitors, resembles that of the human CD49d-high CLL cohort; (iii) the in vivo distribution of TCL1-tg leukemic cells is dependent on VLA-4-mediated interactions with the microenvironment. Our data demonstrate the value of TCL1-tg mice as a model for CD49d-high CLL and corroborate the importance of retained BCR-induced VLA-4 activation in a clinical setting.
First, to compare VLA-4 expression patterns between human CLL cells and TCL1-tg cells, we measured CD49d expression of human or murine healthy B cells and leukemic cells by flow cytometry (antibodies: see Supplementary Table 1). Human blood samples (N = 406; Supplementary Table 2) were obtained upon written informed consent and ethical approval. Blood-derived B cells from age-matched healthy individuals (N = 32) and CD49d-high CLL cells (N = 116) expressed similar levels of CD49d (Fig. 1ai). Previous reports have shown that CD49d-high CLL cells express lower CD49d levels than healthy B cells. However, our healthy donor group consisted of strictly age-matched individuals (average of healthy cohort: 65.8; CLL cohort: 69.63 years), indicating that CD49d expression might be reduced by age in humans. Moreover, CD49d expression of blood-derived B cells from age-matched C57BL/6J wild-type (WT) mice was similar to that of leukemic cells from TCL1-tg mice (Fig. 1aii). To investigate whether the expression of the VLA-4 subunits (CD49d and CD29) had an organ-specific pattern, we measured B cells and leukemic cells of spleen (SPL), bone marrow (BM), lymph nodes (LN), blood (BL), liver (LIV), and the peritoneal cavity (PC) from WT and TCL1-tg mice. Remarkably, expression levels of both integrin subunits were similar in all organs. CD49d mean fluorescence intensity ratio (MFIR) was comparable in WT and TCL1-tg mice (mean MFIR of WT = 9.893; TCL1-tg = 9.710) (Supplementary Fig. 1a), while the expression of the CD49d-coupled beta subunit CD29 (beta1) was higher in TCL1-tg mice (mean MFIR of WT = 7.261; TCL1-tg = 21.920) (Supplementary Fig. 1b). Altogether, our data demonstrate that the TCL1-tg model can be used as a murine surrogate for the CD49d-high CLL cohort, with stable expression. As expression levels in the spleen and blood were comparable, and their cellular compositions are highly similar, we used murine splenocytes in subsequent experiments.
To relate the CD49d expression pattern to function, we investigated the mechanistic basis of VLA-4 activation by BCR stimuli. In human CLL, BCR-induced VLA-4 activation can be detected by the conformation-sensitive anti–human CD29 antibody HUTS-21 [4, 7], which does not recognize the murine epitope. Therefore, here we used the small molecule VLA-4 ligand LDV  and measured its binding to leukemic cells by flow cytometry as outlined in Supplementary Fig. 2. With this experimental layout, we obtained real-time information about the kinetics of VLA-4 affinity regulation resulting in enhanced ligand binding. The speed by which unlabeled LDV replaces its FITC-conjugated counterpart (koff) is used to determine integrin affinity . If VLA-4 and its ligand bind with high affinity, dissociation of LDV-FITC is slower and the replacement time is longer, resulting in lower koff values (below 0.02 s−1). Conversely, low VLA-4-ligand affinity leads to fast LDV-FITC dissociation, shorter replacement time, and higher koff values (above 0.06 s−1). When purified TCL1-tg leukemic cells were stimulated with anti-mouse-IgM, VLA-4 affinity increased, with a decrease in koff from 0.055 to 0.04 s−1. This shows for the first time that in TCL1-tg mice BCR stimulation induces changes in VLA-4 conformation. Ibrutinib pretreatment did not fully prevent this inside-out activation of VLA-4 (Fig. 1bi), recapitulating the observations in human CD49d-high CLL cells . Next, we treated the cells with the phosphoinositide-3-kinase delta (PI3Kδ) inhibitor idelalisib . Treatment of leukemic cells with idelalisib (Fig. 1bii) and the ibrutinib/idelalisib combination (Fig. 1biii) was highly effective with no difference between the affinities of unstimulated compared with stimulated + idelalisib-treated or stimulated + combination-treated cells. In human CLL, neither ibrutinib nor idelalisib is sufficient on its own to prevent inside-out signaling via the BCR , suggesting that the stimulation can bypass both BTK and PI3Kδ. In TCL1-tg mice, PI3Kδ seems to be a dominant molecule that mediates inside-out signaling between the BCR and VLA-4.
To test that the residual BCR-induced VLA-4 activation was not a result of insufficient inhibition by ibrutinib, we measured the phosphorylation of signaling molecules of the BCR pathway in TCL1-tg splenocytes treated with/without BCR inhibitors by phosflow. Ibrutinib and idelalisib showed specific and expected patterns of inhibition. Single-agent ibrutinib treatment significantly reduced the phosphorylation of BTK (Fig. 1ci) and spleen tyrosine kinase (SYK, Fig. 1cii) after BCR stimulation. Extracellular signal-regulated kinase 1/2 (ERK1/2, Fig. 1ciii) phospho signal was decreased by both inhibitors. The phosphorylation of Akt (Fig. 1civ), the downstream signaling component of PI3K, was efficiently inhibited by idelalisib. Both inhibitors showed expected inhibition, therefore, the residual BCR-mediated VLA-4 activation in ibrutinib-treated cells was not a result of suboptimal BTK blocking.
Having examined the connection between the BCR and VLA-4 in vitro, we next investigated the importance of VLA-4 in engraftment and early disease onset of TCL1-tg-driven murine leukemia in vivo. Therefore, we transplanted WT mice with primary TCL1-tg splenocytes  and treated the recipients with vehicle or the small molecule CD49d-inhibitor firategrast . Recipient mice were sacrificed and their organs analyzed by flow cytometry 21 days post transplantation. Firategrast-treated mice showed an overall reduction of leukemic cells in the spleen (Fig. 2ai), accompanied by significant spleen weight reduction (Fig. 2aii). Since leukemic cells need to enter the follicles in the white pulp of the spleen and access fDCs expressing VCAM-1 and providing proliferative signals , VLA-4 inhibition prevents the cells from staying in this supportive niche. Indeed, we found a significant increase of CD5+/CD19+ cells in the blood (Fig. 2aiii) upon VLA-4 inhibition, without changes in tumor infiltration in the BM (Fig. 2aiv). This finding was surprising given the prominent role of VLA-4 in CLL retention in the BM , but at this stage of engraftment the amount of leukemic cells was still low, and TCL1-tg tumor growth is driven by the spleen, especially in early stages .
To investigate a more advanced disease stage, we repeated the treatment study using the inhibitory rat-anti-mouse CD49d antibody PS/2 (Fig. 2b, treatment schedule: see legend). For long-term treatment, we switched to antibody therapy because firategrast has a fast turnover in the organism, although the in vitro inhibitory effects on VLA-4 ligand binding of the two substances were comparable (confirmed by adhesion assays, data not shown). The mice were sacrificed 6 weeks post transplantation. Analysis of the organs revealed decreased tumor burden in the spleen (Fig. 2bi), along with slightly decreased spleen weight (Fig. 2bii). At this stage, a significant decrease of the tumor load in the blood of PS/2-treated mice was also observed (Fig. 2biii); furthermore, tumor burden in the BM of PS/2-treated mice was significantly decreased (Fig. 2biv). Further investigations are needed to determine whether the redistribution by CD49d inhibition might overcome drug resistance or improve the effectiveness of other therapeutic agents by detaching CLL cells from VCAM-1-expressing stromal cells.
Based on our findings, we propose that TCL1-tg mice represent an adequate model for CD49d-high CLL patients, and provide evidence that BCR-induced VLA-4 inside-out activation remains functional upon treatment with ibrutinib, being in turn hampered upon exposure to idelalisib. This indicates that PI3K, rather than BTK, is an essential part of the signaling between the BCR and VLA-4 in the TCL1-tg model, in part confirming the observation made in humans . Furthermore, we specified the role of CD49d in disease development by directly targeting this integrin in vivo, which led to a redistribution of TCL1-tg cells transplanted into WT mice. From a clinical perspective, we suggest that VLA-4 inhibition or additional targeting of BCR pathway molecules, e.g., by PI3K inhibitors other than idelalisib to avoid its severe side effects  and potentially targeting different isoforms, can be promising additions to ibrutinib therapy for high-risk CD49d-high CLL patients. This could help to overcome residual BCR-mediated signaling bypassing BTK and mediating tissue retention facilitated by VLA-4.
Deeks ED. Ibrutinib: a review in chronic lymphocytic leukaemia. Drugs. 2017;77:225–36.
Barrientos JC, Burger JA, Byrd JC, Hillmen P, Zhou C, Ninomoto J, et al. Characterizing the kinetics of lymphocytosis in patients with chronic lymphocytic leukemia treated with single-agent ibrutinib. Leuk Lymphoma. 2019;60:1000–5.
Bulian P, Shanafelt TD, Fegan C, Zucchetto A, Cro L, Nuckel H, et al. CD49d is the strongest flow cytometry-based predictor of overall survival in chronic lymphocytic leukemia. J Clin Oncol. 2014;32:897–904.
Tissino E, Benedetti D, Herman SEM, Ten Hacken E, Ahn IE, Chaffee KG, et al. Functional and clinical relevance of VLA-4 (CD49d/CD29) in ibrutinib-treated chronic lymphocytic leukemia. J Exp Med. 2018;215:681–97.
Arana E, Harwood NE, Batista FD. Regulation of integrin activation through the B-cell receptor. J Cell Sci. 2008;121:2279–86.
Bichi R, Shinton SA, Martin ES, Koval A, Calin GA, Cesari R, et al. Human chronic lymphocytic leukemia modeled in mouse by targeted TCL1 expression. Proc Natl Acad Sci USA. 2002;99:6955–60.
Chigaev A, Waller A, Amit O, Halip L, Bologa CG, Sklar LA. Real-time analysis of conformation-sensitive antibody binding provides new insights into integrin conformational regulation. J Biol Chem. 2009;284:14337–46.
Chigaev A, Blenc AM, Braaten JV, Kumaraswamy N, Kepley CL, Andrews RP, et al. Real time analysis of the affinity regulation of alpha 4-integrin. The physiologically activated receptor is intermediate in affinity between resting and Mn(2+) or antibody activation. J Biol Chem. 2001;276:48670–8.
O’Brien SM, Lamanna N, Kipps TJ, Flinn I, Zelenetz AD, Burger JA, et al. A phase 2 study of idelalisib plus rituximab in treatment-naive older patients with chronic lymphocytic leukemia. Blood. 2015;126:2686–94.
Hofbauer JP, Heyder C, Denk U, Kocher T, Holler C, Trapin D, et al. Development of CLL in the TCL1 transgenic mouse model is associated with severe skewing of the T-cell compartment homologous to human CLL. Leukemia. 2011;25:1452–8.
Miller DH, Weber T, Grove R, Wardell C, Horrigan J, Graff O, et al. Firategrast for relapsing remitting multiple sclerosis: a phase 2, randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2012;11:131–9.
Heinig K, Gatjen M, Grau M, Stache V, Anagnostopoulos I, Gerlach K, et al. Access to follicular dendritic cells is a pivotal step in murine chronic lymphocytic leukemia B-cell activation and proliferation. Cancer Discov. 2014;4:1448–65.
Brachtl G, Sahakyan K, Denk U, Girbl T, Alinger B, Hofbauer SW, et al. Differential bone marrow homing capacity of VLA-4 and CD38 high expressing chronic lymphocytic leukemia cells. PLoS ONE. 2011;6:e23758.
Gutjahr JC, Szenes E, Tschech L, Asslaber D, Schlederer M, Roos S, et al. Microenvironment-induced CD44v6 promotes early disease progression in chronic lymphocytic leukemia. Blood. 2018;131:1337–49.
Moreno OBT, Zann V, Willson A, Leung P, Connor A. Safety, pharmacokinetics, and pharmacodynamics of ME-401, an oral, potent, and selective inhibitor of phosphatidylinositol 3-kinase P110δ, following single ascending dose administration to healthy volunteers. Clin Ther. 2018;40:1855–67.
We thank all patients for their participation in this study. This work was supported by the PhD program Immunity in Cancer and Allergy (W1213, Austrian Science Fund, FWF) to TNH and RG; the Austrian Science Fund (FWF) P26421, the Paracelsus Medical University Salzburg Project E-17-25-130-HAH and the Forschungskomission Freiburg to TNH; the Associazione Italiana Ricerca Cancro (AIRC), Investigator Grant IG-21687 and the Progetto Ricerca Finalizzata PE 2016-02362756, Ministero della Salute, Rome, Italy to VG; and the Austrian Science Fund projects P26719-B19, I1299-B21 and I3282-B26 (FOR 2036) and I2795-B28 (ERA-NET TRANSCAN-2 program JTC 2014—project FIRE-CLL) to AE. We thank Alexandra Hödlmoser and Ursula Denk for mouse handling support.
Conflict of interest
The authors declare that they have no conflict of interest.
Human blood samples were obtained upon written informed consent (Ethics Committee Salzburg approval: 415-E/1287/4—2011, 415-E/1287/13—2016) from CLL patients at the Department of Internal Medicine III, Paracelsus Medical University, Salzburg, Austria. All mouse experiments were performed under the approval of the Austrian or German Animal Ethics Committee (BMWF-66.012/0030-II/3b/2013, BMWF-66.012/0009-II/3b/2012, 20901-TVG/52/11—2012, 20901-TVG/89/7—2014, BMBWF-66.012/0002-V/3b/2018, G18/155).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Szenes, E., Härzschel, A., Decker, S. et al. TCL1 transgenic mice as a model for CD49d-high chronic lymphocytic leukemia. Leukemia 34, 2498–2502 (2020). https://doi.org/10.1038/s41375-020-0759-3