The murine Tcl1 oncogene: embryonic and lymphoid cell ... - Nature

Oncogene (1997) 15, 919 ± 926  1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00

The murine Tcl1 oncogene: embryonic and lymphoid cell expression Maria Grazia Narducci1, Laura Virgilio2, Julie B Engiles2, Arthur M Buchberg2, Linda Billips3, Antonio Facchiano1, Carlo M Croce2, Giandomenico Russo1 and Jay L Rothstein2 1

Istituto Dermopatico dell' Immacolata-IRCCS, Via dei Monti di Creta 104, 00167 Rome; 2Departments of Microbiology /Immunology and Otolaryngology-Head and Neck Surgery, Kimmel Cancer Institute, Thomas Jeerson Medical College, 233 S. 10th Street, BLSB 909, Philadelphia, Pennsylvania 19107 USA; 3Howard Hughes Medical Institute, University of Alabama, Birmingham, Alabama 35294, USA

In human leukemias and lymphomas nonrandom chromosomal rearrangements cause changes in cell growth and/or survival in such a way as to promote malignancy. The detailed study of the biochemical and genetic pathways altered in human cancer requires the identi®cation or development of models to allow the study and manipulation of cancer gene function. Recently, the breakpoint gene TCL1, involved in chromosome translocations observed mostly in mature T-cell proliferations and chronic lymphocytic leukemias (CLL), was isolated and characterized, and showed to be part of a new gene family of proteins involved in these tumors. The murine Tcl1 gene, is similar in sequence to the murine and human MTCP1 gene also involved in T cell leukemias. The murine Tcl1 gene was shown to reside on mouse chromosome 12 in a region syntenic to human chromosome 14. Furthermore, we show that the murine Tcl1 gene is expressed early in mouse embryonic development and demonstrates expression in fetal hematopoietic organs as well as in immature T and B cells. Characterization of the murine Tcl1 gene will help in developing a mouse model of CLL and would provide the best opportunity to study and decipher the role of TCL1 in malignant transformation. Keywords: ATM; ataxia-telangiectasia; TCL1; chronic lymphocytic leukemia; MTCP1; T-PLL

Introduction The molecular basis of cancer is steadily being uncovered due to the wealth of investigations focusing on genes whose altered expression leads to abnormal cellular dierentiation and/or proliferation. Identification of the genes responsible has been possible due to the large number of chromosomal rearrangements that have been found in speci®c types of hematopoietic tumors (Isobe et al., 1985; Baer et al., 1987; Bertness et al., 1990; Cleary, 1991; Croce, 1991). These chromosomal rearrangements have provided cytogenetic landmarks for studies directed at identifying and cloning the genes involved in cancer. The characterized chromosomal translocations have been shown to cause tumorigenesis through the activation of cellular protooncogenes or through the formation of novel chimeric genes capable of transforming hematopoietic cells (Heim and Mitelman, 1987; Savage et al., 1988; Correspondence: JL Rothstein Received 7 March 1997; revised 30 April 1997; accepted 1 May 1997

Rowley et al., 1990; Croce, 1991; Cimino et al., 1991; Rowley, 1992). Analysis of the chromosomal alterations from tumors revealed that breakpoints were clustered at a de®ned loci suggesting that protooncogene(s) resided at or near these positions (Croce, 1991). Speci®c chromosomal rearrangements involving human chromosome 14q32 have been found in Tprolymphocytic leukemia (T-PLL) in 28% of adult T cell leukemias (ATL) which are associated with infection by human T-lymphotropic virus type I (HTLV-I) or chronic T cell leukemias in patients with ataxia telangiectasia (AT; Taylor, 1992; Savitsky et al., 1995). The most common rearrangement is an inversion involving q11 and q32 or reciprocal translocation between chromosomes 14q11 and 14q32 (Brito-Babapulle and Catovsky, 1991). The breakpoint region of these chromosomal alterations was cloned and a gene found adjacent to these altered regions was subsequently isolated, sequenced and named TCL1 (Virgilio et al., 1994). Interestingly, the only other gene with signi®cant amino acid or nucleotide sequence identity with TCL1 is a gene named mature T cell proliferation-1 (MTCP1; Stern et al., 1993) which showed 61% similarity at the amino acid level. MTCP1 is a recently cloned chromosomal translocation t(X;14)(q28;q11) breakpoint gene involved in a case of benign clonal proliferation in an AT patient (Stern et al., 1993). The TCL1 gene was shown to be expressed in T cell acute leukemias and high grade lymphomas carrying rearrangements at 14q32.1 (Virgilio et al., 1993, 1994). The gene is also expressed in pre B cells, in B cells expressing surface IgM, and in the related B cell tumors (Virgilio et al., 1994). Analysis of TCL1 mRNA in normal tissue revealed expression in pre-B cells and CD47CD87 fetal thymocytes (Virgilio et al., 1994), but not in poly(A)+ RNA derived from any of the adult organs tested. Moreover, cellular localization studies revealed that the Tcl1 protein was restricted to the microsomal fraction suggesting that TCL1 performs its function as an intracellular protein (Fu et al., 1994). Thus, the newly identi®ed TCL1 gene and the related MTCP1 gene, represent two members of a new lymphoid-speci®c gene family which show restricted cellular expression in developing lymphocytes and tumors. Further, these ®ndings strongly suggest that the TCL1 gene functions to regulate the development, dierentiation or survival of lymphoid cells. Although the exact function of TCL1 is not known, aberrant gene expression likely results in alterations in cell survival or growth rate in a subset of lymphoid cells harboring an inversion or translocation on chromosome 14. This may result in the observed T cell clonal expansion that precedes malignancy in AT

Embryonic and lymphoid expression of the murine Tcl1 gene MG Narducci et al

920

patients (Narducci et al., 1995). One hypothesis is that this proliferative event is accompanied by additional genetic changes leading to a more progressive malignancy. However, without an appropriate animal model it would be dicult to decipher the role of TCL1 in cellular transformation. Thus, to provide a better understanding of TCL1 function we sought to identify and map the murine Tcl1 gene and evaluate mRNA expression during mammalian embryogenesis. The isolation of the murine gene and the elucidation of these data is a prerequisite to the development of a mouse model of chronic lymphocytic leukemia.

Results Cloning of the murine Tcl1 gene Initial attempts to obtain a murine cDNA directly from a mouse embryonic library using the human TCL1 cDNA did not yield sequences corresponding to the murine Tcl1 gene. Consequently, a TCL1 cDNA probe was used to screen a mouse 129/SVJ genomic library to obtain murine genomic sequences. From this primary screening a clone was identi®ed that showed signi®cant sequence homology (*60%) to exon 1 of the human TCL1 gene (data not shown). From these and other exon sequences, PCR primers were designed to amplify the entire Tcl1 coding region from mouse embryonic RNAs. A full length 1.3 kb cDNA clone was isolated from a mouse embryonic stem (ES) cell

cDNA library. This sequence comprises the 5' untranslated (UTR), coding and the 3' UTR regions of the murine Tcl1 gene (Figure 1). Using these newly developed probes the entire genomic region of the Tcl1 locus was isolated from a mouse 129/SvJ lambda library. The genomic clones were ordered and partially sequenced to deduce the Tcl1 genomic structure (Figure 2a). The intron/exon organization was determined from identi®ed canonical splice site acceptor and donor sequences (Figure 2b) and through the cross hybridization of the murine cDNA clone. Homology within the new family Analysis of the nucleotide and deduced amino acid sequence of the murine Tcl1 gene shows that it shares 55% nucleotide and 50% amino acid sequence homology to the human TCL1 gene (Figure 3) and lower, but signi®cant, homology to the related murine and human MTCP1 genes (Figure 3). Analysis of the murine Tcl1 coding region revealed signi®cant nucleotide and amino acid similarity with the human Tcl1 gene as well as to the human and murine Mtcp1 genes (Figure 3). Comparison of the mouse and human TCL1 and Mtcp1 sequences also showed the presence of several high homology domains containing up to 87% amino acid similarity within the gene family (Figure 3, boxed regions). These data suggest that the Tcl1 and Mtcp1 genes are members of a novel lymphocyte-speci®c gene family that may share a

Figure 1 Nucleotide sequence and deduced amino acid sequence of the murine Tcl1 cDNA. Nucleotide and amino acid positions are numbers in the left and right hand margins. The nucleotide sequence was derived from the full length Tcl1 cDNA clone derived from the ES cell cDNA library and the partial (ORF only) cDNA clone isolated from 15.5 day murine embryo total RNA. Boxed sequences represent ATG start and TAA stop codons. Underline regions denote sequences involved in mRNA stability (Chen et al., 1994)


common primordial ancestor as indicated by CLUSTAL analysis of evolutionary conservation (Figure 4). Indeed this similarity may suggest even though the human and murine Tcl1 genes are not 490% identical, they may share a high amount of structural similarity. To investigate this notion, we evaluated the hydrophilicity for the Tcl1 protein. Hydrophilicity pro®le generally indicates the behavior of dierent regions of the proteins in solution. This pro®le is similar to that of the human Tcl1 gene previously described (Virgilio, 1994) and suggests that Tcl1 is a hydrophilic molecule, much like the human Tcl1 protein (Figure 5). Chromosomal localization of the murine Tcl1 gene To map the murine Tcl1 cDNA we have made use of an interspeci®c backcross panel that has been typed for many markers (Staats, 1981; Green, 1989; Dietrich et al., 1992; Nadeau et al., 1992). Restriction enzyme digestion of chromosomal DNA using KpnI demonstrated a clear polymorphic fragment (RFLP) between Mus musculus

Figure 2 (a) Restriction map of the genomic murine lambda phage lfXIT22 containing the murine Tcl1 gene derived from a 129SVJ library. Restriction enzyme recognition sites are indicated: X=XbaI, P=PstI, H=HindIII, E=EcoRI, B=BamHI, Hc=HincII (incomplete) S=SalI, N=NotI (the SalI and NotI sites pictured at the ends are from vector sequences). (b) Exon intron structure and sequence boundaries at 3' and 5' splicing signals (lowercase) are represented

and Mus spretus strains (not shown). Thus, DNA from AEJ and Spretus backcross progeny were compared by Southern analysis using a mouse Tcl1 probe. Results from the analysis of 195 backcross progeny yielded a concordant map location on mouse chromosome 12 in close proximity to the murine immunoglobulin locus and to the murine B94 TNF response gene, tnfb94 (Wolf et al., 1994). The syntenic region of human chromosome 14q32 is the telomeric region of murine chromosome 12 (proximal to the immunoglobulin locus). Thus, the murine Tcl1 clone maps to the region in the mouse genome that is syntenic to human chromosome 14q32 (Figure 6). These data, together with the high nucleotide and amino acid sequence homology, indicate that the isolated cDNA and genomic clones represent the authentic murine homologue of the human TCL1 gene. Expression of Tcl1 in embryonic tissues The partial homology of human TCL1 to human MTCP1 and the high homology of the murine Tcl1 with the human TCL1 suggest that it is part of a conserved gene family potentially important in lymphoid development. Consequently, we sought to investigate the pattern of TCL1 expression during hematopoiesis. Using a speci®c RT ± PCR assay, we analysed the expression of Tcl1 at several stages of murine development and in neonatal liver and adult hematopoietic organs. Figure 7 shows (and summarized in Tables 1 and 2) the result of agarose gel (Figure 4a) and Southern hybridizations (Figure 7b and c) of PCR products derived from murine mRNA. These results suggest that Tcl1 is expressed early in mouse embryogenesis with expression detected as early as 7.5 days (at the onset of blood development). Although expression of Tcl1 was detected in 15.5 day fetal liver, this product was only visible after Southern analysis suggesting a low level of mRNA expression at this stage. Tcl1 was detected in 15.5 day yolk sac, a tissue highly enriched for hematopoietic progenitors in the blood islands, suggesting that Tcl1 is expressed in primitive hematopoietic organs. To con®rm this we evaluated Tcl1 expression in embryonic thymi derived from 18.5 day embryos. We ®nd that Tcl1 mRNA is expressed in the prenatal mouse

Figure 3 Comparison of the murine Tcl1 deduced amino acid sequence with the human TCL1, murine Mtcp1 and human MTCP1 sequences. Light blue, white lettered, regions represent complete identity for all family members (conserved domains); light blue, black lettered, regions represent homology within the TCL1 and MTCP1 genes and yellow regions represent amino acid sequences conserved in 75% of the sequences. Boxed regions represent TCL1 high-homology domains conserved between mouse Tcl1 and human TCL1 (*80% similarity) and the dotted boxes represent the corresponding regions in the MTCP1 genes

921


922

a

Figure 4 CLUSTAL genetic evolutionary analysis of human (Hu) TCL1, human MTCP1, murine (Mu) Tcl1 and murine Mtcp1 sequences using the DNASTARTM program. Forks in the tree indicate likely branch points during evolution suggesting a common ancestor

b

c Figure 5 Hydrophilicity pro®le of the deduced Tcl1 amino acid sequence. The Tcl1 deduced amino acid sequence was analysed for hydrophilicity using the MacVectorTM program (Kodak Scienti®c Imaging Systems, New Haven, CT)

M

1

2

3

4

5

6

7

8

9 10 11

Figure 7 Tcl1 expression in murine embryonic RNA as measured by nested RT ± PCR. PCR reactions were performed on cDNA synthesized from RNA extracted from: (a) Lane 1, 15.5 day fetal liver, lane 2, 15.5 day yolk sac, lane 3, 15.5 day whole embryo, lane 4, 13.5 day fetal liver, lane 5, 13.5 day whole embryo, lane 6, 12.5 day whole embryo, lane 7, 11.5 day whole embryo, lane 8, 10.5 day whole embryo, lane 9, 7.5 day whole embryo, lane 10, empty well, lane 11, water (negative control). All Tcl1-speci®c products are 360 bp in length. (b) Represents the same gel as a blotted and hybridized with the Tcl1 cDNA probe. Lane M is fX174 digested DNA as molecular weight marker. (c) Represents the corresponding cDNA samples ampli®ed with a G3PDH control primers set (1035 bp product). Lane M is fX174 digested DNA as molecular weight marker

Table 1 Whole embryo expression of murine Tcl1 and MTCP1 genes RNA derived from embryo at days post coituma 9.5 10.5 11.5 12.5 13.5 15.5

Murine Gene

7.5

Tcl1 Mtcp1

+b NDc

+ ND

+ ±

+ ND

+ +

+ +

+ +

a Embryos were isolated and RNA extracted as described in Materials and methods and using standard protocols (Sambrook et al., 1989). b + Indicates positive PCR product as visualized on ethidium bromide stained gel and veri®ed by Southern hybridization (Sambrook et al., 1989). cND; not determined

Figure 6 Schematic of updated mouse chromosome 12. Results of RI mapping using a murine Tcl1 probe and DNAs derived from 195 AEJ x Spretus backcross progeny. Position of the murine Tcl1 gene is placed along the distal region of the chromosome in close proximity to the immunoglobulin locus. Numbers on left indicate approximate distance (in CM) from Igh locus. Tcl1 map distance is *6.4 cM from Igh and 5.4 cM from the chromosomal marker d12mit8. The tnfb94 gene is *3.5 cM from the Igh locus. No recombinations were identi®ed between Tcl1 and the d12mit28 marker (linked loci). Distances shown on schematic are approximate and Figure is not drawn to scale. This region of the murine chromosome is syntenic to the distal region of human chromosome 14 where the human TCL1 gene has been localized (Virgilio et al., 1993, 1994)

thymus (not shown). The expression of Tcl1 in adult tissue was examined to investigate the expression in adult hematopoietic organs. Tcl1 expression was found in adult thymus and spleen but not heart, liver, or kidney (Table 2). The human TCL1 gene is related to the T cell proliferation gene MTCP1 at the nucleotide, amino acid and in mRNA expression (Stern et al., 1993). This similarity is also seen with the murine homologue (Madani et al., 1996). Thus, to compare the expression patterns of the murine Tcl1 gene with the related murine Mtcp1 gene (Madani et al., 1996) we performed an RT ± PCR analysis from embryonic and adult tissues. Interestingly, both of these related genes demonstrated early expression in the embryos (as early as 10.5 days of development), although Tcl1 was expressed one day


923

Table 2 Embryonic and adult organ expression data for the murine Tcl1 and MTCP1 genes Murine Gene

13.5FLa

Tcl1 Mtcp1

14.5FL b

7 +

RNA derived from embryo at days post coitum 14.5YS 15.5YS Heart Kidney

15.5FL + 7

+ NDc

+ ND

+ +/7

7 7

7 +

Liver

Spleen

Thymus

7 +

+ +

+ 7

CD4+CD8+

CD8+CD4–

CD4+CD8–

CD4–CD8–

B220+αµhi

B220+αµlo

B220hiαµ–

FL; fetal liver, YS; yolk sac, Heart, Kidney, Liver, Spleen and thymus represent organs taken from 8 week old adult mice. b+ Indicates positive PCR product as visualized on ethidium bromide stained gel and veri®ed by Southern hybridization (Sambrook et al., 1989). cND; not done, see text for details

B220loαµ–

a

Figure 8 Expression of Tcl1 in ¯ow sorted bone marrow and fetal thymic cell populations. Tcl1 RT ± PCR products were resolved on a 2% agarose gel and transferred to nylon membrane. The membrane was hybridized with an internal Tcl1 speci®c probe and exposed to X-ray ®lm for 24 h. The PCR results are shown. Lanes are labeled with phenotype of ¯ow sorted cell populations. For example, lanes 1 ± 4 correspond to RNA derived from ¯ow sorted B lymphocytes which stained low (lo) or high (hi) with antibodies (a) speci®c for B220 and or surface IgM (m). Lanes 5 ± 8 represent T lymphocytes sorted with CD4 or CD8 speci®c antibodies and separated by the surface phenotype indicated

earlier than MTCP1. In contrast to Tcl1, we ®nd that MTCP1 is expressed in adult kidney and liver (Table 1). In addition, while Tcl1 is expressed in yolk sac, only faint bands are observed in 13.5 and 15.5 day fetal liver; revealed only after Southern analysis of the PCR gel (Figure 7b). This reduced expression coincides with the emigration of thymocytes out of the liver and into the fetal thymus. We ®nd that MTCP1 is also expressed in 13.5 day fetal liver but not 15.5 day fetal liver (not shown). The summary of expression data related to murine Tcl1 and murine MTCP1 are shown in Tables 1 and 2. Expression of Tcl1 in fetal B and T lymphocytes The expression of Tcl1 in hematopoietic organs in the developing mouse embryo suggested that Tcl1 may be restricted to hematopoietic cells. Since Tcl1 is dysregulated in lymphocytes we sought to determine if Tcl1 was physiologically expressed in isolated lymphocytes from embryonic thymus. Fetal B and T lymphocytes were isolated from 12 day fetal bone marrow and thymus as described (Virgilio et al., 1994). B lymphocytes were identi®ed by ¯ow cytometry using antibodies to surface IgM and the B220 marker expressed on immature B cells. T lymphocytes were isolated on the basis of their CD4 and CD8 positively. RT ± PCR was performed from cDNA derived from these sorted populations. The results of a representative experiment show that Tcl1 mRNA can be detected in B cells expressing low to high levels of the B220 marker and negative for surface IgM(m) (Figure 8). The

signal appears most intense, however, in the B220+ population expressing low levels of m suggesting higher expression levels of the Tcl1 mRNA. The persistence of expression at a low level in B220+mhi cells may be due to contaminating B220+mlo cells or to authentic low level expression in the B220+mhi population. This pattern of expression is similar to that observed for human TCL1 in fetal B-cells (Virgilio et al., 1994). The expression of murine TCL1 was also evaluated in the in fetal T cell populations. cDNAs were prepared from ¯ow sorted thymocytes and semi-nested PCR was performed using Tcl1 speci®c primers. The results of these analysis are shown in Figure 8 indicating that Tcl1 is expressed in CD4+CD8+ (double positive) and CD47CD87 (double negative) but not CD4+ CD87 cells. A faint, but discernible, band was observed in the lane containing CD8+CD47 cells suggesting low level expression in this population or possible residual double positive cells contaminating this ¯ow sorted population. Discussion In this report we have presented the cloning, mapping and the embryonic expression pattern of the murine homologue of the T-cell leukemia/lymphoma 1 gene TCL1. This gene is known to be involved in the pathogenesis of human mature T-cell proliferations, such as T-PLL and in similar leukemias originating with high frequency in patients with AT (Taylor, 1982; Baer et al., 1987; Croce et al., 1985; Heim and Mitelman, 1987; Russo et al., 1988; Bertness et al., 1990; Wei et al., 1990; Brunning, 1991; Croce, 1991). This gene is also overexpressed in Adult T-cell leukemias, a T-lymphoproliferation that shares many similarities to T-PLL and associated with endemic infection of the HTLV-I virus (Narducci et al., unpublished results). Further this gene is transcriptionally regulated in the lymphoid compartment, suggesting that TCL1 could play a signi®cant role in the development of lymphocytes much like that observed in other leukemia-causing genes which lead to altered growth or dierentiation of lymphocytes (see Croce, 1991 for review). Since sequence of this gene does not reveal known structural similarities with other proteins that could provide insight into the function of the Tcl1 protein, it is important to develop animal models to evaluate the eects of Tcl1 perturbation. The genomic structure of the murine and human genes demonstrates the similarity of the two genes. By comparison, the intron/exon structure (boundaries) are identical to those in the human gene and the extent of 5' and 3' UTR regions are the same. This is particularly important since TCL1 and MTCP sequences do not show signi®cant similarity to other sequences in PIR


924

database. This, and the information related to the structural analyses, strongly support the hypothesis that TCL1 and MTCP1 sequences belong to a newly identi®ed gene family. The overall high similarity between TCL1 and MTCP1 require only minimal gap insertion. The analysis revealed that both TCL1 and MTCP are proteins rich in tryptophans. Interestingly, almost all tryptophan residues (i.e. four out of ®ve) occur in matching positions. Further, the similar hydropathy plots indicate that dierences in the sequence are almost silent as the exposition to the aqueous solvent is concerned. The expression of murine Tcl1 was not evident in 18 day fetal liver (not shown), a time when developing lymphocytes are emigrating out of the liver and into the fetal thymus. Thus, the TCL1 gene may contain promoter/enhancer elements exclusive to lymphoid cells throughout development. Indeed, tight transcriptional regulation of Tcl1 is likely required for normal Tcl1 function since even increased expression in immature lymphocytes results in transformation. These transformed cells would then be hypothesized to accumulate, predisposing them to a second genetic hit, causing the leukemia observed in CLL. Analysis of leukemia in humans supports this hypothesis since TCL1 dysregulation via translocation precedes transformation (Virgilio et al., 1994). This restricted expression may also be important for normal lymphocyte function during hematopoietic development. Consistent with this is the ®nding of Tcl1 expression at low levels in the spleen and thymus, but no other organs of adult mice (Table 2). Expression of murine Tcl1 also follows a pattern similar to that observed with human TCL1 (Virgilio et al., 1994). TCL1 expression in double negative (CD47CD87) but not in CD8+ BALB/c thymocytes is consistent with the pattern of Tcl1 expression observed in human cells. However it must be noted that the expression in double positive cells has been observed in murine but not in human fetal double positive T-cells. At the moment we do not know the signi®cance of this discrepancy. Identical patterns of Tcl1 expression were observed in B-cell from mice and those from previous reports in humans (Virgilio et al., 1994). Thus, murine Tcl1 expression parallels that observed in humans and represents expression in lymphoid progenitors in the developing embryo and young mice. These data also suggest that the expression of Tcl1 is tightly regulated in lymphoid/hematopoietic precursors by the Tcl1 gene promoter. The map location of the murine Tcl1 gene placed it in close proximity with the B94 gene which is a tumor necrosis factor-a inducible protein expressed in fetal liver, spleen and thymus (Wolf et al., 1994). In addition, this gene product was found to be exclusively expressed in lymphopoietic organs of the adult (Wolf et al., 1994). Although the relationship of B94 and Tcl1 is not clear, the close proximity of these genes in the genome, like immunoglobulin, T cell receptor and HOX genes, may be to control speci®c developmental pathways during mammalian embryonic development. Interestingly, mice de®cient in the ataxia telangiectasia or ATM gene (ATM7/7) develop T cell leukemias with karyotypic abnormalities involving chromosome

12 (Barlow et al., 1996). These data suggest that leukemias may be developing at a high rate in these ATM7/7 mice due to the transcriptional activation of Tcl1, much like that observed in human CLL. Future analysis of the leukemias in ATM7/7 mice as well as other in vivo models of Tcl1 function (e.g. Tcl1 transgenics and Tcl17/7 mice) will assist in studying the mechanistic relationship between ATM and Tcl1 gene products. Furthermore, the study of Tcl1 gene expression may be important for the understanding of the physiology of early T and B cell development. Indeed, the isolation of the Tcl1 gene will not only be useful for studying the maturation of the mammalian immune system, but also for the construction of a murine model of chronic lymphocytic leukemia. This model may someday enable the testing and analysis of better diagnostics and treatments for patients aicted with chronic leukemia.

Materials and methods Sample tissue, RNA and DNA isolation, Northern analysis and RT ± PCR RNA derived from mouse embryos at sequential stages during development were collected from 7.5 d.p.c. through neonate at daily intervals, extracted and prepared as described (Nagasaka et al., 1983; Freeman, 1990; Rothstein et al., 1993). After RNA isolation, ®rst-strand cDNA synthesis was performed as described for the construction of cDNA libraries (Rothstein et al., 1993). The synthesized single-stranded cDNA was ampli®ed in a standard PCR reaction (Innis et al., 1990; Ausubel et al., 1991) containing 2 m M dNTPs, 2.5 units Taq Polymerase (Perkin-Elmer Cetus), and 2 pmole each of the 3' and 5' gene-speci®c oligonucleotide primers, in a total volume of 50 ml in 16PCR reaction buer (supplied by manufacturer; Perkin-Elmer Cetus). Following 30 cycles of denaturation at 948C for 30 s, 50 ± 608C annealing for 30 s and elongation at 728C for 1 ± 2 min, the PCR products were electrophoresed in a 1.5% agarose gel and, in some cases, blotted onto nylon membranes and hybridized with 32P dCTP-labeled gene-speci®c probes as described (Innis et al., 1990; Ausubel et al., 1991). The embryonic RNA was adjusted so that each PCR reaction contained the same amount of template for RT ± PCR (Innis et al., 1990). Lymphocyte isolation and analysis The expression of the murine TCL1 gene was evaluated at selected stages of normal B and T cell dierentiation. Murine fetal bone marrow B cell subpopulations (1.56107 cells) were separated by two color immuno¯uorescence cell sorting as described (Virgilio et al., 1994). cDNA from selected lymphocyte populations were prepared and PCR was carried out with mouse Tcl1 exon 1 speci®c primers as described below and previously (Virgilio et al., 1994). Primers and sequence analysis For the analysis of lymphocyte populations Tcl1 exon 1 speci®c primers were used (Ex1 5' primer: CAG/GGC/AGA/ GAC/ACC/TGC/ACA and Ex1 3' primer: TTA/ATG/ TGG/ACA/GGA/TCT/CCA) to yield a fragment of 300 bp. For all other expression analysis by RT ± PCR, the open reading frame primers (orfp) were used 5' orfp1 primer ATG/GCT/ACC/CAG/CGG/GCA/CAC and 3' orfp2 pri-


mer TTA/TTC/ATC/GTT/GGA/CTC/CGA) which yields a PCR product 350 bp in length. Sequences of human TCL1 and human MTCP1 were taken from PIR database (accession #'s Hstcl1sp1 and Hsmtcplf respectively). Mouse MTCP1 sequence (accession # Mmu32332) was derived from a previously published report (Madani et al., 1996). Multiple alignment was performed by means of the CLUSTAL routine from the software package PCGENE (IntelliGenetics Inc.), according to Higgins and Sharp (1989). The multiple alignment obtained was carried out with the following setting parameters: k-tup 2, gap penalty 3, ®ltering level 2, open gap cost 15, unit gap cost 10. The hydropathy (hydrophilicity) pro®le of Tcl1 sequence was performed according to Kyte and Doolittle, by using a shifting window of seven residues (Kyte and Doolittle, 1982). Isolation of murine and genomic cDNAs To isolate the murine TCL1 gene, a genomic library derived from the murine 129/SVJ library (Stratagene, Inc.) was screened with the human TCL1 cDNA at low stringency (26SSC) as described (Sambrook et al., 1989). Positively hybridizing clones were selected and DNA extracted using the lambda WizardTM extraction kit (Promega, Corp.). Isolated lambda DNA was digested with appropriate restriction enzymes and overlapping clones oriented as described (Sambrook et al., 1989). Fragments containing Tcl1 exons were identi®ed by cross hybridization and subcloned into the pBluescript-II SK+TM (Stratagene) plasmid for sequencing (Innis et al., 1990; Ausubel et al., 1991). Embryonic RNA isolation The procedures for embryo isolation and RNA extraction are as outlined in previous reports (Rothstein et al.,

1993). Brie¯y, 6- to 8-week old male and female (B6D2)F1 mice (Jackson Laboratories, Bar Harbor, ME) were naturally mated, timed and embryos isolated at speci®ed times after observation of a vaginal plug (day of vaginal plug was considered equal to day 0.5 of embryonic development). Isolated embryos were ®rst placed in phosphate buered saline (PBS) and then transferred into embryo lysis buer (100 mM NaCl, 50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 0.5% sodium dodecyl sulfate, 5 mg E. coli tRNA, 5 mg/ml proteinase K), homogenized and incubated at 378C for 30 ± 60 min. Embryo lysates were then phenol-chloroform extracted twice and chloroform extracted once, followed by ethanol precipitation as previously described (Sambrook et al., 1989; Rothstein et al., 1992, 1993). RNA was quanti®ed by optical density and concentrations con®rmed by agarose gel electrophoresis (Sambrook et al., 1989).

Abbreviations ATM, ataxia telangiectasia; CLL, chronic lymphocytic leukemia; MTCP1, mature T cell proliferation-1; T-PLL, T cell prolymphocytic leukemia; TCL1, T cell locus-1.

Acknowledgements This work was supported by the Public Health Service grant CA-21124 (to JLR) from the National Cancer Institute, and by the Outstanding Investigator Grant CA39869 from the National Institutes of Health (to CMC) and by the Human Frontier Grant 350/96 (GR). The Kimmel Cancer Center shared research facilities were supported by a Cancer Center Grant CA56336 from the National Institutes of Health.

References Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA and Struhl K. (1991).Current Protocols in Molecular Biology. John Wiley and Sons: New York. Baer R, Hepple A, Taylor AMR, Rabbitts PH, Bonllier B and Rabbitts TH. (1987). Proc. Natl. Acad. Sci. USA, 84, 9069 ± 9073. Barlow C, Hirotsune S, Paylor R, Liyanage M, Eckhaus M, Collins F, Shiloh Y, Crawley JN, Ried T, Tagle D and Wynshaw-Boris A. (1996). Cell, 86, 159 ± 171. Bertness VL, Felix CA, McBride WD, Morgan R, Smith SD, Sandberg AA and Kirsch LR. (1990). Cancer Genet. Cytogenet., 44, 47 ± 54. Brito-Babapulle V and Catovsky D. (1991). Cancer Genet. Cytogenet., 55, 1 ± 9. Brunning RD. (1991). Blood, 78, 3111 ± 3113. Calzone FJ, Britten RS and Davidson EH. (1987). Methods in Enzymology, 152, 611 ± 632. Chen C-YA, Chen T-M and Shyu A-B. (1994). Mol. Cell. Biol., 14, 416 ± 426. Cimino G, Moir DT, Canaani O, Williams K, Crist WM, Katzav S, Cannizzaro L, Lange B, Nowell PC, Croce CM and Canaani E. (1991). Cancer Res., 51, 6712 ± 6714. Cleary ML. (1991). Cell, 66, 619 ± 622. Croce CM. (1991). Origins of Human Cancer: A Comprehensive Review. Cold Spring Harbor Laboratory Press, New York. pp. 527 ± 542. Croce CM, Isobe M, Palumbo A, Puck J, Erickson J and Rovera G. (1985). Science, 227, 1044 ± 1047.

Dietrich W, Katz H, Lincoln SE, Shin HS, Friedman J, Dracopoli NC and Lander ES. (1992). Genetics, 131, 423 ± 447. Fu T, Virgilio L, Narducci MG, Facchiano A, Russo G and Croce C. (1994). Can. Res., 54, 6297 ± 6301. Freeman SJ. (1990). Postimplantation Mammalian Embryos, A practical Approach. Copp AJ and Cockroft DL (eds). Oxford University Press. pp. 249 ± 265. Green MC. (1989). Genetic Variants and Strains of the Laboratory Mouse. Lyon MF and Searle AG (eds). Oxford University Press: Great Britain. pp. 12 ± 403. Heim S and Mitelman F. (1987). Cancer Cytogenetics. Alan R. Liss: New York. Higgins DG and Sharp PM. (1989). Comp. Appl. Biol., 5, 151 ± 153.21. Innis MA, Gelfand DH, Sninsky JJ and White TJ. (ed). (1990). PCR Protocols: A Guide to Methods and Applications. Academic Press, New York. Isobe M, Erikson J, Emanual BS, Nowell PC and Croce CM. (1985). Science, 228, 580 ± 582. Kyte J and Doolittle RF. (1982). J. Mol. Biol., 157, 105 ± 132. Madani A, Choukroun V, Soulier J, Cacheux V, Claisse J-F, Valensi F, Daliphard S, Cazin B, Levy V, Leblond V, Daniel M-T, Sigaux F and Stern M-H. (1996). Blood, 87, 1923 ± 1927. Mengle-Gaw L, Willard HF, Smith CIE, Hammerstrom L, Fisher P, Sherrington P, Lucas G, Thompson PW, Baer R and Rabbitts TH. (1987). EMBO J., 6, 2273 ± 2280.

925


926

Nadeau JH, Davisson MT, Doolittle DP, Grant P, Hillyard AL, Kosowsky MR and Roderick TH. (1992). Mammalian Genome 3: 480 ± 536. Nagasaka M, Macda H, Chen H-L, Kita R, Makuelin O, Misu H, Matsuo T and Sugigama T. (1983). Blood, 61, 1174 ± 1181. Narducci MG, Virgilio L, Isobe M, Stoppacciaro A, Elli R, Fiorilli M, Corbonari M, Autorelli A, Chessa L, Croce CM and Russo G. (1995). Blood, 86, 2358 ± 2364. Rothstein JL, Johnson D, Jessee J, Skowronski J, DeLoia JA, Solter D and Knowles BB. (1993). Meth. Enzymol., 225, 587 ± 610. Rothstein JL, Johnson D, DeLoia JA, Skowronski J, Solter D and Knowles BB. (1992). Genes and Dev., 6, 1190 ± 1201. Rowley JD, Diaz MD, Espinosa III R, Patel VD, VanMelle E, Ziemin S, Taillon-Miller P, Lichter P, Evans GA, Kersey JH, Ward DC, Domer PH and LeBeau MH. (1990). Proc. Natl. Acad. Sci. USA, 87, 9358 ± 9362. Rowley JD. (1992). Genes, Chromosomes and Cancer, 5, 264 ± 266. Russo G, Isobe M, Pegoraro L, Finan J, Nowell PC and Croce CM. (1988). Cell, 53, 137 ± 144. Russo G, Isobe M, Gatti R, Finan J, Batuman O, Huebner K, Nowell PC and Croce CM. (1989). Proc. Natl. Acad. Sci. USA, 86, 602 ± 606. Sambrook J, Fritsch EF and Maniatis T (eds). (1989). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, New York. Savage PD, Jones KC, Silver J, Geurts Von Kessel AHM, Gonzales-Sarmiento R, Palm L, Hanson CA and Kersey JH. (1988). Cytogenet. Cell Genet., 49, 289 ± 292.

Savitsky K, Bar-Shira A, Gilad S, Rotman G, Ziv Y, Vanagaite L, Tagle DA, Smith S, Uziel T, Sfez S, Ashkenazi M, Pecker I, Frydman M, Harnik R, Patanjali SR, Simmons A, Chessa L, Sanal O, Lavin MF, Jaspers NGJ, Taylor MR, Arlett CF, Miki T, Weissman SM, Lovett M, Collin FS and Shiloh Y. (1995). Science, 256, 1749 ± 1753. Staats J. (1981). The Mouse in Biomedical Research. Foster HL, Small JD and Fox JG (eds). Academic Press: New York, pp. 177 ± 213. Stern MH, Soulier J, Rosenzwajg M, Nakahara K, CankiKlain N, Aurias A, Sigaux F and Kirsch IR. (1993). Oncogene, 8, 2475 ± 2483. Taylor AM. (1992). Ataxia Telangiectasia, A Cellular and Molecular Link between Cancer Neuropathy and Immunode®ciency. Bridges BA and Harnden DG (eds). John Wiley and Sons, Inc: Chichester: United Kingdom; pp. 53 ± 81. Virgilio L, Isobe M, Narducci MG, Cartenuto P, Camerini B, Kurosawa N, Rushdi A, Croce CM and Russo G. (1993). Proc. Natl. Acad. Sci. USA, 90, 925 ± 929. Virgilio L, Narducci MG, Isobe M, Billips LG, Cooper MD, Croce CM and Russo G. (1994). Proc. Natl. Acad. Sci. USA, 91, 12530 ± 12534. Wei S, Rocchi M, Archidiacon N, Sacchi N, Romeo G and Gatti RA. (1990). Cancer Genet. Cytogenet., 46, 1 ± 8. Wolf FW, Sarma V, Seldin M, Drake S, Suchard SJ, Shao H, O'Shea KS and Dixit VM. (1994). J. Biol. Chem., 269, 3633 ± 3640.