Deregulated expression of TCL1 causes T cell leukemia in mice

Abstract

The TCL1 oncogene on human chromosome 14q32.1 is involved in the development of T cell leukemia in humans. These leukemias are classified either as T prolymphocytic leukemias, which occur very late in life, or as T chronic lymphocytic leukemias, which often arise in patients with ataxia telangiectasia (AT) at a young age. The TCL1 oncogene is activated in these leukemias by juxtaposition to the α or β locus of the T cell receptor, caused by chromosomal translocations t(14:14)(q11:q32), t(7:14)(q35:q32), or by inversions inv(14)(q11:q32). To show that transcriptional alteration of TCL1 is causally involved in the generation of T cell neoplasia we have generated transgenic mice that carry the TCL1 gene under the transcriptional control of the p56^lck promoter element. The lck-TCL1 transgenic mice developed mature T cell leukemias after a long latency period. Younger mice presented preleukemic T cell expansions expressing TCL1, and leukemias developed only at an older age. The phenotype of the murine leukemias is CD4⁻CD8⁺, in contrast to human leukemias, which are predominantly CD4⁺CD8⁻. These studies demonstrate that transcriptional activation of the TCL1 protooncogene can cause malignant transformation of T lymphocytes, indicating the role of TCL1 in the initiation of malignant transformation in T prolymphocytic leukemias and T chronic lymphocytic leukemias.

Nonrandom translocations, involving either the Ig or the T cell receptor (TCR) loci, juxtapose cellular protooncogenes to strong enhancer elements, leading to oncogene deregulation and tumor initiation (1–3). Overexpression of the TCL1 gene in humans has been implicated in the development of mature T cell leukemia, in which chromosomal rearrangements bring the TCL1 gene in close proximity to the TCRα or TCRβ regulatory elements (4, 5). T cell leukemias associated with TCL1 activation are most commonly classified as T prolymphocytic leukemia (T-PLL), a very aggressive mature T cell proliferation, and sometimes as T chronic lymphocytic leukemia (T-CLL) (6). T cell leukemias arising in patients with ataxia telangiectasia often carry TCL1 rearrangements. These rearrangements and TCL1 activation are already present in T cell clonal expansions of these patients at the preleukemic stage (7). These findings suggest that TCL1 activation may be the first step in the neoplastic process and might be present at the preleukemic stage for an extended period of time before the occurrence of secondary genetic events, which would then result in the onset of overt leukemia. It has already been shown that the development of aggressive leukemia or lymphoma is a multistep process. For example, follicular lymphoma with BCL2 rearrangements can progress to a more malignant form by acquiring a c-MYC rearrangement (8, 9). Another example is the transition from the chronic to the acute phases in chronic myelogenous leukemia (CML), in which p53 mutations and deletions are observed in 20–30% of the cases of CML in blastic crisis (10).

TCL1 codes for a protein of 114 aa that is expressed in the cytoplasm and in the nucleus of lymphoid cells (5). In normal T cells TCL1 is expressed in CD4⁻CD8⁻ cells, but not in cells at later stages of differentiation (5). We have also shown that the Tcl1 protein has a high similarity to p13^MTCP1 (11), a protein encoded by one of the ORFs of the MTCP1 gene and involved in the t(14:X)(q32:q28) chromosome translocation in T cell proliferative diseases (12). To understand the effect of TCL1 overexpression in the lymphoid compartment and whether the TCL1 deregulation produces lymphoid tumors in experimental models, we generated transgenic mice carrying the TCL1 gene under the control of the T cell-specific gene promoter p56^lck whose expression specifically targets gene expression in the thymus (13). The development of TCL1 transgenic mice represents an optimal system to study the multiple genetic events involved in the leukemogenic process of T-PLL and T-CLL.

METHODS

Constructs and Transgenic Mice.

A schematic representation of the construct used to generate lck-TCL1 transgenic mice is given in Fig. 1. The original plasmid containing the lck proximal promoter (lck^pr), the 3′ untranslated region, and the poly(A) site for the human growth hormone gene (hGH) was kindly provided by Roger Perlmutter, University of Washington, Seattle (13). The construct was obtained by inserting a 350-bp fragment containing the entire human TCL1 coding sequences into a BamHI restriction site downstream to the lck promoter. The construct was freed from vector sequences by digestion with NotI, purified from agarose gel, and injected in fertilized mouse oocytes essentially as described (14). The fertilized mouse oocytes were derived from superovulated (B6C3)F₁ animals. Successful integration of the injected DNA was monitored by Southern analysis of tail tip DNA as described (14), by using as probe the same DNA fragment used to inject the oocytes. Founder mice were backcrossed with C57BL/6J mice.

Figure 1.

(A) DNA construct used in generating human lck-TCL1 transgenic mice. A 0.35-kb cDNA fragment containing the entire human TCL1 coding region was inserted into the BamHI site of a vector containing the lck proximal promoter and the hGH 3′ untranslated region and poly(A) addition site. (B) Southern blot analysis of DNA from tail clips from the first transgenic progeny. Blots were hybridized with the same DNA fragment used for injection of zygotes. Lane 1, DNA from a positive founder mouse; lanes 2–4, DNA from nontransgenic progeny.

Protein Extracts from Mouse Tissues.

Tissues were homogenized with a hand homogenizer for 30–60 sec in a solution containing 50 mM Tris⋅HCl (pH 7.6), 150 mM NaCl, 1% NP40, and 1 mM phenylmethylsulfonyl fluoride. The samples were kept on ice for 15 min followed by centrifugation at 12,000 rpm for 15 min at 4°C. Supernatants were analyzed by Western blot.

Flow Cytometry.

Single-cell suspensions were made at the time of autopsy from spleen, thymus, and lymph nodes of transgenic and control mice in PBS supplemented with 1% bovine calf serum and 0.01% sodium azide (staining solution). Cells were washed in this solution, and for direct labeling 1 × 10⁶ cells were resuspended with 100 μl of fluorescein isothiocyanate (FITC)- or phycoerythrin (PE)-conjugated antibodies diluted 1:100 in staining solution and incubated for 30 min at 4°C. Cells were washed three times with cold PBS, fixed with 1% paraformaldehyde for 15 min, and washed once with PBS. Conjugated FITC anti-CD4, FITC anti-TCR, PE anti-CD8, and T cell receptor Vβ region-specific antibodies (Vβ2–8, Vβ8.2, Vβ8.3, Vβ9, Vβ10b, and Vβ11–14, PharMingen) were used according to manufacturer protocols. Cells were analyzed on a Coulter Profile II flow cytometer. Dyes were exited at 488 nm, and the emissions were collected at 525 nm (FITC) and 575 nm (PE).

Tissue Immunohistochemistry.

Tissues isolated from transgenic mice and control mice were fixed in 4% paraformaldehyde and paraffin-embedded. Five-micron-thick serial sections were cut from formalin-fixed, paraffin-embedded specimens and mounted on Plus slides (Fisher Scientific). These sections were deparaffinized and rehydrated through graded xylene and alcohol series and placed in citrate-buffered solution (pH 6.0), and then heated at 100°C by microwave oven for 20 min for antigen retrieval as previously described (15). Endogenous peroxidase was blocked with 3% hydrogen peroxide, and nonspecific binding was blocked with 10% normal serum. Polyclonal rabbit anti-mouse antibodies specific for recombinant human Tcl1 were used at 1:200 dilution. Immunohistochemical staining was performed with Vectastain ABC kit (Vector Laboratories). Diaminobenzidine was used for coloration, and nuclei were counterstained with hematoxylin. Secondary antibody alone was consistently negative on all sections. Normal goat serum was substituted for the primary antibody in negative controls.

TCR Rearrangements.

DNA isolated from splenocytes and thymocytes was digested with EcoRI restriction enzyme, and Southern blots were obtained as described (16). The blots were hybridized with a DNA probe specific for the Jβ₂ segment of the TCR gene described by Kronenberg et al. (17), followed by autoradiography.

RESULTS

lck^pr-TCL1 Transgenic Mice.

The TCL1 gene is normally expressed in CD4⁻CD8⁻ thymocytes and in pre-B/IgM⁺ cells (5). Based on TCL1 tissue specificity and its involvement in leukemias of T cell origin, the lck proximal promoter was used to target TCL1 expression in the T cell compartment of transgenic mice. A transgenic construct was generated by introducing the coding sequences of human TCL1 cDNA into the BamHI cloning site, which is located downstream of the p56^lck promoter region (Fig. 1A). The 3′ untranslated portion of this construct provided introns, exons, and the poly(A) site for the human growth hormone gene (hGH). Purified DNA of the construct was microinjected into zygotes and transplanted into pseudopregnant females. Founder mice were identified by screening of the transgenic progeny tail DNA by Southern hybridization with a human TCL1 DNA probe, which does not cross-hybridize with mouse sequences (Fig. 1B). Transgenic lines were successfully established and analyzed by Western blot by using antibodies specific for human Tcl1 protein. Fig. 2 shows the immunoblot result of total protein extracts from the thymus, liver, spleen, and kidney from a representative transgenic (TG) and nontransgenic mouse (non-TG). High expression of the human Tcl1 protein was observed in the thymus of the transgenic mouse. No expression was visible in the spleen whole extracts by immunoblot analysis; however, a very low level of expression could be detected in the spleen after enrichment for lymphocytes (data not shown).

Figure 2.

Western blot analysis of protein extracts from transgenic (TG) and nontransgenic (non-TG) mouse tissues. Proteins (100 μg) were loaded per lane: 1, kidney; 2, thymus; 3, liver; and 4, spleen. The control lane, C, contains extracts from the pre-B leukemic cell line 697, which expresses high levels of the Tcl1 protein (5).

lck^pr-TCL1 Transgenic Mice Develop Lymphocytic Leukemia with Very Late Onset.

Transgenic animals were analyzed at 1–3 months of age to determine the presence of developing malignancies. No pathological signs of tumor development were evident, and flow cytometry of the thymocytes showed normal distribution of CD3⁺, CD4⁺CD8⁻, CD4⁻CD8⁺, and CD4⁺CD8⁺ populations. Splenic populations were not analyzed because the organs appeared normal in size and color. However, peripheral blood of some mice showed increases in the percentage of lymphocytes (75–90%) and in the cell counts compared with normal mice (60%). The transgenic mice were kept under close supervision and checked biweekly for signs of illness. Mice became visibly ill around the age of 15–20 months. These mice presented with enlarged spleens, enlarged lymph nodes and thymus, and high white blood cell (WBC) counts (Table 1). An example of a representative enlarged spleen is shown in Fig. 3. The weight of the transgenic spleen of a leukemic animal (TG 39) shown in this figure was 2 g, approximately 30 times the weight of a normal spleen (normal splenic weight was 0.07 ± 0.01 g). Immunohistochemical analysis of tumorigenic spleens with anti-Tcl1 polyclonal antibodies showed staining in numerous clusters of lymphocytes showing a uniform morphology, indicating that leukemic cells express Tcl1 (Fig. 4). Flow cytometric analysis of splenocytes indicated that the phenotype of the leukemic cells is CD3⁺ and CD4⁻CD8⁺ in all the cases analyzed (Table 2). This observation contrasts with the situation in humans where T cell leukemias with TCL1 involvement are most often CD3⁺ and CD4⁺CD8⁻. Pathological examination of peripheral blood smears classified the malignancies as lymphocytic leukemias with the predominant cell type represented by a large (10–15 μ in diameter) lymphoid cell.

Table 1.

Phenotype of leukemias from lck-TCL1 transgenic mice

Age	No. mice/total	Phenotype
1–12	0/5	Leukemia*
12–14	9/13	T cell expansion†
	2/13	Lymphocytic leukemia‡
15–20	6/10	Lymphocytic leukemia

^*Mice were analyzed for clonal expansions, increased lymphoid count, or other abnormalities characteristic of leukemia. 

^†Characterized by an increase in CD3⁺ cells and skewed CD8/CD4 ratio compared with normal mice and no detectable clonal TCR rearrangement (see text for details). 

^‡Leukemia was determined by clonal rearrangement of the T cell receptor as detected by a Jβ2 probe, splenomegly, increased WBC count, and any lymph node or thymic abnormalities (increased size or altered architecture). 

Figure 3.

Whole spleen specimen isolated from a leukemic mouse (Left, TG 39) and a control mouse (Right).

Figure 4.

Immunocytochemical analysis of Tcl1 protein in transgenic mouse spleens. Normal mouse spleen (negative control) stained for Tcl1 protein using polyclonal anti-Tcl1 antibody (A). A young transgenic mouse (<1 year old) stained for Tcl1 protein (B). Arrow in B shows a small cluster of positively stained cells. Small numbers of stained cells can also be observed scattered throughout the specimen. Spleen specimen from a representative animal containing extensive tumor infiltration stained with secondary antibody alone (C) or together with anti-Tcl1 polyclonal antibodies (D). A low (×10) magnification (E) and high (×40) magnification (F) depicting a foci of Tcl1-stained cells in pre- (or early) leukemic mice.

Table 2.

Cell surface phenotype of transgenic lymphocytes

Animal number	Percent of spleen cells positive for			WBC*, × 106
	CD3	CD4	CD8
N 1	36.0	21.0	12.0	1.8
N 2	35.0	24.0	12.0	1.8
N 3	36.0	22.0	16.0	ND
TG 37	63.0	17.0	29.0	2.6
TG 54	57.0	21.0	34.0	6.2
TG 60	54.0	33.0	17.0	11.6
TG 68	55.0	34.0	21.0	ND
TG 72	64.0	37.0	22.0	ND
TG 73	67.0	31.0	27.0	5.6
TG 74	70.0	32.0	30.0	2.4
TG 83	49.0	23.0	30.0	ND
TG 108	65.0	27.0	24.0	2.4
TG 109	63.0	22.0	35.0	1.8
TG 116	52.0	27.0	26.0	24.8
TG 117	55.0	24.0	22.0	ND
TL 25	65.0	1.0	60.0	ND
TL 39	99.0	1.4	97.0	168.0
TL 47	65.0	20.0	40.0	ND
TL 53	75.0	3.0	74.0	ND
TL 71	63.0	21.0	40.0	ND
TL 82	65.0	10.0	73.0	ND

N, normal; TG, transgenic; TL, T cell leukemic mice; ND, not done. 

^*The mean WBC/ml blood for normal adult mice (10–18 months old) was 2.6 ± 0.4 × 10⁶ cells/ml (n = 10). 

Lymphocyte Populations in Transgenic Mice.

Transgenic mice between the age of 12 and 14 months were also analyzed for pathological abnormalities. Some of the animals presented slightly enlarged spleens and a high white blood cell count but with no hyperplasia of the thymus or lymph nodes (Table 2). In two of the mice, TG71 and TG82, the presence of full-blown malignancy was observed (Table 1). Flow cytometric analysis of the T cell populations in transgenic spleens indicated that the CD3⁺ population is high in all the animals examined, ranging between 50 and 70% (Table 1). The percentage of CD8⁺ T cells is nearly doubled, 21–34%, compared with control mice (nontransgenic), whereas the percentage of CD4⁺ cells shows only a slight increase in some specimens. The ratio CD4⁺/CD8⁺ in the spleens of the transgenic mice was 1.0 compared with 1.7 in control mice. Immunocytochemical analysis of the spleens with TCL1 antibodies showed the presence of large numbers of lymphoid cells expressing TCL1 (Fig. 4). It is possible that an early or preleukemic stage exists in these mice, denoted by the increased white blood cell count, expansion in the CD3⁺ and CD8⁺ populations, and increased numbers of TCL1⁺ cells clustered within the spleen (Fig. 4 E and F). We predict that these mice are at increased risk for leukemia as they age. Future experiments will be needed to determine the relationship between each stage of disease.

T Cell Receptor Rearrangements in lck-TCL1 Mouse Tumors.

Southern blot analysis of DNA isolated from leukemic cells with probes for the TCRβ chain gene joining region Jβ2 showed clonal rearranged bands (Fig. 5A). DNA from healthy transgenic spleens showed the TCRβ gene in its germ-line configuration, whereas DNA from leukemic splenocytes show the presence of extra rearranged bands, indicating the presence of clonal T cell populations. As expected, DNA from spleen, thymus, and lymph nodes from the same leukemic animals all present the same TCR gene rearrangements (Fig. 5A). In one representative case the expressed Vβ gene was found to be the Vβ6 gene by flow cytometry (Fig. 5B). The presence of clonal TCR gene rearrangements in conjunction with the phenotypic data confirms the diagnosis of T cell leukemia. DNA from spleens of asymptomatic transgenic mice (mice <1 year old) showed expansions of the CD3⁺ and CD8⁺ cells although clonal TCR gene rearrangements were not detected (Fig. 5A and data not shown), suggesting that these T cell expansions were not clonal at this stage.

Figure 5.

Southern analysis of TCR gene rearrangements in leukemias from transgenic mice. (A) Clonal TCR rearrangements. Southern blot analysis of DNA isolated from the spleens or thymuses of leukemic and nonleukemic mice. DNAs were digested with EcoRI. Blots were hybridized with a probe specific for the TCR Jβ₂ gene segment. Lanes: 1, spleen TG25; 2, spleen non-TG littermate; 3, spleen TG28 (partial digest); 4, spleen TG 43; 5, spleen TG 45; 6, spleen TG 47; 7, thymus TG 47; 8, spleen TG 39; and 10, thymus TG39. The strong 2.2-kb bands represent the TCR Jβ₂ gene in its germ-line configuration; arrows point to weaker hybridizing bands indicating the presence of clonal TCR gene rearrangements. (B) Expression of a specific TCR-Vβ on the surface of mouse leukemic cells. Single-cell suspensions of splenocytes from a representative tumor were stained with anti-Vβ6 antibodies (black) or with control antibody (FITC control, gray) followed by fluorescence-activated cell sorter analysis.

DISCUSSION

The TCL1 gene is involved in the development of T-PLL and T-CLL in humans (5). One T-PLL characteristic is the very late onset in life with a median age of 69 years (6). Leukemias carrying TCL1 rearrangements also frequently occur in AT patients at a younger age (18, 19). These patients have been reported to carry premalignant T cell expansions coincident with TCL1 activation (7). It is possible that the same scenario may be true for individuals developing T-PLL, where the premalignant clones may go undetected in asymptomatic individuals. Whereas these individuals may carry the T cell expansions for an extended period before the occurrence of secondary genetic events, AT patients may develop overt leukemia much earlier because of secondary mutations occurring more frequently. lck^pr-TCL1 transgenic mice, which express the transgene specifically in the T cell compartment, develop T cell leukemia after a long latency period closely resembling the leukemia in humans. The leukemias observed in transgenic mice are mature in phenotype, as in humans, but are restricted to the CD3⁺, CD4⁻CD8⁺ subset. This phenotype is at variance with that of human leukemias with TCL1 involvement, which are most often CD3⁺ and CD4⁺CD8⁻ (6). Although we cannot explain why the tumors developing in mice are exclusively CD8⁺, it cannot be attributed to the promoter specificity because lck-BCL2 transgenic mice develop leukemia of the CD4⁺CD8⁻ phenotype after a similarly long latency period (20). Young transgenic mice (<1 year old) already show an increase in the CD3⁺ and CD8⁺ populations, suggesting that TCL1 may influence the ratio of CD4 and CD8 lymphocytes in the spleen. These early expansions could not be considered clonal, because analysis of TCR gene rearrangement failed to detect clonal rearrangements. However, this indicates a selective advantage for the cells expressing TCL1 resembling the preleukemic clonal expansions in humans. In fact, whereas in human leukemia all T cells overexpressing TCL1 carry a specific rearrangement juxtaposing the TCRα to the TCL1 gene and hence are clonal, in lck-TCL1 transgenic mice TCL1 expression is forced in all T cells leading to the possible expansion of the entire T cell repertoire. Thus, we hypothesize that additional genetic changes are needed to cause the development of overt leukemia. In this model, TCL1 overexpression gives the cells a growth advantage increasing the probability for the occurrence of secondary genetic events responsible for the development of aggressive malignancy.

The development of TCL1 transgenic mice represents a very interesting model to study multiple genetic events involved in leukemogenesis, and we are in the process of establishing cell lines from the leukemic cells to study these events. For these studies secondary genetic events may also be identified by crossing TCL1 transgenic mice with Atm^−/− or p53^−/− mice.

ABBREVIATIONS

AT

ataxia telangiectasia

ATL

adult T cell leukemia

TCL1

T cell leukemia-1

T-CLL

T chronic lymphocytic leukemia

T-PLL

T prolymphocytic leukemia

TCR

T cell receptor

WBC

white blood cell