Generation of Inducible BCL11B Knockout in TAL1/LMO1 Transgenic Mouse T Cell Leukemia/Lymphoma Model

The B-cell CLL/lymphoma 11B gene (BCL11B) plays a crucial role in T-cell development, but its role in T-cell malignancies is still unclear. To study its role in the development of T-cell neoplasms, we generated an inducible BCL11B knockout in a murine T cell leukemia/lymphoma model. Mice, bearing human oncogenes TAL BHLH Transcription Factor 1 (TAL1; SCL) or LIM Domain Only 1 (LMO1), responsible for T-cell acute lymphoblastic leukemia (T-ALL) development, were crossed with BCL11B floxed and with CRE-ER/lox mice. The mice with a single oncogene BCL11Bflox/floxCREtg/tgTAL1tg or BCL11Bflox/floxCREtg/tgLMO1tg were healthy, bred normally, and were used to maintain the mice in culture. When crossed with each other, >90% of the double transgenic mice BCL11Bflox/floxCREtg/tgTAL1tgLMO1tg, within 3 to 6 months after birth, spontaneously developed T-cell leukemia/lymphoma. Upon administration of synthetic estrogen (tamoxifen), which binds to the estrogen receptor and activates the Cre recombinase, the BCL11B gene was knocked out by excision of its fourth exon from the genome. The mouse model of inducible BCL11B knockout we generated can be used to study the role of this gene in cancer development and the potential therapeutic effect of BCL11B inhibition in T-cell leukemia and lymphoma.


Introduction
T-cell neoplasms, including T-cell acute lymphoblastic leukemia (T-ALL) and T-cell lymphoma (TCL), divided into peripheral T-cell lymphomas (PTCL) and cutaneous T-cell lymphomas (CTCL), comprise a heterogeneous group of hematopoietic tumors. T-ALL is an aggressive malignancy of early T-cell progenitors characterized by high numbers of blast cells in the bone marrow and peripheral blood, enlargement of mediastinal lymph nodes, and often central nervous system involvement. T-ALL has an incidence of 0.3/100,000 and accounts for approximately 15% of pediatric and 25% of adult ALL cases [1]. PTCL and CTCL originate from mature T-cells, have an incidence of 1.0/100,000 and 0.8/100,000 respectively, and together constitute 15% of all non-Hodgkin lymphomas (NHL) in the USA [1]. Due to the ageing of the population, the incidence of TCL is increasing. These neoplasms often present at an advanced stage at the time of diagnosis, and most commonly have an aggressive clinical course requiring prompt treatment. In contrast to B-cell precursor ALL, which has a very good prognosis with conventional therapy, and B-cell NHL, in which substantial clinical progress has been made with the introduction of monoclonal antibodies such as Rituximab, no comparable advances have been seen in T-ALL and PTCL.
Similar to other types of hematopoietic malignancies, T-cell neoplasms are caused by genetic alterations in hematopoietic precursors (T-ALL) or more mature T-cells (PTCL and CTCL), leading to a variety of changes, including loss of cell cycle control, unlimited self-renewal capacity, impaired differentiation, hyperproliferation and loss of sensitivity to death signals [2,3]. To improve the prognosis in T-cell malignancies, new, gene-targeted therapies have to be developed.
The B-cell CLL/lymphoma 11B gene (BCL11B) is a key player in T-cell development [4], but its role in T-cell malignancies is still unclear. Our group originally identified a chromosomal rearrangement in T-ALL involving BCL11B [5]. We found a high expression of BCL11B in T-ALL and showed that the in vitro suppression of the BCL11B gene leads to massive apoptosis in malignant, but not in normal T-cells [5,6]. Therefore, BCL11B might be an attractive target for the specific therapy of T cell leukemia and lymphomas.
TAL BHLH Transcription Factor 1 (TAL1;SCL) and LIM Domain Only 1 (LMO1) are oncogenes whose activation plays an important role in the development of a fraction of human T-ALL. TAL-1 is an essential transcription factor in normal and malignant hematopoiesis. It is required for the specification of the blood program during the development, adult hematopoietic stem cell survival and quiescence, and terminal maturation of select blood lineages [7]. Following ectopic expression, caused by a chromosomal translocation t(1;14) or a fusion to the STIL Centriolar Assembly Protein (STIL) exogenous promoter, TAL-1 contributes to oncogenesis in T-ALL. TCL1 activities are all mediated through the nucleation of a core quaternary protein complex (TCL1:E-protein:LMO1/2:LDB1) and the recruitment of additional regulators in a lineage-and stage-specific context. LMO1 belongs to a large family of proteins that are required for many developmental processes and are implicated in the onset or the progression of several cancers, including T cell leukemia, breast cancer and neuroblastoma. It contains two protein-interacting LIM domains that operate through nucleating the formation of new transcriptional complexes and/or by disrupting existing transcriptional complexes to modulate gene expression programs. Through these activities, LMO1 has important cellular roles in processes that are relevant to cancer such as self-renewal, cell cycle regulation and metastasis [8]. Transgenic mice bearing both TCL1 and LMO1 human oncogenes spontaneously develop T cell malignancies [9]. Here, we report the generation of a new model of inducible BCL11B knockout in TCL1/LMO1 transgenic mice to study the role of this gene in T cell malignancies.

Results
Four transgenic mouse strains: BCL11B flox/WT , Cre tg , SCL tg and LMO1 tg were sequentially crossed in order to obtain a BCL11B flox/flox Cre tg/tg SCL tg LMO1 tg mouse model of inducible BCL11B knockout in spontaneously developing T cell malignancies ( Figure 1).

Generation of the Final Experimental Mouse Model of Inducible BCL11B Knockout in T-Cell Malignancies
The BCL11B flox/flox Cre tg/tg TAL1 tg and BCL11B floxfloxl CRE tg/tg LMO1 tg animals were mated, and one month old progeny were genotyped via PCR from tail DNA. One fourth of the progeny carried both oncogenes (BCL11B flox/flox LMO1 tg /TAL1 tg /Cre tg/tg ), and >90% of them developed T-ALL and were used for studies. The animals carrying only one oncogene, either LMO1 (BCL11B flox/flox CRE tg/tg LMO1 tg ) or TAL1 (BCL11B flox/flox CRE tg/tg TAL1 tg ), were healthy, bred normally and were used to maintain the mice in culture. Mice born without oncogenes (BCL11B flox/flox CRE tg/tg ) were used as a control.
2.4. Creation of the BCL11B flox/del CRE tg/tg LMO1 tg and BCL11B flox/del CRE tg/tg TAL1 tg Mice To increase the efficacy of the BCL11B knockout, a heterozygous mouse strain with an inherited deletion of the BCL11B gene (BCL11B flox/del ) was generated by inducing gene deletion in one of the parents by tamoxifen administration. Surprisingly, almost all (29 out of 30) offspring were born with one deleted allele ( Figure 3). For further breeding, BCL11B flox/del and BCL11B flox/flox mice were coupled to avoid the bi-allelic deletion of BCL11B, which was previously described [10] and also proven by us to be lethal. Similarly to the previously described animals with an intact BCL11B locus, the animals with BCL11B flox/del CRE tg/tg LMO1 tg mice were mated with the BCL11B flox/flox Cre tg/tg TAL1 tg , or BCL11B flox/flox CRE tg/tg LMO1 tg with BCL11B flox/del CRE tg/tg TAL1 tg . One month old progeny were genotyped via PCR from tail DNA. One fourth of the progeny carried both oncogenes, and half of them carried the BCL11B deletion (BCL11B flox/del LMO1 tg /TAL1 tg /Cre tg/tg ).
To increase the efficacy of the BCL11B knockou an inherited deletion of the BCL11B gene (BCL11B flo deletion in one of the parents by tamoxifen administr of 30) offspring were born with one deleted alle BCL11B flox/del and BCL11B flox/flox mice were coupled to av which was previously described [10] and also prove previously described animals with an intact BCL11B flox/del CRE tg/tg LMO1 tg mice were mated wit BCL11B flox/flox CRE tg/tg LMO1 tg with BCL11B flox/del CRE tg/tg T genotyped via PCR from tail DNA. One fourth of t and half of them carried the BCL11B deletion (BCL11

Diagnosis of Hematologic Malignancies
As originally reported by Aplan et al., within carrying both oncogenes (LMO1 and TAL1), spontane leukemia/lymphoma [9]. Starting from the fourth LMO1/TAL1 mice were monitored for the developme

Diagnosis of Hematologic Malignancies
As originally reported by Aplan et al., within 3-6 months after birth the progeny, carrying both oncogenes (LMO1 and TAL1), spontaneously developed an aggressive T cell leukemia/lymphoma [9]. Starting from the fourth month of life, every two weeks the LMO1/TAL1 mice were monitored for the development of malignancy. Lymphoma development was assessed by palpation and magnetic resonance imaging (MRI) (Figure 4), and the leukemia development was determined based on white blood cell (WBC) counting and peripheral blood smear analysis ( Figure 5). Leukemia was diagnosed when the WBC was >10,000 cell/µL and/or the blast content >20%.

Determination of the Immunophenotype of the Malignant Disease by Flow Cytometry
The diagnosis of T cell malignancy was further confirmed by FC of the peripheral blood. In contrast to healthy mice, in animals that developed T cell malignancies, besides an increased ratio of CD3+ cells, double-positive CD4+/CD8+, CD25+ or terminal deoxynucleotidyl transferase (TdT+) subsets of T cells were detected ( Figure 6). The most common feature of all cases was the increased (30-90%) subset of CD3+ leukocytes (17/23; 74%). In 61% of the cases (14/23), double-positive CD4+/CD8+ T cells were observed, normally not present in the peripheral blood. In the majority of the CD4+/CD8+ cases, abnormal TdT+ (10/14; 71%) and CD25+ (8/14; 57%) cell subsets were also observed. In 10/13 (77%) of the analyzed cases, NK cell-specific antigen NK-1.1 was expressed, in half of the cases as a sole abnormal phenotype, in the other half accompanied by the CD4/CD8 phenotype, of which three also expressed TdT.

Determination of the Immunophenotype of the Malignant Disease by Flow Cytometry
The diagnosis of T cell malignancy was further confirmed by FC of the peripheral blood. In contrast to healthy mice, in animals that developed T cell malignancies, besides an increased ratio of CD3+ cells, double-positive CD4+/CD8+, CD25+ or terminal deoxynucleotidyl transferase (TdT+) subsets of T cells were detected ( Figure 6). The most common feature of all cases was the increased (30-90%) subset of CD3+ leukocytes (17/23; 74%). In 61% of the cases (14/23), double-positive CD4+/CD8+ T cells were observed, normally not present in the peripheral blood. In the majority of the CD4+/CD8+ cases, abnormal TdT+ (10/14; 71%) and CD25+ (8/14; 57%) cell subsets were also observed. In 10/13 (77%) of the analyzed cases, NK cell-specific antigen NK-1.1 was expressed, in half of the cases as a sole abnormal phenotype, in the other half accompanied by the CD4/CD8 phenotype, of which three also expressed TdT.

Determination of the Immunophenotype of the Malignant Disease by Flow Cytometry
The diagnosis of T cell malignancy was further confirmed by FC of the peripheral blood. In contrast to healthy mice, in animals that developed T cell malignancies, besides an increased ratio of CD3+ cells, double-positive CD4+/CD8+, CD25+ or terminal deoxynucleotidyl transferase (TdT+) subsets of T cells were detected ( Figure 6). The most common feature of all cases was the increased (30-90%) subset of CD3+ leukocytes (17/23; 74%). In 61% of the cases (14/23), double-positive CD4+/CD8+ T cells were observed, normally not present in the peripheral blood. In the majority of the CD4+/CD8+ cases, abnormal TdT+ (10/14; 71%) and CD25+ (8/14; 57%) cell subsets were also observed. In 10/13 (77%) of the analyzed cases, NK cell-specific antigen NK-1.1 was expressed, in half of the cases as a sole abnormal phenotype, in the other half accompanied by the CD4/CD8 phenotype, of which three also expressed TdT.

Discussion
During the last two decades, extensive knowledge has been accumulated about the physiological role of BCL11B and the role of its mutations in inherited diseases. Although there are several reports on the involvement of BCL11B in hematological malignancies, including gene rearrangements, mutations and overexpression, the role of BCL11B in the development of the disease has not yet been conclusively clarified [11,12]. Additionally, the hypotheses on the possible mechanism of malignant transformation caused by BCL11B alterations are contradictory. While some studies postulated that BCL11B acted as a tumor suppressor and that inactivating mutations played a role in tumor development [13,14], others, including our studies, suggested BCL11B as being an oncogene [6,15] and its downregulation resulting in apoptosis of malignant T cells. To better understand the role of BCL11B in the pathogenesis of T cell malignancies and to test the efficacy of new therapeutic approaches, animal models of disease are needed.
In this study, we report the generation of a new model of inducible BCL11B knockout in LMO1/TAL1 transgenic mice spontaneously developing T cell malignancies. In this model, the animals carrying only one oncogene, LMO1 tg or TAL1 tg , are healthy and immunocompetent. They are easy to maintain in culture and breed normally. When crossed with each other, 90% of the double transgenic animals, LMO1 tg /TAL1 tg , within 3-6 months from birth, spontaneously develop T cell leukemia/lymphoma, resembling human T cell malignancies [9]. Since the animals possess the BCL11B flox/flox CRE tg/tg genetic modification, the BCL11B gene can be inactivated at any time by induction of the Cre-lox knockout system, using synthetic estrogen (tamoxifen) administration.
To improve the efficacy of the BCL11B knockout, we modified that system and introduced an inherited knockout of one of the alleles. The animals were born with one floxed

Discussion
During the last two decades, extensive knowledge has been accumulated about the physiological role of BCL11B and the role of its mutations in inherited diseases. Although there are several reports on the involvement of BCL11B in hematological malignancies, including gene rearrangements, mutations and overexpression, the role of BCL11B in the development of the disease has not yet been conclusively clarified [11,12]. Additionally, the hypotheses on the possible mechanism of malignant transformation caused by BCL11B alterations are contradictory. While some studies postulated that BCL11B acted as a tumor suppressor and that inactivating mutations played a role in tumor development [13,14], others, including our studies, suggested BCL11B as being an oncogene [6,15] and its downregulation resulting in apoptosis of malignant T cells. To better understand the role of BCL11B in the pathogenesis of T cell malignancies and to test the efficacy of new therapeutic approaches, animal models of disease are needed.
In this study, we report the generation of a new model of inducible BCL11B knockout in LMO1/TAL1 transgenic mice spontaneously developing T cell malignancies. In this model, the animals carrying only one oncogene, LMO1 tg or TAL1 tg , are healthy and immunocompetent. They are easy to maintain in culture and breed normally. When crossed with each other, 90% of the double transgenic animals, LMO1 tg /TAL1 tg , within 3-6 months from birth, spontaneously develop T cell leukemia/lymphoma, resembling human T cell malignancies [9]. Since the animals possess the BCL11B flox/flox CRE tg/tg genetic modification, the BCL11B gene can be inactivated at any time by induction of the Cre-lox knockout system, using synthetic estrogen (tamoxifen) administration.
To improve the efficacy of the BCL11B knockout, we modified that system and introduced an inherited knockout of one of the alleles. The animals were born with one floxed and one knocked-out BCL11B allele (BCL11B flox/del CRE tg/tg LMO1 tg and BCL11B flox/del CRE tg/tg TAL1 tg ), were healthy, bred normally, and passed the knocked-out allele to half of their offspring. As originally reported by Wakabayashi et al. [10], the germline knockout of both BCL11B alleles was lethal, and the animals died shortly after birth.
Besides the direct pro-apoptotic effect on malignant T cells, BCL11B suppression was shown to cause the transformation of normal T cells into induced T-to-natural killer (ITNK) cells [16]. Since ITNK cells killed tumor cells in vitro and in vivo, the inhibition of BCL11B may additionally exert its anti-tumor activity by inducing normal T cells to kill the remaining malignant T cells. In our previous studies, we showed that an increased expression of BCL11B leads to chemoresistance accompanied by G1 accumulation [17]. It can be speculated that the opposite, the suppression of BCL11B, should lead to an increased chemotherapy susceptibility. This suggests that combining BCL11B inhibition with chemotherapy could increase the efficacy of targeting cancer cells.

Mouse Models
Heterozygous transgenic mice bearing constructs which express TAL BHLH Transcription Factor 1 (TAL1;SCL) mRNA driven by a fusion to the STIL Centriolar Assembly Protein (STIL) exogenous promoter and LIM Domain Only 1 (LMO1), mimicking common gene dysregulations associated with human T-ALL, were kindly provided by Peter Aplan (National Institute of Health/National Cancer Institute, Bethesda, MD, USA). These mice, bearing human oncogenes, TAL1 or LMO1, responsible for T-cell acute lymphoblastic leukemia (T-ALL) development, are healthy and breed normally. When crossed with each other, the double transgenic mice TAL1 tg /LMO1 tg , within 3 to 6 months from birth, spontaneously develop T-cell leukemia/lymphoma [9].
The C57BL/6-Gt(ROSA)26Sor tm9(Cre/ESR1)Arte (Cre tg ) mice, with inducible Cre recombinase, were purchased from Artemis (Koeln, Germany) [19]. Upon administration, synthetic estrogen (tamoxifen) binds to the estrogen receptor and activates the Cre recombinase. All studies involving animals were performed with the approval of the Local Ethical Committee for Animal Experiments in Poznan (approval code LKE 2/2018).

Genotyping
Genetic modifications were determined via PCR from tail DNA in one month old mice. For the determination of the presence of LoxP sequences within the BCL11B gene, the 5 BCL6912f (mm39; chr12:107,883,573-592) TCGGAAGCCATGTGTGTTCT and the 3 BCL7202r (mm39; chr12:107,883,863-844) TAGATCCCGTGTTCCCTTGC primers were used, amplifying a 291 bp fragment of the intron/fourth exon border of BCL11B in the wild type animals. In the BCL11B floxed mice, with the inserted LoxP sequence and fragments of the cloning vector, the amplicon was 342 bp long.
For the presence of Rosa 26 Cre locus 1242_1: CCATCATCGAAGCTTCACTGAAG and 1242_2: GGAGTTTCAATACCCGAGATCATGC primers were used, amplifying a 315 bp fragment; and as a reaction control 1260_1: GAGACTCTGGCTACTCATCC and 1260_2: CCTTCAGCAAGAGCTGGGGAC primers were used, amplifying a 585 bp fragment of CD79b, as recommended by the supplier (Artemis).

Magnetic Resonance Imaging (MRI)
MRI experiments were carried out using a preclinical horizontal scanner operating at 9.4 T (Agilent) equipped with a 600 mT/m gradient system, and a 40-mm i.d. quadrature millipede type coil was used. During the imaging experiment, animals were anesthetized with 2% isoflurane in a 20/80 air-oxygen mixture (induction 4% isoflurane) and put in a specially designed holder. The temperature of the animal was kept at 37 • C, and the respiration of the animal was also monitored and used to synchronize MRI experiments (1030, SA Instruments Inc., Stony Brook, NY, USA).
MRI images of the proton spin density were collected at two localizations of each animal (abdomen area and neck area) using a fast spin-echo sequence (FSEMS) with the following parameters: TR = 5 s, effective TE = 10 ms, ETL = 8, FOV 35 × 35 mm, 2 averages, matrix size 256 × 256, slices from 12 to 30 depending on animal and localization (slice thickness 1 mm).

Induction of BCL11B Knockout by Tamoxifen Administration
Tamoxifen (Sigma-Aldrich, Burlington, MA, USA) was dissolved in corn oil at a concentration of 10 mg/mL by shaking overnight at 37 • C. To induce Cre recombinase synthesis followed by knockout of the floxed BCL11B gene, the mice were injected intraperitoneally with 1 mg of tamoxifen (100 µL of oil solution) for 3 subsequent days. The efficacy of deletion was checked 7 days after the third administration.

Conclusions
The mouse model we created can be used to study the function of BCL11B in T cell malignancies, especially the effect of BCL11B suppression. Since, to date, no specific BCL11B inhibitor has been invented, this model can be used as a proof of principle for the rationale of therapeutic BCL11B suppression in T cell neoplasms.