Wnt signaling in B-cell neoplasia - Nature

Oncogene (2003) 22, 1536–1545

& 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc

Wnt signaling in B-cell neoplasia Ya-Wei Qiang, Yoshimi Endo, Jeffrey S Rubin and Stuart Rudikoff* Laboratory of Cellular and Molecular Biology, National Cancer Institute, NIH, Bethesda, MD 20892, USA

Wnts comprise a family of secreted proteins that interact with receptors consisting of a Frizzled (Fz) family member alone or complexed with LDL receptor-related proteins (LRP5/6). Wnt signaling plays a crucial role in both development and differentiation, and activation of a ‘canonical’ Wnt pathway resulting in b-catenin stabilization is associated with several types of human cancers. To date, little is known about potential Wnt signaling in mature lymphocytes or lymphoid neoplasia. Herein, we have analysed Wnt signaling in mature B cells (lymphomas) and plasma cells (multiple myeloma). Both Fz and LRP5/6 mRNAs were expressed in myeloma lines, but LRP5/6 were not observed in lymphomas. In myelomas, a canonical Wnt signaling pathway was activated following treatment with Wnt-3a as assessed by accumulation of bcatenin, but b-catenin levels actually decreased in lymphoma cells. Wnt-3a treatment further led to striking morphological changes in myeloma cells accompanied by rearrangement of the actin cytoskeleton. Morphological changes were associated with a second Wnt pathway dependent on Rho activation. These results suggest that Wnt responsiveness is a stage-specific phenomenon in Bcell development and that the morphological changes associated with Wnt signaling may play a role in the motility and metastatic potential of myeloma cells. Oncogene (2003) 22, 1536–1545. doi:10.1038/sj.onc.1206239 Keywords: Wnt; b-catenin; myeloma; morphology; Rho

Introduction Lymphocytes respond to a wide variety of cytokines and growth factors that play pivotal roles in lymphoid development and function. Recently, Wnt family members have been demonstrated to contribute to lymphopoiesis and early stages of both B- and T-cell development, but a signaling role in mature lymphocytes or lymphoid neoplasia has yet to be determined. Wnts mediate signaling through their cognate Frizzled (Fz) receptors and associated LDL receptor-related protein 5 or 6 (LRP5/6) resulting in the activation of intracellular cascades that have profound biological effects. In *Correspondence: S Rudikoff; E-mail: [email protected] Received 23 September 2002; revised 8 November 2002; accepted 12 November 2002 This article is a United States Government Work Paper as defined by the US Copyright Act

Drosophila embryogenesis, Wnts determine cellular polarity and fate (Wodarz and Nusse, 1998). Wnt effects are not restricted to early development as, for example, in adult tissues Wnts have been suggested to regulate epithelial cell proliferation (Polakis, 2000). This regulation assures that replacement of cells lost because of death is tightly controlled and results in a precise maintenance of cell numbers. Defects in Wnt signaling have been associated with several forms of cancer (Peifer and Polakis, 2000; Polakis, 2000). Many Wnt effects are mediated by the downstream effector b-catenin, which plays a pivotal role in the ‘canonical’ Wnt pathway. In the absence of Wnt signaling, b-catenin is phosphorylated by GSK3b marking it for ubiquitination and proteosome degradation. Phosphorylation of b-catenin takes place in a complex including adenomatous polyposis coli (APC) and axin. Both of these proteins are also phosphorylated by GSK3b and required for efficient phosphorylation of b-catenin. Upon interaction of Wnt proteins with the coreceptors Fz and LRP5/6 (Pinson et al., 2000; Tamai et al., 2000; Wehrli et al., 2000), the APC/axin/GSK3b complex is antagonized by a mechanism that is not clear, but involves dishevelled (Dvl) family members thought to be downstream of Fz receptors (Yanagawa et al., 1995; Lee et al., 1999). As a result, Dvl, in a complex with a second protein FRAT, inhibits GSK3b phosphorylation of b-catenin and augments dissociation of the destruction complex. b-catenin thus accumulates in the cytoplasm and translocates to the nucleus where it functions as a transcription factor in conjunction with TCF/LEF family members to activate target genes that regulate many cellular processes including cell cycle progression and differentiation. In addition to its role as a transcription factor, b-catenin performs a second major function as a structural protein at cell–cell adherens junctions linking adhesion receptors of the cadherin family to the actin cytoskeleton (Zhurinsky et al., 2000). Thus, b-catenin levels in the cell reflect a complex dynamic involving multiple cellular compartments and pathways associated with quite different effects ranging from transcription regulation to membrane structure and cell shape. In addition to the roles described above for b-catenin in normal cellular functions, activation of this pathway has been described in a variety of human cancers, the most well studied of which are colorectal tumors, but also include fibromatosis, gastric and hepatocellular carcinoma (Peifer and Polakis, 2000; Polakis, 2000). bcatenin involvement in cancer is commonly associated with mutations in the amino-terminal region that make

Wnt signaling in multiple myeloma Y-W Qiang et al

1537

the molecule resistant to processing through the degradation pathway. Mutations in other proteins in this pathway, most notably the APC gene in colon cancer, similarly lead to b-catenin accumulation. A role for other Wnt-activated (noncanonical) pathways in disease has yet to be determined. While importance of the Wnt/b-catenin pathway has been clearly established in embryogenesis, development and some forms of cancer, relatively little is known about the potential involvement of this cascade in lymphoid development and lymphoid neoplasia. Multiple myeloma is an invariably fatal form of cancer characterized by the development of malignant plasma cells in the bone marrow that are representative of terminally differentiated, antibody-secreting B-lineage cells. During the past several years, a variety of studies have suggested that cytokines and growth factors are likely to play important roles in the survival and growth of these cells. IL-6 has been demonstrated to act as a proliferative and antiapoptotic factor for both primary explants and a number of myeloma cell lines (Bergui et al., 1989; Zhang et al., 1994). Importantly, in animal models IL-6 null mice fail to develop plasma cell tumors (Hilbert et al., 1995; Lattanzio et al., 1997). Similar effects on myeloma lines have also been reported for IGF-I (Ferlin et al., 2000; Ge and Rudikoff, 2000) and both of these factors have been shown to promote myeloma tumor growth in in vivo models (Lu et al., 1995; Ge and Rudikoff, 2000). In contrast to multiple myeloma, B-cell lymphomas represent an earlier stage of B-cell development corresponding to mature B cells that express high levels of surface immunoglobulin, but secrete relatively little. Herein, we have undertaken to examine the question of Wnt signaling in mature B-lineage cells and describe functional Wnt/b-catenin and Wnt/RhoA signaling in myeloma lines that is absent in B-cell lymphomas.

Results The general insolubility of Wnt proteins has hampered the production of appropriate antibodies to facilitate Table 1

experimental detection. We, therefore, initially used RT–PCR to evaluate the presence of Wnt receptor RNA in both myeloma and lymphoma cells. Analysis using primers for all Fz family members (Table 1) revealed expression of multiple Fzs at relatively high levels in myeloma cells, but a clear lack of such expression in lymphpomas wherein only Fz-3 was detected at comparable levels. Importantly, neither LRP-5 nor -6 was expressed in any B cell line whereas expression of one, or both, was observed in all myeloma lines. These results suggest that multiple Wnt receptors are expressed specifically in plasma cells, but not B cells, and may reflect stage-specific regulation. Having demonstrated the presence of Fz and LRP receptors, we sought to determine whether a canonical b-catenin pathway was functional in myeloma cells. We first examined Dvl isoforms known to be downstream of the receptor and to play a critical role in inhibiting bcatenin processing by the GSK3b/APC/axin complex. Dvl-1 was barely detectable (not shown), whereas Dvl-2 and Dvl-3 were readily observed in all the lines tested. The protein levels of Dvl-2 increased with time following Wnt-3a treatment, with maximum levels attained by 6 h (Figure 1). Dvl-3 appeared to become phosphorylated in the same time frame as indicated by a gel mobility shift. This change in mobility was dependent on the concentration of Wnt-3a as demonstrated by the treatment with varying dilutions of conditioned medium (CM). Phosphorylation was confirmed by phosphatase treatment of immunoprecipitated Dvl-3. CM from parental L cells had no effect on either Dvl-2 or Dvl-3 at any time point (not shown). Detailed examination of the effect of Wnt-3a revealed b-catenin protein accumulation in a time-dependent manner. In the H929 line, increases in b-catenin levels were seen at 3 h and continued through the 12 h time point (Figure 2). Similar increases in OPM-2 were observed by 1 h and peaked between 3 and 6 h. Wnt-3a treatment had no effect on either a- or g-catenin protein levels (not shown). Interestingly, in 3/3 B-cell lymphoma lines b-catenin levels evidenced a continuous decline. It should be noted that exposure time of the film for the B cell lines was 12 that of the myeloma lines. Thus, the protein level of b-catenin in lymphomas was much lower

Fz and LRP expression in myeloma and lymphoma lines Myeloma

Lymphoma

8226

Delta47

H929

MM144

OPM-2

Daudi

Namalwa

ST486

Fz-1 Fz-2 Fz-3 Fz-4 Fz-5 Fz-6 Fz-7 Fz-8 Fz-9 Fz-10

+ +++ + + + ++

+++ + +++ + + +++ +

++ + +++ + +++ +++ + ++ ++

+++ + +++ + + +++ + ++

+++ +++ +++ + +++ +++ ++ + ++

+++

+ +++ +

+++ + +

LRP5 LRP6

+ +++

+ +

+++ +++

+ +++

+++

Oncogene


1538

a

0

1

3

6

9

12

hr

H929

Dvl-2

OPM-2

Dvl-2 0

b

1

3

6

9

12

a

0

1

3

6

9

12

hr H929

OPM-2

hr

Daudi

Dvl-3

H929

Nam alwa

Dvl-3

OPM-2

b

c % Wnt-3a 0.0 6.25

12.5 25.0 50.0 100.0

H929

Wnt-3a AP

+

H929

+ +

+

Dvl-3

+

+

+ +

+ +

β-tubulin

c

Li C l

0

10

20

40

µM

H929

H929 OPM-2

Figure 1 Wnt-3A modulation of Dvl proteins. Cells were cultured in serum-free growth medium for 12 h and then stimulated with Wnt-3a-CM for the times indicated (a, b). Cell lysates were resolved on SDS–PAGE gels, transferred to membranes and blotted with the indicated antibodies. (c) Cells were cultured at varying concentrations of Wnt-3a-CM diluted in con-CM and lysates prepared at 3 h for H929 and 6 h for OPM-2. (d) Lysates were prepared from con-CM (lane 1) or Wnt-3a-CM (lanes 2–4)-treated cells after 3 h and incubated with anti-Dvl-3 antibodies. Immunoprecipitates were collected and left untreated (lanes 1 and 2), treated with alkaline phosphatase (AP) buffer (lane 3) or AP plus AP buffer (lane 4). Samples were analysed by immunoblotting with anti-Dvl-3 antibodies

than in myeloma cells in spite of comparable mRNA levels (not shown). These results suggest that b-catenin protein synthesis and/or degradation is likely to be differentially regulated in myeloma versus lymphoma cells. The observed stabilization of b-catenin in myeloma lines was completely inhibited by secreted Fz-related protein-1 (sFRP-1), a putative antagonist of Wnt binding (Figure 2b). Additionally, lithium chloride (LiCl), which has also been shown to inhibit b-catenin phosphorylation by GSK3b, produced a similar stabilization of b-catenin (Figure 2c). Oncogene

OPM-2 Wnt-3a sFRP1

Dvl-3

OPM-2

d

ST486

OPM-2 Figure 2 Wnt-3a treatment leads to increased b-catenin protein levels in myeloma but not lymphoma cells. (a) Cells were cultured in serum-free growth medium for 12 h, then stimulated with Wnt3a-CM for the indicated time. Cell lysates were prepared and blotted with antibody to b-catenin. Films for myeloma lines H929 and OPM-2 were exposed for 20 s while those for the three B cell lines (Daudi, Namalwa and ST486) were exposed for 4 min. (b) Wnt-3a-CM or con-CM was incubated with sFRP-1 (10 mg/ml) for 30 min and then added to cultures for an additional 6 h followed by blotting as in (a). (c) Cells were treated with LiCl (at indicated concentrations) for 6 h. Lysates were analysed as in (a)

Stabilization of b-catenin in other systems leads to the formation of complexes with members of the TCF/LEF families and subsequent transcriptional activation. RT– PCR analysis of the TCF/LEF families revealed expression of only LEF-1 (Figure 3a) in either myeloma or B cells. Using a reporter construct (pTOPFLASH) containing binding sites for the b-catenin/TCF/LEF transcription complex, a significant increase in luciferase activity was observed indicating that transcriptional activation was a downstream effect of Wnt-3a treatment in myeloma cells (Figure 3b). Given the characterization of a functional Wnt signaling pathway in myeloma cells, we next sought to identify biological effects associated with the activation of Wnt signaling. Initial experiments using a number of


Daudi

ST486

Namalwa

OPM-2

MM144

H929

1539

a

LEF1 Actin b

3500

Relative Luciferase Units

3000 Con 2500

Wnt3a

2000 1500 1000 500 0 pTOPFLASH

pFOPFLASH

Figure 3 Wnt-3a induced activation of LEF/TCF transcription. (a) LEF-1 mRNA expression in myeloma and B cells. Total RNA was extracted from indicated cell lines (myelomas MM144, H929 and OPM-2, and B cell lymphomas Namalwa, ST486 and Daudi). RT–PCR was performed with primers for LEF1 and b-actin. (b) OPM-2 cells were transiently transfected with wild-type (TOPFLASH) or mutant (FOPFLASH) LEF/TCF luciferase reporter constructs. Cells were treated with Wnt-3a-CM or con-CM. Values represent the mean of six independent transfections for each construct

myeloma cell lines revealed no proliferative effect associated with Wnt-3a treatment (not shown). However, striking morphological changes were observed in 4/ 4 myeloma lines, associated with the attachment of cells to culture plates (Figure 4a–d). These changes were clearly visible by 1 h and increased over a 24 h period, resulting in large numbers of cells exhibiting extruded cytoplasmic structures resembling filopodia such that the cells acquired a fibroblast-like appearance. These projections were located at sites of attachment to the culture plates. Cells treated with control-CM (con-CM) retained their typical rounded appearance and did not attach to culture dishes at any time point for as long as 24 h. A marked rearrangement of the actin cytoskeleton (Figure 4e–h) accompanied the morphologic alterations, wherein actin staining was concentrated at the attachment points of the filopodia-like structures. Morphologic changes could be prevented by preincubation with

sFRP-1 (not shown). Again, such changes were not observed in a series of B-cell lymphomas. A series of experiments were next performed to assess the intracellular pathways associated with the observed morphological change. LiCl treatment of myeloma cells, previously shown (Figure 2c) to stabilize b-catenin, failed to induce these cellular alterations. To confirm that the canonical b-catenin pathway was not responsible for the morphological alterations, we treated myeloma cells with supernatants containing the LRP inhibitors, Dickkopf (Dkk)-1 and -2. As shown in Figure 5, both Dkk-1 and -2 inhibited Wnt-3a-induced accumulation of nonphosphorylated (active) b-catenin and transcriptional activation, but neither prevented the morphological changes (not shown). Identical results were obtained with two additional myeloma lines (not shown). Wnts have also been found to activate at least two additional pathways resulting in either calcium flux Oncogene


1540

a

Dkk-1 Wnt-3a

0

0.5

1

2

+ + +

3

+

+

hr β-cat Non-p- β-cat β-tubulin

b

0

Dkk-2 Wnt-3a

0.5

1

2

3

hr β-cat Non-p-β-cat

β-tubulin

c Relative Luciferase Units

3000 2500 2000 1500 1000 500

Figure 4 Morphological and cytoskeletal alterations in myeloma cells following Wnt-3a treatment. Microscopy of OPM-2 (left) or H929 cells (right). (a, b) con-CM-treated cells; (c, d) cells treated with Wnt-3a-CM for 4 h.; e, f F-actin staining of con-CM-treated cells; (g, h) F-actin staining of Wnt-3a-CM-treated cells

(Wodarz and Nusse, 1998; Miller et al., 1999) or Rho activation (Habas et al., 2001; Winter et al., 2001). We were unable to detect a calcium flux associated with Wnt-3a stimulation, but activation of Rho from the GDP to the GTP form could be directly demonstrated in a pull-down assay using GST-Rho binding domain (Figure 6). This activation was not blocked by Dkk-1. Moreover, treatment with Y27632, an inhibitor of Rhoassociated kinase, completely inhibited the morphological changes (Figure 7).

Discussion Wnt signaling has been demonstrated to have profound effects in embryogenesis, cell growth and certain forms Oncogene

kk -2 /C D kk -2 /W nt -3 a

D

kk -1 /C D kk -1 /W nt -3 a

D

V

ec to r/C V ec to r/W nt -3 a

0

Figure 5 Dkk-1 and Dkk-2 inhibition of Wnt-3a-induced bcatenin stabilization and activation of LEF/TCF transcription. H929 cells were treated with Dkk-1 (a) and Dkk-2 CM (b) for the indicated times, then stimulated with Wnt-3a or con-CM. Cell lysates were resolved on 8% SDS–PAGE gels and analysed with antibodies directed against pan-b-catenin (a, b top panels) or nonphosphorylated (active) b-catenin (a, b middle panels). Blots were stripped and reprobed with anti-b-tubulin (a, b bottom panels). (c) Effect of expression of Dkk-1 and Dkk-2 on Wnt-3ainduced transcription of LEF/TCF in OPM-2 cells. OPM-2 cell clones expressing PEF6/V5-His-TOPO (vector), PEF6/V5-HisTOPO-Dkk-1 (Dkk-1) or PEF6/V5-His-TOPO-Dkk-2 (Dkk-2) were transiently cotransfected with wild-type (TOPFLASH) LEF/ TCF luciferase reporter construct. Cells were then treated with Wnt-3a or con-CM (c). Values represent the mean of six independent transfections for each construct

of cancer. However, little is known about potential Wnt signaling in mature lymphocytes or lymphoid neoplasia. Expression of Wnt-16 following a chromosomal translocation has been reported in pre-B-cell leukemia (McWhirter et al., 1999). Wnts have additionally been shown to play a role in hematopoiesis in which Wnt-11 induced bone marrow cells to develop into a variety of different lymphoid cell types (Brandon et al., 2000). In contrast, Wnt-3a was shown to decrease the production of B and myeloid cells in bone marrow cultures. The observed suppression required stromal cells implying an


1541

a Wnt-3a Dkk-1

0

15

30

60

180 60

min

+ GTP-RhoA

Total R hoA b

Wnt-3a Dkk-1

15

30

60

180 60

min

+ GTP-RhoA Total RhoA

Figure 6 Wnt-3a induces formation of Rho-GTP in myeloma cells. Lysates from OPM-2 (a) and H929 (b) cells treated for the indicated times with Wnt-3a-CM, con-CM or Dkk-1-CM followed by Wnt-3a-CM were incubated with GST-RBD glutathione beads as described in Materials and methods. GTP-bound Rho protein was analysed by Western blotting using RhoA-specific antibody

indirect mechanism (Yamane et al., 2001). This pathway has been further implicated in B- and T-cell development as LEF-1 (a member of the TCF family of transcription factors that forms complexes with bcatenin) null mice have defects in pro-B-cell proliferation and survival (Reya et al., 2000) and TCF-1 null mice are defective in thymocyte differentiation (Ioannidis et al., 2001). However, it should be noted that Wnt signaling has only been directly demonstrated in pro-B cells (Reya et al., 2000). In the present studies, we have identified functional Wnt signaling pathways in myeloma cells that are absent in B-cell lymphomas. All myeloma lines expressed mRNA corresponding to multiple Wnt receptors (Table 1) as well as either or both of the Wnt coreceptors LRP 5 and 6. In striking contrast, only Fz-3 was expressed at comparable levels in a series of B-cell lymphomas and none of these expressed either of the LRP coreceptors. The absence of LRP coreceptor required for canonical Wnt signaling (Pinson et al., 2000; Tamai et al., 2000; Wehrli et al., 2000) suggests that B-cell lymphomas are not likely to be able to signal through a canonical Wnt/b-catenin pathway, but does not rule out the possibility of Wnt signaling through noncanonical pathways. The lack of receptors may further represent a stage-specific difference between mature B and plasma cells. Clearly, similar studies using normal B and plasma cells will be required to ascertain the generality of the distinct receptor expression patterns observed in B lineage tumors. Functional Wnt/b-catenin signaling involves downstream effectors including Dvl proteins that disrupt the

b-catenin degradation complex. Wnt treatment of myeloma cells led to increases in protein levels of Dvl2 and phosphorylation of Dvl-3 (Figure 1) accompanied by stabilization of b-catenin (Figure 2). In B lymphomas no stabilization was observed and the markedly lower endogenous levels appeared to be continuously degraded. b-catenin accumulation in myeloma cells was inhibited by sFRP-1, a putative antagonist that prevents Wnt binding to Fz receptors. Wnt-mediated accumulation of b-catenin could also be mimicked by LiCl, a known inhibitor of GSK3b in the b-catenin degradation complex. Accumulation of b-catenin leads to nuclear translocation and binding to members of the TCF/LEF families to form transcription-activating complexes (Korinek et al., 1997). RT–PCR analysis (Figure 3) revealed that the only member of the TCF/LEF families expressed in myeloma or B cells was LEF-1 and that functional-activating complexes could be formed as determined by a luciferase reporter construct assay. It should be noted that earlier reports using Northern blot analysis have suggested that TCF/LEF family members are not expressed in mature B cells (Oosterwegel et al., 1993; Van de Wetering et al., 1996). It is possible that the greater sensitivity of RT–PCR analysis may account for the differences between these previous reports and the current studies. Taken together, the present experiments reveal the presence of a functional canonical Wnt/ b-catenin pathway in myeloma cells capable of initiating transcriptional activation upon exposure to Wnt proteins. Treatment of myeloma lines with Wnt-3a resulted in pronounced morphological changes wherein cells rapidly became fibroblast-like in appearance, formed filopodia and attached to culture plates (Figure 4). The morphological alterations were accompanied by rearrangement of the actin cytoskeleton with actin protein becoming localized at the tips of filopodia forming attachment points. The observation that LiCl stabilized b-catenin, but did not induce morphological changes, suggested the possibility that this pathway was not sufficient (and possibly not necessary) for cytoskeletal rearrangement. Since sFRP-1 completely inhibited the changes, any cellular signaling pathways participating in this phenomenon must still be activated through Fz receptors. To determine whether LRP coreceptors were also required, myeloma cells were exposed to conditioned media containing Dkk-1 or -2 proteins that have been shown to inhibit signaling specifically through the LRP receptors (Mao et al., 2001; Semenov et al., 2001). Dkk and sFRP regulation of Wnt signaling have been similarly compared in kidney model systems (Y Endo, unpublished). Both Dkk-1 and -2 inhibited b-catenin stabilization (Figure 5), but neither prevented morphological changes indicating that LRP coreceptors do not participate in the required pathway. Subsequent experiments demonstrated that Rho was activated by Wnt-3a treatment, and that the morphological alterations could be completely blocked by the Rho-associated kinase inhibitor Y27632 (Figures 6 and 7). This compound inhibits two Rho-associated kinases (ROCK-I and -II) downstream of Rho and prevents stress fiber formation in 3T3 cells (Ishizaki et al., 2000). Furthermore, Rho Oncogene


1542

OPM-2

H929

a

b

c

d

e

f

Figure 7 Inhibition of RhoA-associated kinase blocks Wnt-3a-induced morphology alterations. OPM-2 (left) and H929 (right) were cultured in serum-free medium for 12 h. Cells were then treated with con-CM (a, b) or Wnt-3a-CM (c, d, e, f) for 6 h. Cells in (e) and (f) were preincubated in media containing Y27632 (10 mm) for 1 h prior to Wnt-3a treatment

activation was not inhibited by Dkk-1. Thus, Rho activation is critical for the observed cytoskeletal rearrangement and is independent of LRP coreceptors. Recent studies have demonstrated that Rho can be activated by both Wnt-1 and Wnt-11 and that this pathway, termed Wnt/RhoA, is necessary for proper Xenopus gastrulation and cell polarity (Habas et al., 2001; Winter et al., 2001). Cytoskeletal alterations in response to Wnt-3a have also previously been noted in epithelial cells (Shibamoto et al., 1998). Such changes suggest alterations in motility and likely reflect rearrangements that will alter cell–cell or cell–substrate interactions. Experiments to evaluate changes in adhesion molecules and migratory properties associated with Wnt signaling in myeloma cells are Oncogene

currently being undertaken and should provide additional information concerning the biological consequences associated with the observed cytoskeletal rearrangements. It is possible that signaling through Fz receptors may play a significant role in the interaction between myeloma cells and the bone marrow microenvironment. This hypothesis becomes even more attractive in the light of recent studies suggesting a key role for Wnt signaling in bone matrix deposition associated with osteoblasts (Gong et al., 2001; Little et al., 2002). Given the well-documented relation between myeloma cells and the induction of osteoclasts resulting in highly debilitating bone lesions in myeloma development (Bataille et al., 1992; Niesvizky and Warrell, 1997), it will be intriguing to examine the possibility that Wnts


1543

may also play a role in this aspect of bone metabolism. Irrespective of potential biological implications, the current studies demonstrate, for the first time, two functional Wnt signaling pathways in mature lymphocytes and suggest that these pathways may be important in both normal and abnormal lymphocyte development.

Materials and methods Cell lines and CM Human MM cell lines ANBL6, Delta-47, H929, MM144, OPM-2, RPMI8226, and B-cell lines Daudi, Namalwa and ST486 (kindly provided by Drs Michael Kuehl and Kishor Bhatia) were cultured in RPMI1640 containing 10% fetal calf serum (FCS), 2 mmol/l l-glutamine, 0.1% penicillin (100 u/ ml), streptomycin (100 mg/ml). CM or con-CM was prepared from Wnt-3a-producing L cells (stably transfected with Wnt3a cDNA kindly provided by Dr Shinji Takada) or control L cells grown to confluency in EMEM medium supplemented with 10% FCS after which the medium was replaced with serum-free EMEM. Culture supernatants were collected after 72 h and designated Wnt-3a-CM and con-CM, respectively. sFRP-1 was prepared as described previously (Uren et al., 2000).

con-CM for 3, 6, 9 and 12 h or for the indicated time points. For inhibition studies, cells (1X107) were grown in serum-free media for 12 h and pretreated with Y27632, or Dkk-1 or -2 CM, or con-CM for 1 h. Fresh media consisting of a 1 : 1 mixture of inhibitor-containing and Wnt-3a-containing CM (or con-CM) was then added for an additional 3 h or indicated times. Following treatment, cells were lysed as previously described (Qiang et al., 2002). Cell lysates were separated on sodium dodecyl sulfate–polyacrylaminde gels (SDS–PAGE) and transferred to Immobilon polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA). Immunoblotting was performed using indicated antibodies. Immunoprecipitation and phosphatase treatment Whole cell lysates (500 mg protein) from cells treated with Wnt3a-CM or con-CM for 3 h were prepared as described above and precleared by incubation with protein G-Sepharose. Lysates were incubated with anti-Dvl-3 antibodies for 2 h at 41C. Immune complexes were then adsorbed to protein Gsepharose beads and washed three times. Phosphatase treatment of immune complexes was performed as described (Semenov and Snyder, 1997). Treated complexes were subjected to immunoblot analysis with anti-Dvl-3 antibody. Rho-GTP pull-down assay

Reagents and antibodies Antiactive-b-catenin antibody was a gift from Dr Hans Clevers (University Medical Center Utrecht, Utrecht, the Netherlands). The following antibodies were purchased from the indicated sources: anti-a-, b- and g-catenin, Transduction Laboratories (Lexington, KY, USA); anti-Dvl-1, -Dvl-2, -Dvl3, -RhoA, Santa Cruz Biotechnology, Inc. (Beverly, MA, USA); anti-V5, Invitrogen-Life Technologies, Inc. (Carlsbad, CA, USA). Horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies were from Transduction Laboratories. The Rho-associated kinase inhibitor, Y27632, was purchased from CalBiochem (San Diego, CA, USA) and dissolved in water. Constructs, transfectants and reporter gene assay Dkk-1 and -2 cDNAs were derived by PCR amplification and ligated into the expression plasmid PEF6/V5-His-TOPO (Y Endo, unpublished) containing a V5 tag sequence. Constructs were transfected into OPM-2 cells using Lipofectamine (Invitrogen-Life Technologies, Inc.) according to the manufacturer’s instructions. Clonal cell lines were generated by limited dilution in a growth media containing 2.5 mg/ml blasticidin. Dkk-1 and Dkk-2 conditioned media were harvested from positive clones grown in serum-free medium for 72 h and recombinant protein detected by immunoblotting with anti-V5. For reporter assays, plasmids with wild-type (TOPFLASH) or mutated (FOPFLASH) LEF/TCF binding sites (Korinek et al., 1997) were transiently transfected into OPM-2 cells using Lipofectamine. Following transfection, cells were exposed to Wnt-3a-CM or -con-CM for 24 h prior to luciferase assay. pSV-b-galactosidase vector was used to normalize for transfection efficiency. Luciferase values were determined following cell lysis in Glo lysis buffer. Luciferase activity was measured with the Bright-Glo luciferase assay kit (Promega). Immunoblot analysis Cells (1 107) were grown in serum-free media for 12 h, and then incubated with lithium chloride (LiCl), Wnt-3a-CM or

Measurement of GTP-bound Rho was performed as previously described (Ren et al., 1999). Briefly, RhoA-binding domain as a GST-fusion protein (GST-RBD) was purified using GST beads. GST-RBD was used to precipitate GTPbound Rho from cells lysed in 50 mm Tris, pH 7.2, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 500 mm NaCl, 10 mm MgCl2 and a cocktail of protease inhibitors. Precipitated Rho-GTP was detected by Western blotting using monoclonal anti-Rho antibody. RT–PCR analysis First-strand cDNA synthesis was performed using ProSTARt Ultro HF RT–PCR Kit (Stratagene) primed with random hexamer in a 50 ml reaction mixture containing 1 mg total RNA. The first-strand cDNA mixture (1 mg) was subjected to PCR using PCR Kit (Applied Biosystem) in a 50 ml volume according to the manufacturer’s instructions. All PCR reactions were initiated with a first cycle at 941C for 3 min and a final cycle at 721C for 10 min. Reactions were carried out for 40 cycles under the following conditions: Fz primers – 941C/ 30 s, 601C/45 s, 721C/1 min; LRP-5 and -6, LEF-1, TCF-1, -3, and -4 primers – 941C/15 s, 601C/30 s, 721C/45 s. Primer sequences were as follows: Fz-d, 50 -NNNGAATTCTAYCCNGARMGNCCNAT-30 / 50 -NNNAAGCTTNGCNGCNARRAACCA-30 ; Fz-1, 50 -GA ACTTTCCTCCAACTTCATGGC-30 /50 -CATTTCCATTTTACAGACCGG-30 ; Fz-2, 50 -GGTGAGCCAGCACTGCAAGAG-30 /50 -CCTAAAAGTGAAATGGTTTCGATCG-30 ; Fz-3, 50 -GCTGTACTCACAGTTAACATG-30 /50 -GCTAAAATACCCTTGCTGATTT-30 ; Fz-4, 50 -TGCCTTTTCAGGGCAAAGTG-30 /50 -ACAGGAAGAGATTTATGGAATG-30 ; Fz-5, 50 -TACCCAGCCTGTCGCTAAAC-30 /50 AAAACCGTCCAAAGATAAACTGC-30 ; Fz-6, 50 -ACATCTCTGCTTGTTCAC-30 /50 -GATCTGTGAAATTCCTAA-30 ; Fz-7, 50 -GTTTGGATGAAAAGATTTCAGGC-30 /50 -GACCACTGCTTGACAAGCACAC-30 ; Fz-8, 50 -ACAGTGTTGATTGCTATTAGCATG-30 /50 -GTGAAATCTGTGTATCTGACTGC-30 ; Fz-9; 50 -CCCTAGAGACAGCTG ACTAGCAG-30 /50 -CGG GGGTTTATTCCAGTCACAGC-30 ; Fz-10, 50 -ACACGTCCA Oncogene


1544 ACGCCAGCATG-30 /50 -ACGAGTCATGTTGTAGCCGATG30 ; LRP5, 50 -GACATCTACAGCCGGACACTG -30 , 50 -CACAAGTCAGCAGGTTCTGCAGG-30 ; LRP6, 50 -GATTATCCAGAAGGCATGGCAG-30 /50 -CAATCACCATGCGGTTGATGGC-30 ; LEF1, 50 -CCAGCTATTGTAACACCTCA30 /50 -TTCAGATGTAGGCAGCTGTC -30 ; TC-F1, 50 -TGACCTCTCTGGCTTCTACT-30 /50 -TTGATGGTTGGCTTCTTGGC-30 ; TCF3, 50 -AGGAAATCACCAGTCACCGT-30 /50 GTACTTGGCCTGTTCTTCTC-30 ; TCF4, 50 -TTCAAAGACGACGGCGAACAG-30 /50 -TTGCTGTACGTGATAAGAGGCG-30 ; b-actin: 50 -CCACTGGCATC GTGATGGAC-30 /50 GCGGATGTCCCACGTCACACT-30 ; GAPDH, 50 -CTGAGAACGGGAAGCTTGTCAT-30 /50 -CAGCCTTCTCCATGGTGGTGAAGA-30 . Subcloning of PCR fragments and DNA sequence analysis PCR fragments were separated on 1.2% agarose gels and purified using QIAEX Gel Extraction Kit (QIAGEN Inc.). The PCR fragments were subcloned using TOP-TA cloning

vector according to the manufacturer’s instructions (Invitrogen). Candidate clones from each PCR fragment were subjected to DNA sequence analysis using vector M13 primers. Sequences were determined at the NCI DNA Sequencing Minicore. Data were analysed using Sequencert 3.1 software and compared with Gene Bank using NCBI BLAST (http://www.ncbi.nlm.nih.gov/blast/). Fluorescence microscopy analysis Cells were incubated in 24-well plates with Wnt-3a-CM or conCM for varying times and photographed using an inverted phase contrast microscope equipped with a cooled chargecouple device (CCD) camera. For the detection of F-actin, cells with or without Wnt-3a-CM were cultured in a two-well chamber for 12 h, fixed with 3.7% formaldehyde solution, permeabilized in 201C acetone for 3 min and stained with Alexa Fluor 488 phalloidin (Molecular Probes). The fluorescence intensities of F-actin were detected using an excitation wavelength of 488 nm.

References Bataille R, Chappard D and Klein B. (1992). Hematol. Oncol. Clin. North Am., 6, 285–295. Bergui L, Schena M, Gaidano G, Riva M and CaligarisCappio F. (1989). J. Exp. Med., 170, 613–618. Brandon C, Eisenberg LM and Eisenberg CA. (2000). Blood, 96, 4132–4141. Ferlin M, Noraz N, Hertogh C, Brochier J, Taylor N and Klein B. (2000). Br. J. Haematol., 111, 626–634. Ge NL and Rudikoff S. (2000). Blood, 96, 4091–4095. Gong Y, Slee RB, Fukai N, Rawadi G, Roman-Roman S, Reginato AM, Wang H, Cundy T, Glorieux FH, Lev D, Zacharin M, Oexle K, Marcelino J, Suwairi W, Heeger S, Sabatakos G, Apte S, Adkins WN, Allgrove J, ArslanKirchner M, Batch JA, Beighton P, Black GCM, Boles RG, Boon LM, Borrone C, Brunner HG, Carle GF, Dallapiccola B, Paepe AD, Floege B, Halfhide ML, Hall B, Hennekam RC, Hirose T, Jans A, Ju¨ppner H, Kim CA, KepplerNoreuil K, Kohlschuetter A, LaCombe D, Lambert M, Lemyre E, Letteboer T, Peltonen L, Ramesar RS, Romanengo M, Somer H, Steichen-Gersdorf E, Steinmann B, Sullivan B, Superti-Furga A, Swoboda W, Boogaard M, Hul WV, Vikkula M, Votruba M, Zabel B, Garcia T, Baron R, Olsen BR and Warman ML. (2001). Cell, 107, 513–523. Habas R, Kato Y and He X. (2001). Cell, 107, 843–854. Hilbert DM, Kopf M, Mock BA, Kohler G and Rudikoff S. (1995). J. Exp. Med., 182, 243–248. Ioannidis V, Beermann F, Clevers H and Held W. (2001). Nat. Immunol., 2, 691–697. Ishizaki T, Uehata M, Tamechika I, Keel J, Nonomura K, Maekawa M and Narumiya S. (2000). Mol. Pharm., 57, 976– 983. Korinek V, Barker N, Morin P, van Wichen D, de Weger R, Kinzler K, Vogelstein B and Clevers H. (1997). Science, 275, 1784–1787. Lattanzio G, Libert C, Aquilina M, Cappelletti M, Ciliberto G, Musiani P and Poli V. (1997). Am. J. Pathol., 151, 689– 696. Lee JS, Ishimoto A and Yanagawa S. (1999). J. Biol. Chem., 274, 21464–21470. Little RD, Carulli JP, Del Mastro RG, Dupuis J, Osborne M, Folz C and Manning SP. (2002). Am. J. Hum. Genet., 70, 11– 19. Oncogene

Lu ZY, Brailly H, Wijdenes J, Bataille R, Rossi JF and Klein B. (1995). Blood, 86, 3123–3131. Mao B, Wu W, Li Y, Hoppe D, Stannek P, Glinka A and Niehrs C. (2001). Nature, 411, 321–325. McWhirter JR, Neuteboom ST, Wancewicz EV, Monia BP, Downing JR and Murre C. (1999). Proc. Natl. Acad. Sci. USA, 96, 11464–11469. Miller J, Hocking A, Brown J and Moon R. (1999). Oncogene, 18, 7860–7872. Niesvizky R and Warrell RJ. (1997). Cancer Invest., 15, 85–90. Oosterwegel M, van de Wetering M, Timmerman J, Kruisbeek A, Destree O and Meijlink F. (1993). Development, 118, 439– 448. Peifer M and Polakis P. (2000). Science, 287, 1606–1609. Pinson KI, Brennan J, Monkley S, Avery BJ and Skarnes WC. (2000). Nature, 407, 535–538. Polakis P. (2000). Genes Dev., 14, 1837–1851. Qiang Y, Kopantzev E and Rudikoff S. (2002). Blood, 99, 4138–4146. Ren X, Kiosses W and Schwartz M. (1999). EMBO J., 18, 578–585. Reya T, O’Riordan M, Okamura R, Devaney E, Willert K, Nusse R and Grosschedl R. (2000). Immunity, 13, 15–24. Semenov MV and Snyder M. (1997). Genomics, 42, 302–310. Semenov M, Tamai K, Brott B, Kuhl M, Sokol S and He X. (2001). Curr. Biol., 11, 951–961. Shibamoto S, Higano K, Takada R, Ito F, Takeichi M and Takada S. (1998). Genes Cells, 3, 659–670 Tamai K, Semenov M, Kato Y, Spokony R, Liu C, Katsuyama Y, Hess F, Saint-Jeannet JP and He X. (2000). Nature, 407, 530–535. Uren A, Reichsman F, Anest V, Taylor WG, Muraiso K, Bottaro DP, Cumberledge S and Rubin JS. (2000). J. Biol. Chem., 275, 4374–4382. Van de Wetering M, Castrop J, Korinek V and Clevers H. (1996). Mol. Cell Biol., 16, 745–752. Wehrli M, Dougan ST, Caldwell K, O’Keefe L, Schwartz S, Vaizel-Ohayon D, Schejter E, Tomlinson A and DiNardo S. (2000). Nature, 407, 527–530. Winter C, Wang B, Ballew A, Royou A, Karess R, Axelrod J and Luo L. (2001). Cell, 105, 81–91


1545 Wodarz A and Nusse R. (1998). Annu. Rev. Cell Dev. Biol., 14, 59–88. Yamane T, Kunisada T, Tsukamoto H, Yamazaki H, Niwa H, Takada S and Hayashi SI. (2001). J. Immunol., 167, 765–772. Yanagawa S, van Leeuwen F, Wodarz A, Klingensmith J and Nusse R. (1995). Genes Dev., 9, 1087–1097.

Zhang XG, Gaillard JP, Robillard N, Lu ZY, Gu ZJ, Jourdan M, Boiron JM, Bataille R and Klein B. (1994). Blood, 83, 3654–3663. Zhurinsky J, Shtutman M and Ben-Ze’ev A. (2000). J. Cell Sci., 113, 3127–3139.

Oncogene