Peptide containing the BCR oligomerization domain - Nature

Oncogene (1998) 17, 825 ± 833  1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00 http://www.stockton-press.co.uk/onc

Peptide containing the BCR oligomerization domain (AA 1-160) reverses the transformed phenotype of p210bcr ± abl positive 32D myeloid leukemia cells Xiang-Yang D Guo1, Jean-Marie Cuillerot2, Tao Wang1, Yun Wu2, Ralph Arlinghaus2, David Claxton2, Carlos Bachier5, Joel Greenberger3, Ilyas Colombowala4 and Albert B Deisseroth1 1

Medical Oncology Section, Department of Internal Medicine and the Gene Therapy Program of the Yale Cancer Center, Yale University School of Medicine, New Haven, Connecticut 06520-8032; 2UT MD Anderson Cancer Center, Houston, Texas; 3 University of Pittsburg Medical Center; 4Dartmouth College, Hanover, New Hampshire; 5South Texas Cancer Institute, San Antonio, Texas, USA

We ®rst showed that the introduction of a bcr ± abl transcription unit into the 32D murine myeloid cell line (P210bcrabl32D) converts this cell line from an IL3 dependent cell line to an IL3 growth independent cell line. We next cloned a fragment of the bcr ± abl cDNA, which codes for the bcr oligomerization domain and neighboring regions. To test for a transformation inhibitory eect of this oligomerization inhibitory peptide transcription unit on the p210bcr ± abl mediated IL3 independent growth of the P210bcrabl32D cell line, we transiently co-electroporated into the growth factor dependent 32D cells, mixtures of plasmids which contained varying ratios of the plasmid expression vectors for the bcr oligomerization inhibitory peptide along with a smaller amount of the plasmid expression vector for the full length p210bcr ± abl. (The P210bcr ± abl protein converts the 32D from a growth factor dependent into a growth factor independent cell line.) We then showed that the oligomerization domain containing fragment from the bcr and bcr ± abl proteins, can be used to inhibit the IL3 independent growth of p210bcr ± abl positive 32D cells. These studies may be of eventual interest for those investigators whose goal is to design molecular therapeutic approaches to CML based on the use of peptidomimetic chemical functionalities, which mimic the structure and the inhibitory binding properties of the oligomerization domain containing fragment so as to inhibit the transforming function of the P210bcr ± abl oncoprotein. Keywords: p210bcr ± abl BCR; CML

oncoprotein;

tetramerization;

Introduction Transforming function of bcr ± abl bcr ± abl is an oncogene derived from a reciprocal chromosomal translocation between chromosomes 9 and 22 (Ben-Neriah et al., 1986; Heisterkamp et al., 1985), which de®nes Ph1 chronic myelogenous leukemia (CML). The p210bcl ± abl chimeric protein generated by this chromosomal translocation is associated with Correspondence: AB Deisseroth Received 16 January 1998; revised 23 March 1998; accepted 23 March 1998

deregulated tyrosine speci®c protein kinase activity (Konopka et al., 1984; Kloetzer et al., 1985), and major changes in the phenotype of the myeloid hematopoietic cell: (1) growth factor independent growth in previously growth factor dependent cells; (2) anchorage independent growth in previously anchorage dependent cells; and (3) genetic instability (Daley and Baltimore, 1988; Renshaw et al., 1995; Lugo and Witte, 1989). Expression of the p210bcr ± abl protein along with the myc protein results in transformation of the Rat-1 ®broblast cell line (Lugo et al., 1990). Transduction of normal marrow cells by retroviral vectors containing a bcr ± abl transcription unit can produce a CML-like syndrome when transplanted into lethally irradiated recipient mice, although the expression of a leukemic myeloid phenotype may depend on the genetic background of the inbred strain of mice used for the experiment (Daley et al., 1990; Gishizky et al., 1993; Elefanty et al., 1992). Overexpression of antisense oligonucleotides, which block the translation of the p210bcr ± abl protein, or result in a reduction of bcr ± abl mRNA levels, inhibits the growth of CML cells (Szczylik et al., 1991; Bedi et al., 1994) and sensitizes even blast crisis CML cells to chemotherapy. Cells expressing a temperature sensitive mutant of p210bcr ± abl show growth factor dependent growth at the non-permissive temperature, whereas the cells revert to growth factor independent growth at the permissive temperature (Carlesso et al., 1994). Functions of c-abl which promote and oppose the transformation of the myeloid cell The carboxyl terminal half of the p210bcr ± abl is derived from c-abl tyrosine speci®c protein kinase, which is tightly regulated in a cell cycle speci®c manner, and is also found in the cytoplasm bound to PI 3-kinase (Van Etten et al., 1989; Pendergast et al., 1991; Wetzler et al., 1993; Jain et al., 1997; Skorski et al., 1997). Overexpression of c-abl can induce cell cycle progression inhibition, and apoptosis, especially following irradiation or chemotherapy (Raitano et al., 1995; Kharbanda et al., 1995; Yuan et al., 1996). In contrast, a transforming function of c-abl can be unmasked by deletion of the Src homology region 3 (SH3) domain or by deletion of the carboxyl terminus of the abl protein including the DNA binding domain, which also activates the transforming activity of the protein

BCR fragment suppresses p210bcr ± abl function X-YD Guo et al

826

(Mayer and Baltimore, 1994; Goga et al., 1993; Wang, 1988; Shore et al., 1990; Jackson and Baltimore, 1989). Alternately the abl kinase and transforming activity can be activated by fusing the c-abl cDNA with the retroviral GAG cDNA or with the 5' half of the bcr cDNA (Rosenberg and Witte, 1988; Muller et al., 1991). This structural change converts a protein which promotes apoptosis, and opposes cell cycle progression, into a transforming protein which promotes cell cycle progression and rescues cells from apoptosis. The marked elevation of the abl kinase activity upon which the transforming activity of the P210bcr±abl protein depends, may be due to structural changes or the change in the localization of the c-abl fusion protein from the intranuclear location of the P145c ± abl to the extranuclear compartment of the cell. The P210bcr±abl is diusely present in the cytoplasm (Wetzler et al., 1993; Muller et al., 1991; McWhirter and Wang, 1991). Functional regions of c-abl The abl protein contains regions which bear homology to the Src oncogene: SH1, which is a tyrosine speci®c protein kinase, SH2, which is a phosphoprotein binding region (which in the case of the P210bcr±abl may interact with other proteins such as PI3-kinase), and SH3, which may negatively auto-regulate the activity of the abl tyrosine speci®c protein kinase, and which binds other proteins such as PI3-kinase in the case of P145c-abl. The c-abl also contains a DNA binding region, and a F-actin binding domain in the carboxyl terminus (Ren and Baltimore, 1994; Kipreos and Wang, 1992; Van Etten et al., 1994; McWhirter et al., 1993a; Welch and Wang, 1993). A number of cell cycle regulatory proteins (CRKL, GRB2, Abi-1, Abi2, the retinoblastoma tumor suppressor protein, p53, Cdk2 and Jun kinase have all been shown to interact with the kinase of the abl protein (Raitano et al., 1995; Kharbanda et al., 1995; Welch and Wang, 1993; Dai and Prendergast, 1995; Shi et al., 1995; Hoeve et al., 1994; Schwartzberg et al., 1991; Maru et al., 1991; Pendergast et al., 1993). In CML cells, the CRKL has been found to be phosphorylated by p210bcr ± abl and to bind to the bcr ± abl oncoprotein (Hoeve et al., 1994). Targeted disruption of the carboxyl terminus of the c-abl protein in mouse cells results in a loss of function of the c-abl protein indicating that this portion of the abl protein is essential for its normal function (Schwarzberg et al., 1991). Deletion of the carboxy-terminal abl actin binding motif from the P210bcr±abl protein can result in loss of bcr ± abl transforming activity (McWhirter et al., 1993a). Functional regions of the bcr region At the bcr N-terminal portion of the chimeric p210bcr ± abl protein, there is a serine/threonine kinase catalytic activity (Maru and Witte, 1991). Multiple tyrosine residues are also found in the bcr portion of bcr ± abl oncoprotein which have been shown to be potential targets of the abl tyrosine kinase activated in p210bcr ± abl (Pendergast et al., 1993; Puil et al., 1994; Lu et al., 1993). Tyrosine177 of the bcr protein at the amino terminal end of the p210bcr ± abl protein has been shown to be a substrate for the p210bcr ± abl tyrosine kinase.

Tyrosine177 may bind to the SH2 domain of the Grb2 adaptor protein and eventually lead to activation of the ras pathway (Pendergast et al., 1993; Puil et al., 1994). However the signi®cance of tyrosine177 phosphorylation in hematopoietic cells remains unresolved (Cortez et al., 1995). Recently phosphorylation of tyrosine360 has been shown to be involved in regulating bcr serine/threonine kinase activity and binding of p160bcr to p210bcr ± abl (Liu et al., 1996). However, little is known with regard to the functional consequences of the bcr serine/ threonine kinase in the bcr ± abl chimeric protein except that ®rst exon of the bcr gene which contains the coding sequences for the oligomerization domain are important in the transformation activity of the bcr ± abl protein (Muller et al., 1991; McWhirter et al., 1993b). Oligomerization domain of the bcr An oligomerization domain has been characterized in the extreme N-terminus of the p160bcr protein, (McWhirter et al., 1993b). Point mutations or insertional mutations which destroy the tetramerization of bcr or bcr ± abl have been observed to lead to inactivation of the p210bcr ± abl transforming activity. It has recently been reported that another form of abl oncoprotein may exist, which fuses the abl in frame with the Tel protein, a member of Ets family (Papadoponlos et al., 1995). The portion derived from the Tel protein encompasses a putative HLH region of the protein which may function in a similar manner to bcr, in that it can act to oligomerize the abl protein (McWhirter et al., 1993b; Muvre et al., 1989). Another piece of evidence which supports the involvement of oligomerization in the function of bcr ± abl comes from a series of experiments in which the hormone-binding domain of the estrogen receptor was fused to the Abl kinase. This results in the activation of the hybrid protein in the presence of estrogen (Jackson et al., 1993). Dominant negative inhibitory eects of peptides which contain oligomerization domains A dominant negative action has been observed with the clinical manifestations of generalized resistance to thyroid hormone when one of their thyroid hormone receptor alleles is mutated (Usala et al., 1988; Zavacki et al., 1993). Similarly, the retinoic acid receptor undergoes dimerization when bound to the ligand. Targeted overexpression of this dominant mutant form of this receptor in mouse epidermis produces the suppression of epidermal maturation indicating the speci®c dominant negative blockade of retinoic acid signaling in the transgenic mice (Mitinori et al., 1995). It has therefore been proposed that overexpression of the bcr ± abl oligomerization domain without abl kinase domain may produce a dominant negative inhibitory eect by disrupting the bcr ± abl homotetramerization of the bcr ± abl protein and thus its transforming activity. In this paper we report that the over expression of a transcription unit, which encodes the bcr N-terminal oligomerization domain, leads to inhibition of the p210bcr ± abl dependent growth factor and anchorage independent growth in a colony formation assay of p210bcr ± abl transformed cells on methyl cellulose.


Results Peptide transcription units which block P210bcrabl The N-terminal ®rst 63 amino acids region of the bcr protein encodes a coiled-coil tetramerization domain which has been reported to be involved in the binding the p210bcr ± abl to F-actin monomers, and to be essential for the transforming activity of the P210bcr ± abl oncoprotein (McWhirter et al., 1993). In addition, the bcr aminoterminal domain of the P210bcr±abl oncoprotein contains a tyrosine177 GRB2 binding site, a SH2 binding domain, and multiple tyrosines which are potential targets of the abl tyrosine kinase. Among these is tyrosine360, the phosphorylation of which is involved in regulating the bcr serine/threonine kinase activity and binding of p160bcr. In order to determine if the expression of the bcr ®rst 160 amino acids of bcr, including the tetramerization domain, is capable of inhibiting the transforming capacity of bcr ± abl, a cDNA fragment which encodes a fragment of bcr containing the ®rst 160 amino acids, including the bcr oligomerization domain, was subcloned into the constitutive expression vector pcDNA3. In this vector, the transcription unit is driven by a CMV promotor (this vector is designated pcDNABcr1-160). In addition, we constructed a similar series of plasmids containing the same transcription units, but this time driven by a tetracycline inducible expression vector (this vector is designated pUHD103). This vector, which contains the ®rst 160 amino acids of the bcr, is designated pUHD-Bcr1-160. The pUHD-Bcr1-160 fragment of bcr does not contain the Tyrosine177 Grb2 binding site or the serine/ threonine kinase activity of bcr, which have been shown to be important in the P210bcrabl mediated transformation of ®broblasts (Pendergast et al., 1993). In contrast, the plasmid pUHD-BCR1-412/Abl32 contains all of the ®rst 412 amino acids of the bcr protein. The protein fragment represented in this vector would include all of the important bcr domains. The cDNA in the transcription unit in the latter vectors was also fused in frame with a cDNA for a protein fragment which contained the last 33 amino acids of the abl protein. The terminal abl sequences were added to stabilize the protein. These transcription units were then introduced into the pcDNA3 vector (this vector is designated pcDNA-Bcrl-412-ABL32) in which the transcription

unit is driven by the CMV promoter, or by the tetracycline enhancer (this vector is designated pUHDBcr1-412/ABL32). The cDNA fragment comprising the codons coding for the ®rst 160 amino acids of bcr contains the tyrosine177, serine/threonine kinase activity and SH2 binding motif (bcr amino acids 192-412) in addition to the tetramerization domain (1 ± 63). As a control for the above vectors, we deleted that portion of the bcr cDNA coding for the N-terminal ®rst 160 amino acids, from the construct coding for amino acids 1 ± 412 and for the last 32 amino acids of c-abl. This plasmid, which lacks the oligomerization domain, is designated pUHD-Bcr161-412/ABL32. Another control transcription unit, pUHD-Bcr1-160/ABL32, encodes the bcr tetramerization motif fused in frame with the abl carboxyl terminal 32 amino acids (see Table 1). These fragments were subsequently cloned into the pcDNA3 (this vector is designated pcDNA-Bcr1-160/ ABL32), in which the transcription unit was driven by the CMV promoter, or into plasmids pUHD10-3 (this vector is designated pUHD-Bcr1-160/ABL32), in which the transcription unit was driven by the tetracycline regulated promoter. All of these constructs are schematically summarized in Figure 1 and Table 1. Each one of them was studied in the transient functional expression system in the K562 or 32D myeloid leukemia cell lines, described in Figures 2 ± 6. We ®rst conducted experiments to determine if the peptides coded for by the transcription units in the plasmids presented in Table 1 and in Figure 1, once transiently expressed in myeloid cell lines, would be detectable with the predicted molecular weights by Western blotting or immunoprecipitation. These latter experiments are described in the next section. CMV and tetracycline inducible bcr transcription units The expression of the bcr transcription units was ®rst studied using plasmids with the pcDNA3 expression vectors in which the transcription units were driven by the CMV promoter. As observed by us in these experiments, and as described by others previously, the level of expression of these peptide transcription units was variable and very low (data not shown). Because of the low expression levels of the CMV driven peptide transcription units shown in Table 1, we next studied a second series of bcr expression vectors (see Figure 1 and Table 1), in which the transgene

Table 1 Explanation of expression vector plasmids Plasmid p GD-p210 pGDX PUHD15-1neo pUHC13-3 PUHD10-3 pUHD-Bcr1-160 pUHD-Bcr1-160/ABL32 pUHD-Bcr1-412/ABL32 PUHD-Bcr161-412/AL32

Coding region of the expresson vector full length Bcr-Abl pGD backbone vector constitutive expression vector encoding Tetracycline repressor and VP16 transactivation domain luciferase reporter gene under the control of tetracyline inducible promoter tetracycline inducible backbone vector tetracyline inducible expression vector encoding ®rst 160 amino acids of Bcr-Abl tetracylin inducible expression vector encoding ®rst 160 amino acids of Bcr fused with last 33 amino acids of Abl tetracycline inducible expression vector encoding ®rst 412 amino acid of Bcr fused with last 33 amino acids of Abl tetracycline inducible expression vector encoding 161-412 amino acids of Bcr fused with last 33 amino acids of Abl

827


828

expression was driven by a minimal CMV promoter and the tetracycline enhancer sequence. These constructs, which contained a tetracycline driven promoter (designated tTA), were then studied by electroporating them transiently into the K562 chronic myelogenous leukemia blast crisis cell line, or into the 32D cell line. The constructs described in Table 1, are also presented in Figure 1. Since the binding of this protein to its cognate recognition sequence is inhibitable by the presence of tetracycline, the peptides coding for the fragments of bcr and abl described above, would be undetectable in the presence of tetracycline, and would be detectable in the absence of tetracycline. Clones of the modi®ed K562 or 32D cell lines containing the tTA cDNA (designated K562tet or 32Dtet), were selected on the basis of their ability to support the expression of a luciferase reporter gene driven by a tetracycline inducible enhancer. The criteria for choosing clones were: (1) no detectable expression in the presence of tetracycline; and (2) expression of the tTA transcriptional regulatory protein in the absence of tetracycline which was an intermediate level, but not so high that a toxic eect on the cell would be generated. We then analysed the transfectants for expression of the peptides by Western blotting or immunoprecipitation. We used monoclonal antibodies or polyclonal antibodies for those regions on bcr which were included in the peptides coded by the transcription units. As shown in Figure 2, these constructs were expressed in the absence of tetracycline at levels

Figure 1 Structure of fragments of bcr, c-abl and bcl ± abl fragments used for study: The human bcr protein contains 1271 amino acids, the ®rst 63 of which constitutes a region called a coiled-coil oligomerization domain (O). The tyrosine177 (Y177) is a binding site for Grb2 adaptor protein. This region, which is homologous to the human Dbl oncoprotein and to the yeast CDC24, is marked Dbl/CDC24. Close to the carboxyl terminus of the bcr is the Rac GTPase activting protein homologous region (Rac GAP). The c-abl region contains 1097 common amino acids encoding regions homologous to the Src oncoprotein SH3, which may inhibit the abl tyrosine speci®c protein kinase, SH2, which may direct the abl to bind phosphoprotein substrates, and SH1, which contains the tyrosine speci®c protein kinase domain (Y kinase). In addition, there is a nuclear localization signal (NTS), a DNA binding domain (DB), and an actin binding motif (AB). The four constructs which were prepared as described in the Materials and methods section, and are indicated in the Figure, are as follows: BCR1-412/ABL32, which contains all of the bcr aminoacids and the carboxyl-terminal 32 amino acids of c-abl, BCR161-412/ABL32, which lacks the bcr aminoacids 1 ± 160 (the oligomerization domain) but contains the rest of the Bcr, and the carboxyl-terminal 32 amino acids of c-Abl, BCR1-160/ABL32, which contains bcr aminoacids 1 ± 160 (which contains the oligomerization domain), in addition to the carboxyl-terminal 32 amino acids of c-abl, and BCR1-160, which contains the ®rst 160 aminoacids of bcr (the oligomerization domain), and no other aminoacids. The last 32 carboxyl-terminal amino acids of c-abl were designated by the letter E

detectable by Western blotting, and which were higher than the transcription units with the pcDNA3 vector. These studies were conducted by introducing the pUHD15-1neo plasmid containing the tTA transactivator fusion gene and a neomycin resistance gene into the K562 line. This cell line was designated K562Tet.

Figure 2 The tetracycline regulated expression of the Bcr ± Abl peptide fragment transcription units in K562 cells. (a) Transient transfection of K562 which contains the Tetracycline/VP16 transcriptional regulatory protein, with the transcription units presented in Table 1 and Figure 1, was carried out by the BTX electroporator. The cells were incubated with (+ above the lane) or without (7 above the lane) tetracycline, following which they were radiolabeled with 35S-methionine for 2 h. The cell lysates were precleared with preimmune rabbit serum and protein A/plus G agarose. The resultant supernatants were incubated with anti-bcr antibody along with protein A/plus G agarose, washed, and subjected to the 12% SDS ± PAGE. After drying, the gel was exposed to autoradiography. Note that the epitope of the anti-bcr antibody is located at bcr aminoacids 103 ± 114. (b) Transiently transfected K562 cells, which carry the tetracycline/VP16 transcriptional regulatory protein, were cultured in the presence (+ above the lane) or absence (7 above the lane) of tetracycline, and were lysed in RIPA buer. 20 mg lysates were loaded on each lane. After electrotransfer onto the Immobillon-P membranes, they were hybridized with anti-bcr antibody which recognizes the epitope on Bcr161-412/ABL32. The membrane was then incubated with HRP conjugated secondary goat anti-rabbit antibody and visualized by enhanced chemiluminescence system

Figure 3 32Dtet cells depend on IL3 for growth: 32Dtet cells, showing an intermediate level of expression of a reporter gene driven by a tetracycline regulated enhancer, were washed and resuspended in the absence or in the presence if IL3. A small number of cells was collected at dierent time intervals and the number of viable cells was determined by trypan blue exclusion. The viable cell number per ml is depicted on the ordinate. The dotted line connected by the solid circles indicates cell growth in the presence of IL3, whereas the line connected by the solid triangle line represents the cell proliferation in the absence of IL3


The expression of these transcription units was monitored by [35S]Methionine and immunoprecipitation. In the presence of tetracycline, which inactivates the tTA tetracycline transcriptional regulatory protein, the expression of all of these peptide transcription units was undetectable or very low, as shown in Figure 2. In the absence of tetracycline, levels of the peptides

detectable by immunoprecipitation and Western blotting, were seen, when measured by [35S]methionine and immunoprecipitation (see Figure 2). The monoclonal anti ± bcr antibody used for most of the studies presented in Figure 2, which was generated by injecting a synthetic peptide containing amino acids 103 ± 114 in Bcr protein, can not recognize the peptide produced by the Bcr161-412/ABL32 vector, in which the ®rst 160 amino acids were deleted. Another polyclonal anti-bcr antibody was used to examine the expression of pUHD-Bcr161-412/ABL32. However, the latter antibody did not precipitate the polypeptide eciently in immunoprecipitation assay. The Western blotting was then used to detect the expression of Bcr161-412/ABL32 (see Figure 2b). These results showed that in the absence of added tetracycline antibiotic, the transcription units introduced into the 32D cells were expressed at detectable levels and in a tetracycline inducible manner. Co-transfection of the bcr oligomerization domain inhibitory peptide transcription unit and the transcription unit for the p210bcr ± abl in a ratio of 5 to 1 Mock transfected 32Dtet cells do not generate colonies in methylcellulose in the absence of IL3. These cells still absolutely depend on IL3 to survive (Figure 3 and data not shown). 32D cells carrying the bcr±abl cDNA coding for the p210bcr ± abl protein

Figure 4 Bcr transcription unit suppresses colony formation induced by P210bcr±abl. 32D tetcells, which contained a CMV driven transcription unit coding for the tetracycline/VP16 transcriptional regulatory protein, were transiently co-transfected with pGD-p210 (a plasmid which contains the bcr±abl cDNA) plus individual plasmids with the bcr fragment cDNAs shown in Figure 1 and Table 1. The bcr±abl and bcr plasmids were cotransfected in a 1/5 ratio. The bcr cDNA fragment plasmids were in pUHD10-3 (tetracycline regulated) based transcription units encoding: Bcr1-160, Bcr1-160/ABL32, Bcr1-412/ABL32, and Bcr161-412/ABL32, and a backbone vector. After recovering for 24 ± 48 h, the cells were then plated in methyl cellulose in the absence of IL3 and tetracycline. After incubation for about 2 weeks, the colonies with more than 100 cells were counted. The average of three experiments was depicted in each histogram. The primary data for this ®gure expressed as percent of control is presented in Table 2

Figure 5 Dose-dependent inhibition of bcr±abl induced IL3 independent colony formation of 32D cells. 32Dtet cells were transiently transfected with the pGD vector as negative control, or were transfected with mixtures of the pGD-p210 plasmid with pUHDBCR1-160 fragments containing the ®rst 160 amino acids in a 1 : 1, 1 : 5 and 1 :10 ratios. The percentage of the colony formation as compared with pGD-p210 plus backbone vector (de®ned as 100%) from three experiments is shown. The primary data expressed as percent of control is presented in Table 3

Figure 6 Schematic representation of tetracycline inducible expression of transcription units encoding the bcr tetramerization domain and inhibition of p210bcr ± abl transforming activity. Fusion protein of the tTA tetracycline/VP16 chimeric transactivator was constitutively expressed in the cells under the control of the CMV promotor. In the presence of tetracycline, the fusion protein will bind to tetracycline, which prevents it from binding to the tetracycline recognition enhancer in the DNA promoter, and the transcription unit under the control of minimal CMV promoter and tetracycline repressor binding elements will not be activated or expression. In the absence of tetracycline however the tTA fusion protein will bind to the upstream tetracycline repressor binding sequences and activate the transcription. The messenger RNA will be translated into the bcr tetramerization domain and bind to the wild type p210bcr ± abl to inhibit the tetramer formation of p210bcr ± abl and thus abrogate the oncogenic eect of p210bcr ± abl

829


830

Table 2 Co Transfection of Bcr AA 1-160 transcription unit with bcr±abl transcription unit suppresses colony formation expressed as percentage of plating eciency seen when the bcr±abl transcription unit is transfected alone Plasimd

Exp 1

Exp 2

Exp 3

Mean

Standard deviation

Mock P210+BCR P210+BCR P210+BCR P210+BCR P210+BCR

0 100 9.4 18.9 43.4 28.0

0 100 26.0 19.0 89.7 60.3

0 100 7.0 41.0 114.0 87.3

0 100 14.1 26.3 82.3 58.5

0 0 6.7 10.5 29.3 19.2

AA 1-160 AA 1-412+ABL32 AA 1-412+ABL32 AA 161-412+ABL 32 AA 1-160+ABL32

AA=amio acid, P210=P210bcr±abl, Exp=experiment, Mock=cells exposed to transfection conditions without plasmid DNA.

Table 3 Dose-dependent inhibition of bcr±abl induced IL3 Independent Colony formation of 32D cells following transfection of mixtures of BCR transcription units (with and without oligomerization unit-AA1-160) with bcr±abl transcription unit in ratios of 1/1, 5/1 and 10/1 (results are expressed as percent of plating eiciency with transfection of bcr±abl transcription unit alone) Plasmid

Exp 1

Exp 2

Exp3

Mean

MOCK P210 Alone BCR/P210 Ratio 1/1 5/1 10/1

0 100

0 100

0 100

0 100

67.0 33.0 0.0

112.5 62.5 25.0

90.0 80.0 40.0

89.8 58.5 21.7

Standard deviation 0 0 18.6 19.3 16.5

Exp=experiment P210= P210bcr±abl, Mock=cells exposed to the conditions of transfection without plasmid DNA

generates an average of 80 colonies/100 000 cells plated. Thus, the P210bcr±abl renders the 32D myeloid cell line IL-3 independent (see Figure 4 and Table 2). A ®xed (1/5) ratio of the plasmid containing the bcr± abl cDNA, and the plasmids coding for the Bcr1-160, BCR1-160/Abl32, Bcr61-412/Abl32, and Bcr1-412/ Abl32, was transiently transfected into 32Dtet cells and plated in methyl cellulose in the absence of IL3. As shown in Figure 4 and in Table 2, the plasmid transcription unit coding for the ®rst 160 amino acids of the bcr (p210+BCR1-160), which contains the oligomerization domain, signi®cantly suppresses the plating eciency of the 32D cells (fourfold reduction of plating eciency). A similar degree of inhibition is seen when the transcription unit containing the cDNA coding for the ®rst 412 amino acids of bcr and the last 32 aminoacids of c-abl (p210+Bcr1-412/BAL32) is transiently co-electroporated along with the plasmid containing the P210bcr±abl transcription unit. In contrast, when the cDNA coding for the bcr aminoacids 161 ± 412 (which lack the bcr oligomerization domain) and the last 32 aminoacids of the c-abl (p210+Bcr161-412/ABL32) is transiently co-electroporated along with the plasmid containing the bcr ± abl cDNA, no signi®cant reduction of the plating eciency is seen, as shown in Figure 4 and Table 2. The plasmid which contains the ®rst 160 bcr aminoacids attached to the last c-abl aminoacids (p210+Bcr1-160/ABL32), when cotransfected along with the plasmid for the bcr±abl, generated an intermediate level of inhibition (80 to 55 colonies). Perhaps this is due to an eect of the last 32 aminoacids of the c-abl, which contains an actin binding domain. The presence of the c-abl carboxy terminal fragment may sequester the bcr oligomerization domain peptide to the actin monomers. This would keep bcr oligomerization domain away from the

Table 4 32Dtet cells were transfected with pSVZeosin plus individual transcription unit in 1:10 ratio Plasmid

Oligomerization domain show non-toxic to 32D cells Colony number

pUHD10-3 pUHD-Bcr1-412/ABL32 PUHD-Bcr1-160 PUHD-Bcr161-412/ABL32 pUHD-Bcr1-160/ABL32

84 265 240 239 234

The transfected cells were then plated in methyl cellulose in the presence of IL3 conditioned medium and 1.5mg/ml Zeosine but lack of tetracycline. After 2 weeks the colonies of more than 100 cells were counted

target P210bcr±abl, which is diusely distributed throughout the cytoplasm. We next transiently co-transfected the transcription unit which coded for the ®rst 160 bcr amino acids into the 32Dtet cells along with the pGD-p210 plasmid in the following varying ratios of 1 : 1, 5 : 1 and 10 : 1. The reduction of the plating eciency of the 32Dtet cells was examined in three independent experiments, in which the experimental design permitted one to test if the protein product of the bcr peptide inhibitory transcription unit inhibits the transforming function of the bcr ± abl protein product. As shown in Figure 5 and in Table 3, the co-transfection resulted in a dosedependent decrease in p210bcr±abl induced IL3 independent colony formation of 32Dtet cells in methylcellulose. Overexpression of the bcr transcription units did not inhibit 32Dtet cell growth in methylcellulose supplemented with IL3 To test for the possible nonspeci®c inhibitory eects of the bcr ± abl fragments on 32D cells, we then


cotransfected a dominant selection marker pSVzeosine with the bcr transcription unit in a 1 : 10 ratio into the 32Dtet cells and plated them in methylcellulose in the presence of IL3. The 32Dtet cells grew into colonies, which were counted if they contained more than 100 cells in the presence of IL3 (Table 4). There was no signi®cant suppressive eect on colony growth seen in those transfectants which contained the bcr aminoacids 1 ± 160 (240 and 234 colonies) as compared to those 32Dtet cells without the ®rst 160 bcr aminoacids (239 colonies). Thus, the bcr oligomerization domain did not non speci®cally inhibit the growth of the 32Dtet cells when the 32Dtet cells were not dependent on the P210bcr±abl protein for growth. Discussion The ®rst 63 amino acids of the P210bcr±abl protein promotes the formation of tetramers of p210bcr ± abl, as shown in Figure 6 (McWhirter et al., 1993b). An association between p210bcr±abl and p160bcr in myeloid leukemia has been detected (Lu et al., 1993). This region has been shown to be essential for P210bcr ± abl transformation in hematopoietic cells. Disruption of this domain by inserting a b-turn sequence renders the P210bcr ± abl non-transforming. It is therefore possible to interrupt the formation of tetramers of P210bcr±abl by over expression of this domain. The data presented in this paper shows that the presence of a stoichiometic excess of a fragment containing the protein protein contact point for tetramerization of the P210bcr±abl, suppresses the transformed phenotype of the P210bcr±abl positive cells. Surprisingly, the ®rst 160 amino acids of bcr was found to be more eective in the suppression of p210bcr ± abl transforming activity, than constructs of the ®rst 160 amino acids with additional amino acids either from the bcr region or those from abl. One possible explanation is that the peptide which is composed of the ®rst 160 amino acids of bcr may be more stable in the cells than other fragments, or alternately, the decrement in suppression of the p210bcr ± abl associated growth factor independent proliferation which is seen when the abl actin binding sequences are added to the bcr oligomerization domain, may result from translocation of the oligomerization domain to F-actin cytoskeleton, thus making these sequences unavailable for binding to and inhibition of the P210bcr ± abl oncoprotein. On the other hand the pUHD-Bcr1-412/ABL32 plasmid contains additional sequences, such as the tyrosine177 (Pendergast et al., 1993), the SH2 binding region, and the serine/threonine kinase activity, although functional consequences of these domains to the p210bcr ± abl transformation activity, have not been established. Recently, tyrosine360 of the bcr portion of the p210bcr ± abl oncoprotein has been found to be involved in regulation of serine/threonine kinase activity (Lu et al., 1993). This tyrosine may be the target of elevated abl kinase. However its role in P210bcr ± abl transformation remains to be de®ned. Targeted disruption of protein-protein interactions may enable us to speci®cally interfere with the signals that transforming proteins deliver. The results described in this report suggest that peptides which bind

to the oligomerization domain of the P210bcr±abl may suppress the transformed phenotype of P210bcr±abl positive cells. In order to translate these ®ndings into a new direction in therapy for CML, the synthesis of peptidomimetic chemicals, which mimic the structure and the inhibition of the oligomerization domain fragment, may provide a new approach to the treatment of CML. Such chemicals, would be stable in acqueous solution, membrane permeable, and would bind at high anity to the oligomerization domain of P210bcr±abl, thus suppressing its transforming function. Materials and methods Cell lines and plasmids The 32D cl3, a mouse myeloid cell line (Greenberger et al., 1983), was cultured at 378C in a 5% CO2 incubator in Dulbecco's modi®ed Eagle's medium supplemented with 10% heat inactivated fetal bovine serum (HyClone, Logan, UT) and 10% WEHI-3B conditioned medium as a source of interleukin-3(IL-3). The K562 chronic myeloid leukemia cell line (American Type Culture Collection) was grown at 378C, 5% CO2 in RPMI 1640 medium supplemented with 10% heat inactivated fetal bovine serum. The plasmids for these studies are summarized in Figure 1 and Table 1. pGD-p210 was a gift from Dr David Baltimore (Daley et al., 1990). The pGDX plasmid was generated by restriction digestion of pGD-p210 with XhoI endonuclease and subsequent re-ligation of the plasmid so that the cDNA sequence coded for p210bcr ± abl was removed. The pUHD151neo, pUHC13-3 and pUHD10-1 plasmids were provided by Dr Hermann Bujard (Gossen and Bujard, 1992; Resnitzky et al., 1994). The pUHD-Bcr1-412/ABL32 was constructed by BglII digestion of pcDNA3-p210 and subsequent religation of DNA fragments encoding the ®rst 412 amino acids of the bcr in frame with the last 32 amino acids of the abl. This pUHDBcr161-412/ABL32 plasmid was subcloned by restriction endonuclease digestion of the pcDNA-Bcr1-412/ABL32 plasmid with KpnI and BamHI. The two ends are then joined by an adaptor containing the ATG start codon. The pUHD-Bcr1-160 was subcloned by ®rst digesting pcDNAp210 with EcoRI and BamHI. We then inserted the fragment encoding ®rst 160 amino acids of bcr into pBluescript SK(7). The resultant pBS-Bcr160Bam was cut with BamHI and XbaI and religated with an adaptor containing a stop codon. The pUHD-Bcr1-160/ABL32 plasmid was prepared by digesting pcDNA-p210 with BglII and SpeI. The resulting fragment encoding the last 32 amino acids of the abl protein was inserted into pBS-Bcr160Bam at compatible BamHI and XbaI sites so that the ®rst 160 amino acids of bcr were fused in frame with the last 32 amino acids of the abl. The XbaI site was regenerated by subcloning the insert into pcDNA3 vector with EcoRI and XhoI. The resultant plasmid was digested with EcoRI and XbaI and inserted into the pUDH10-3 expression vector under the control of seven repeats of the Tet operator linked to cytomegalovirus (CMV) minimal promotor. The structure of all of these constructs is summarized in Table 1 and Figure 1. Transfection The K562 cells were electroporated using the BTX electroporator (BTX, Inc., San Diego) according to the manufacturer's recommended protocol. For establishing the inducible expression system, the cells were plated on methyl cellulose supplemented with 750 mg/ml of G418 for 14 days. The resulting colonies were picked under microscopic visualization and then expanded in liquid medium. Each individual expanded clone was then

831


832

transfected with the pUHC13-3 luciferase reporter construct and incubated in the presence or absence of tetracycline. The luciferase reporter gene activity was then checked using a luminometer and a luciferase assay kit (Promega, Madison, WI). The 32Dtet clones were established in a similar manner, except that the electroporation was carried out at 225V and 960 mFi for 26106 cells. The selections were performed in liquid medium for 2 weeks and cloned by limiting dilution.

of more than 200 cells were counted. For Zeocin selection, the pZeoSV plasmid encoding the Zeocin resistance gene was cotransfected with a series of distinct transcription units expressing bcr ± abl fragment polypeptides in a ratio of 1 : 10. After recovering for 24 ± 48 h the 32D cells were then selected in 1.5 mg/ml Zeocin in methylcellulose plates. The colonies of cells containing more than 100 cells were then counted.

Immunoprecipitation and immunoblotting

Abbreviations IMDM, Iscove's modi®ed Dulbecco's medium; a.a., amino acid; SH1, Src homology region 1; SH2, Src homology region 2; SH3, Src homology region 3; CML, chronic myelogenous leukemia; IL-3, interleukin 3; HLH, helix ± loop ± helix; CMV, cytomegalovirus; SDS, sodium dodecyl sulfate; PBS, phosphate buered saline; PAGE, polyacrylamide gel electrophoresis.

Transfected cells were radiolabeled and immunoprecipitated using monoclonal antibody against Bcr(G6) (Oncogene Science, NY). The expression of the pUHDBcr161-412/ABL32 transcription unit was detected by Western blotting using a polyclonal antibody against bcr which recognizes the short peptide sequence contained in Bcr161-412/ABL32.35 mg of total cell lysate was loaded on each lane. Methyl cellulose plating assay pGDX or pGD-p210 was linearized with NheI and pUHD10-3 based plasmid vectors were linearized by SSpI. 32D cells were exposed to co-electroporation with a mixture of plasmids, for the P210bcr±abl and bcr oligomerization inhibitory domains, which varied from 1/10 to 1/1. After electroporation, 32Dtet cells were allowed to recover in the presence of IL-3 (without tetracycline) for 24 ± 48 h. The cells were then collected and washed three times with serum-free DMEM, and 16106 viable cells were mixed with 3 ml of methyl cellulose complete medium (StemCell, Vancouver, Canada) without IL-3. The cells were then grown in a humidi®ed 5% CO2 incubator at 378C for 2 weeks. The colonies that consisted

Acknowledgements We thank Drs David Baltimore and Martin L Scott for providing the plasmid pGD-p210, Drs Hermann Bujard and Sabine Freundlieb for providing the plasmids pUHD151neo, pUHC13-3 and pUHD10-3. The authors acknowledge support for this work provided to ABD by the NCI (P01 CA 55162 and P01 CA49639), The George and Barbara Bush Fund for Leukemia Research (to AD), The Anderson Chair for Cancer Treatment and Research (to AD), the Stringer Chair in Cancer Research (to RA) at the MD Anderson Cancer Center, the Hull Development Fund of the Gene Therapy Program of the Yale Cancer Center (to AD) and the Ensign Professorship of Medicine (to AD) at the Yale University School of Medicine.

References Bedi A, Zehnbauser BS, Barber JP, Sharkis SJ and Jones RJ. (1994). Blood, 83, 2038 ± 2044. Ben-Nerial Y, Daley GQ, Mes-Mason AM, Witte ON and Baltimore D. (1986). Science, 233, 212 ± 214. Carlesso N, Grin JD and Druker BJ. (1994). Oncogene, 9, 149 ± 156. Cortez D, Kadlec L and Pendergast AM. (1995). Mol. Cell. Biol., 15, 5531 ± 5541. Dai Z and Pendergast AM (1995). Genes and Dev., 9, 2569 ± 2582. Daley GQ and Baltimore D. (1988). Proc. Natl. Acad. Sci. USA, 85, 9312 ± 9316. Daley GQ, Van Etten RA and Baltimore D. (1990). Science, 247, 824 ± 830. Elefanty AG and Cory S. (1992). Mol. Cell. Biol., 12, 1755± 1763. Gishizky ML, Johnson-White J and Witte OM. (1993). Proc. Natl. Acad. Sci. USA, 90, 3755 ± 3759. Goga A, McLaughlin J, Pendergast AM, Parmar K, Muller A, Rosenberg M and Witte ON. (1993). Mol. Cell. Biol., 13, 4967 ± 4975. Gossen M and Bujard H. (1992). Proc. Natl. Acad. Sci. USA, 89, 5547 ± 5551. Greenberger JS, Sakakeeny MA, Humphries RK, Eaves CJ and Eckner RJ. (1983). Proc. Natl. Acad. Sci. USA, 80, 2391 ± 2395. Heisterkamp N, Stam K, Groen J, DeKlein A and Grosveld G. (1985). Nature, 315, 758 ± 761. Hoeve J, Arlinghaus RB, Guo JQ, Heisterkamp N and Groen J. (1994). Blood, 86, 1731 ± 1736. Jackson P and Baltimore D. (1989). EMBO J., 8, 449 ± 456.

Jackson P, Baltimore D and Picard D. (1993). EMBO J., 12, 2809 ± 2819. Jain SK, Langdon WJ and Varticovski L. (1997). Oncogene, 14, 2217 ± 2228. Kharbanda S, Ren R, Pandey P, Shafman TD, Feller SM, Weichselbaum RR and Kufe DW. (1995). Nature, 376, 785 ± 788. Kipreos ET and Wang JYJ. (1992). Science, 256, 382 ± 385. Kloetzer W, Kurzrock R, Smith L, Talpaz M, Spille M, Gutterman J and Arlinghaus R. (1985). Virology, 140, 230 ± 238. Konopka JD, Watanabe SM and Witte ON. (1984). Cell, 37, 1035 ± 1042. Liu J, Wu Y, Ma GZ, Lu D, Haataja L, Heisterkamp N, Groen J and Arlinghaus R. (1996). Mol. Cell. Biol., 16, 998 ± 1005. Lu D, Liu J, Campbell M, Guo JG, Heisterkamp N, Groen J, Canaani E, Arlinghaus R. (1993). Blood, 82, 1257 ± 1263. Lugo TG, Pendergast AM and Witte ON. (1990). Science, 247, 1079 ± 1082. Lugo TG and Witte ON. (1989). Mol. Cell. Biol., 9, 1263 ± 1270. Maru Y and Witte ON. (1991). Cell, 67, 459 ± 468. Mayer BJ and Baltimore D. (1994). Mol. Cell. Biol., 14, 2883 ± 2894. McWhirter JR, Galasso DL and Wang JYJ. (1993a). EMBO J., 12, 1533 ± 1546. McWhirter JR, Galasso DL and Wang JYJ. (1993b). Mol. Cell. Biol., 13, 7587 ± 7595.


McWhirter JR and Wang JYJ. (1991). Mol. Cell. Biol., 11, 1553 ± 1565. Mitinori S, Sugai S, Tanaka T, Shimouchi K, Fuchs E, Narumiya S and Kakizuka A. (1995). Nature, 374, 159 ± 162. Muller AJ, Young JC, Pendergast AM, Pondel M, Landau NR, Littman DR and Witte ON. (1991). Mol. Cell. Biol., 11, 1785 ± 1792. Muvre C, McCaw PS and Baltimore D. (1989). Cell, 56, 777 ± 783. Papadoponlos P, Ridge SA, Boucher CA, Stocking C and Wiedemann LM. (1995). Cancer Res., 55, 34 ± 38. Pendergast AM, Muller AJ, Kavlik MH, Clark R, McCormick F and Witte ON. (1991). Proc. Natl. Acad. Sci. USA, 88, 5927 ± 5931. Pendergast AM, Quilliam LA, Cripe LD, Basing GH, Dai Z, Li N, Batzer A, Rabun KM, Der CJ, Schlessinger L and Gishizky ML. (1993). Cell, 75, 175 ± 185. Puil L, Lui J, Gish G, Mbamalu G, Bowtell D, Giuseppe PP, Arlinghaus R and Pawson T. (1994). EMBO J., 13, 764 ± 773. Raitano AB, Hallpern JR, Hambuch TM and Sawyers CL. (1995). Proc. Natl. Acad. Sci. USA, 92, 11746 ± 11750. Ren R, Ye ZS and Baltimore D. (1994). Genes and Dev., 8, 783 ± 795. Renshaw MW, McWhirter JR and Wang JYJ. (1995). Mol. Cell. Biol., 156, 1286 ± 1293. Resnitzky D, Gossen M, Bujard H and Reed SI. (1994). Mol. Cell. Biol., 14, 1669 ± 1679. Rosenberg N and Witte ON. (1988). Adv. Virus Res., 35, 39 ± 81. Schwarzberg PL, Stall AM, Hardin JD, Bowdish KS, Humaran T, Boast S, Harbison ML, Robertson EJ and Go SP. (1991). Cell, 65, 1165 ± 1175.

Shi Y, Alin K and Go SP. (1995). Genes and Dev., 9, 2583 ± 2597. Shore SK, Bogart SL and Reddy EP. (1990). Proc. Natl. Acad. Sci. USA, 87, 6502 ± 6506. Skorski T, Bellacosa A, Nieborowska-Skorski M, Majewski M, Martinez R, Choi JK, Trota R, Wlodarrski P, Perrotti D, Chan TO, Wasik MA, Tsichlis PN and Calabretta B. (1997). EMBO J., 16, 6151±6161. Szcylik C, Skorski T, Nicolaides NC, Manzella L, Malaguarnera L, Venturelli D, Gerwirtz AM, Calabretta B. (1991). Science, 253, 562 ± 565. Usala SJ, Bale AE, Gesundheit N, Weinberger C, Lash RW, Wondisford FE, McBride OW, Weintraub BD. (1988). Mol. Endocrin., 2, 1217 ± 1220. Van Etten RA, Jackson P and Baltimore D. (1989). Cell, 58, 669 ± 678. Van Etten RA, Jackson P, Baltimore D, Sanders MC, Matsudaira PT and Janmey PA. (1994). J. Cell Biol., 124, 325 ± 340. Wang JYJ. (1988). Oncogene Res., 3, 293 ± 298. Welch PJ and Wang JYJ. (1993). Cell, 75, 779 ± 790. Wetzler M, Talpaz M, Van Etten RA, Hirsh-Ginsberg C, Beran M and Kurzrock RM. (1993). Clin. Invest., 92, 1925 ± 1939. Yuan AM, Huang YY, Whang Y, Sawyers C, Weichselbaum R, Kharbanda S and Kufe D. (1996). Nature, 382, 272 ± 274. Zavacki AM, Harney JW, Brent GA and Larsen PR. (1993). Mol. Endocrin., 7, 1319 ± 1330.

833