(CANCER RESEARCH 58. 2158-2165. May 15. IWK] Glial Cell Line-derived Neurotrophic Factor/Neurturin-induced Differentiation and Its Enhancement by Retinoic Acid in Primary Human Neuroblastomas Expressing c-Ret, GFRa-1, and GFRa-21
1(iinoio Ilisliiki, Yoshinori Nimura, Eriko Isogai, Kazuhiro Kondo, Shingo Ichimiya, Yohko Nakamura, Insilinoli Ozaki, Shigeru Sakiyama, Masayoshi Hîrose,Naohiko Seki, Hideyo Takahashi, Naomi Ohnuma, Masahiro Tanabe, and Akira Nakagawara2
Divisions itf Biochemistry ¡T.H.. Y. Ni.. E. !.. K. K.. S. I., Y. Na.. T. O.. S. S.. A. N.] and Clinical Examinalinn ¡M.H.¡.Chiba Cancer Center Research Institute. Chiha 260-X7I7: ¡jiruinilory of Gene Structure I, Kiizusa DNA Research Institute, C/tiha 292-0812 ¡N.S./; ami Department tif Pediatrie Surgery. Chiba University Settimi of Medicine. Chiba 260- «670¡T.H.. H. T.. N. O.. M. T.]. Japan
ABSTRACT We and other investigators have shown that TRK-A. a high-affinity receptor for NGF, is expressed in neuroblastomas with favorable Ncuroblastomas often undergo spontaneous differentiation and/or re prognosis and may regulate the differentiation and/or regression of the gression in vivo, which is at least partly regulated by the signals through tumor (5-8). On the other hand, TRK-B, a receptor for brain-derived neurotrophins and their receptors. Recently, glial cell line-derived neuro- neurotrophic factor (BDNF) and NT-4. is frequently expressed in trophic factor i<.DM i and a second family member, neurturin (NTN), have been found to mediate their signals by binding to a heterotetramcric tumors with N-/HVCamplification, which may promote cell survival complex of c-Ket tyrosine kinase receptors and glycosylphosphatidylinosi- and increase the invasive activity in an autocrine or paracrine manner tol-linked proteins, GFRa-1 (GDNFR-a) or GFRa-2 (TrnR2/GDNFR-/3/ (9). NTNR-0/RKTL2). Here, we studied the effect of GDNF and NTN on Several secretory proteins that are not structurally related to the human ncurohlastomas in the short-term primary culture system, as well NTs are also able to support the survival of particular neurons. These as the expression of c-Ret, GFRa-l, GFRa-2, GDNF, and NTN. GDNF (1-100 ng/ml) induced morphological differentiation in 34 of 38 primary include ciliary neurotrophic factor (10), basic fibroblast growth factor neuroblastomas and an accompanying increase in c-Fos induction. These (11), leukemia-inhibitory factor (12), midkine (13), pleiotrophin (13, effects were markedly enhanced by treatment with 5 ;<\i all-/ran.v-retinoic 14), and GDNF (15). The second GDNF family member, NTN, has acid. Although GDNF alone induced a rather weak differentiation inde also been cloned (16). GDNF and NTN are distantly related members pendent of the disease stages, the enhancement of neurite outgrowth of the transforming growth factor ßfamily. GDNF is a potent survival induced by treatment with both GDNF and all-fra/i.v-rctinoic acid was factor for midbrain dopaminergic neurons and motoneurons as well as significantly correlated with younger age (less than I year; P = 0.0039), sensory and autonomie neurons of the peripheral nervous system non-stage 4 diseases (/' = 0.0023), a single copy of N-m.vc (/> = 0.027), and (17-22). The functional receptor for GDNF has recently been identi high levels of TRK-A expression (/' = 0.0062). To examine the expression fied as a heterotetrameric complex of a proto-oncogene. Ret. and a levels ofGFRat-I, we cloned a short form of the human GFRa-1 gene with a 15-bp deletion by screening a human adult substantia nigra cDNA novel glycosylphosphatidylinositol-linker protein, GFRa-1 (GD library. Many primary neuroblastomas expressed c-Ret, GFRa-I, and NFR-a; Refs. 23-26). In addition, a novel glycosylphosphatidylinosi- (ìFRa-2as well as their ligands, GDNF and NTN, suggesting the presence tol-linker receptor. GFRa-2 (TrnR2/GDNFR-0/NTNR-a/RETL2), of a paracrine or autocrine signaling system within the tumor tissue. The has also been cloned (27-31 ). effect of NTN on primary culture cells of neuroblastoma was similar to Ret is a receptor tyrosine kinase that is expressed in several cell that of GDNF. These imply that the GDNF(NTN)/c-Ret/GFRa-l(GFR«-2) lineages including sympathetic neurons and adrenal chromaffin cells, signaling may have an important role in regulating the growth, differen from which neuroblastoma originates (32). Ret signaling is necessary tiation, and cell death of neuroblastomas. for the differentiation of neuronal precursor cells to mature autonomie neurons. In familial medullary thyroid carcinomas and multiple en docrine neoplasias type 2A and 2B, germ-line mutations of Ret were INTRODUCTION identified (33, 34), whereas a loss of function mutations of Ret has Neuroblastoma is one of the most common pediatrie tumors and been observed in Hirschsprung's disease, a congenital aganglionosis originates from the sympathoadrenal lineage derived from the neural of the intestine (35, 36). However, in neuroblastomas, the role of Ret crest ( 1). The prognosis of neuroblastoma patients over 1 year of age has been unclear, despite the fact that the Ret proto-oncogene is tends to be poor, whereas that of patients under 1 year of age is usually frequently expressed in primary neuroblastomas and cell lines (8, 37, favorable, with tumors having a potential to differentiate or to regress 38). The discovery that GDNF binds to the receptor complex of Ret spontaneously (2). Recent studies have revealed that growth, differ and GFRa-1 has enabled us to study the functional role of GDNF entiation, and programmed cell death of the developing neural crest- derived cells are strongly regulated by NTs' and their receptors (3, 4). signaling in neuroblastoma. In the present study, we cloned a variant form of human GFRa-1 with a 15-bp deletion (GFRa-l-15del) in the open reading frame. The Received 10/29/97; accepted 3/18/98. genes encoding the receptor complex of c-Ret and GFRa-1 or The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with GFRa-2 and their ligands, GDNF and NTN, were frequently ex 18 U.S.C. Section 1734 solely to indicale this fact. 1Supported in part by a grant-in-aid from the Ministry of Health and Welfare for a pressed in primary neuroblastoma tissues. GDNF as well as NTN New Comprehensive 10-Year Strategy for Cancer Control, Japan. induced morphological differentiation of many neuroblastomas in the " To whom requests for reprints should be addressed, at Division of Biochemistry. Chiha Cancer C'enter Research Institute. 666-2 Nitona. Chuoh-ku. Chiba 260-8717. Japan. primary culture system, which was markedly enhanced by treatment Phone: 81-43-264-5431; Fax: 81-43-265-4459; E-mail: [email protected]. with RA. However, the grade of morphological differentiation in ' The abbreviations used are: NT. neurotrophin: NGF. nerve growth factor: GDNF. duced by the concomitant treatment with GDNF and RA was much glial cell line-derived neurolrophic factor; NTN. neurturin; RA. all-mm.v-retinoic acid: RT-PCR, reverse transcriplion-PCR: GAPDH. glyceraldehyde-3-phosphate dehydro- weaker in aggressive neuroblastomas than in nonaggressive tumors. genase. Our data suggest that the GDNF/NTN signaling may play an impor- 2158
Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. GDNF/NTN-INDUCED DIFFERENTIATIONIN NEUROBLASTOMA
Table 1 Expression of TRK-A and CDNF receptors and response to NGF, RA, and GDNF in iti primary neurobliislomus in priman- culture receptors'1'P^tp/lpcj» of GDNF toGDNF (nNGF^ CaseOrigin"12345678910111213141516MediastinumAdrenalAdrenalAdrenalAbdomenMediastinumAbdomenAbdomenPelvisAdrenalAdrenalAdrenalAdrenalAbdomenAbdomenAdrenal1 Stage ploidy''AAAAAAAAATAADAAAATAAAAADAADDADDTDDDDDDTRK-Aexpression'HHHHHHHHHHHHHMHHHHHHHHHHHHLLHLLLLLLNDNDLExpression (mo)117g87779156779077g973109711747712g12Ig533614314Ig27MSh++++++++++—++—+++++—+++——+—+—-———-——--N-mvr'(copy)AmplifiedAmplifiedAmplified1AmplifiedAmplified1AmplifiedAmplifiedDNAGFRa-l-I5del4+c-Ket GFRa-\ RA2+1 GDNF3+4+3+2+3+3+2+4+3+3+3+4+4+4+4+4+1 +3+ + + +3+3+002+03+2+01 +3+ + + +3 +3+ + + +02+2+1+3+2+2+1 +2+ + + +3+ + + +3+ + + +4+ + + +2+ + + +3+ + +2+ + +4+ + + +1 +2 +3+ + + +2+1 +01 +4+ + + +4+ + + +2+2+0Toxic01+2+2+2+2+3+1 +3+ + + +4+ + + 7 2Abdomen18 +3+ + + +2+3+3+3+1+2+4+3+3+01+1 2Abdomen19 +3+ + 2Mediastinum20 +2+ + + 2Mediastinum21 +1+ + + +1 +2+1 2Mediastinum22 +2+ + + +001+01 2Mediastinum23 +3-1- + + +2+2+1 Mediastinum244s +4+ ++ 3Adrenal25 +0 + + 3Pelvis26 +2+ + + +02+2+2+1+1+2+1 3Pelvis27 +1+ + + +Toxic00Toxic03 4Adrenal28 +1+ + + 4Adrenal29 +3+ + + 4Adrenal30 +2+ + + +2+04+1 4Adrenal31 +0 + + 4Adrenal32 +2+ + + 4Abdomen33 +0 + + +01 4Abdomen34 +0 -1- + +1 +3 4Adrenal35 +0 H- + +Toxic2+0ToxicResponse+01 +03+2+2 4Adrenal36 +3+ + + 4Adrenal37 ND1+ ND ND +1 4Adrenal38 +0 + + +1 4 AdrenalAge + +• +RA''2 ++ + " Abdomen, abdominal sympathetic ganglion; adrenal, adrenal gland. h +, found by a mass screening system for neuroblastomas at the age of 6 months; —¿ .sporadicneuroblastoma. ' Amplified, 210 copies. A, anueploid; D. diploid: T. tetraploid. ' H. high TRK-A expression (higher than the expression level of the CHP-901 neuroblastoma cell line detected by Northern blot analysis); L. low TRK-A expression (lower than CHP-901); ND, not done. ^0, no response; 1+. mild response; 2 + . moderate response; 3 + , strong response; 4 + . very strong response (see "Material and Methods"). * +, detectable by RT-PCR; ND, not done. h Toxic, toxic effect.
tant role in regulating the differentiation and survival of neuroblasto- LA-N-5, SH-SY5Y, OAN, NGP, NLF, CHP901. TGW. and NMB) were mas in vivo in an autocrine and/or paracrine manner. described previously (9. 40). and the cells were cultured in the RPMI 1640 (Life Technologies. Inc.. Gaithersburg. MD) with 107r heat-inactivated fetal bovine serum and 100 fig/ml kanamycin at 5% CO2/959f air at 37°C. MATERIALS AND METHODS Primary Cultures of Neuroblastomas. RPMI 1640 with 10% heat-inac tivated fetal bovine serum. 100 /¿g/mlOP1 (Sigma Chemical Co.. St. Louis, Neurotrophic Factors and cDNA Probes. Mouse NGF was a generous MO), and 100 jig/ml kanamycin was used as a standard culture medium. Fresh gift from Eugene M. Johnson. Recombinant human GDNF was purchased tumor tissue (0.2-0.5 g) was cut into small pieces and incubated in the from Pepro Tech, Inc. (Rocky Hill. NJ), dissolved in 1x PBS with 0.1% BSA. and stored at —¿ 80°Cuntil use. Recomhinant mouse NTN was a gift from standard medium containing 500 units/ml collagenase (Sigma Chemical Co.) for 30 min at 37°Cat 5% CCK/95% air on a 100 X 20-mm polystyrene tissue Genentech, Inc. (South San Francisco, CA). The following cDNA probes were generous gifts: the N-mvc probe was from J. Michael Bishop; the human culture dish (Becton Dickinson. Lincoln Park, NJ). The tumor cells were TRK-A probe was from Luis Parada; the human c-Ret probe was from Minako dispersed using a plastic syringe with an 18-gauge needle, centrifuged at Nagao; and the probes for c-Fos and NGFI-A were from Jeffrey D. Milbrandt. 250 x g for 5 min, resuspended in 10 ml of standard medium, and filtered Primary Tumor Samples and Cell Lines. All tumor samples were ob using an autoclaved metallic tea filter. The supernatant was layered gently on tained from surgical biopsies or resections of the untreated tumors except for top of 5 ml of lymphocyte separation medium (Organon Teknika Co., Durhan, case 35, whose tumor cells were collected by bone marrow aspiration. The NC), and centrifuged at 1000 X g for 10 min. The cells at the interface were samples obtained were stored at 4°Cand immediately sent to the Chiba Cancer collected, washed once or twice with standard medium, and then resuspended Center Research Institute. A total of 38 samples ( 16 stage 1, 6 stage 2. 1 stage in a proper amount of the same medium. The viable nucleated cells were 4s, 3 stage 3, and 12 stage 4) were available for the primary culture testing the counted after staining with trypan blue (Sigma Chemical Co.). To evaluate the effect of GDNF and RA. Because our study had already started before NTN effect of neurotrophic factors and other reagents on primary neuroblastoma cells, 2-5 X IO5 cells in 1 ml of standard medium were incubated in a well was identified, the effect of NTN was examined using another set of 17 primary neuroblastomas. All diagnoses of neuroblastoma were confirmed by coated with type I collagen on 24-well tissue culture plates (Becton Dickin the histological assessment of surgically resected tumor specimens. The tumors son). The cells were treated with or without NGF (100 ng/ml). GDNF (100 were staged according to the International Neuroblastoma Staging System ng/ml), or NTN ( 100 ng/ml) in the absence or presence of 5 /J.MRA (Sigma (39). The characteristics of neuroblastoma cell lines (SMS-KCN. RTBM-1. Chemical Co.). The cells were incubated in 5% CO2/95% air at 37"C for at 2159
Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. GDNF/NTN-INDUCED DIFFERENTIATION IN NEUROBLASTOMA
',<•- ./:|Y>
Fig. 1. Effect of GDNF ( 100 ng/ml) and RA (5 JÃŒM) on surgically resected ncurohlastoma cells in primary culture. A-f-\ primary culture cells obtained from a favorable-prognosis neuroblastoma (case 16, stage 1) on day 5. G-J, primary culture cells obtained from an advanced neu iiblasloma (case 32. stage 4) on day 7. A. untreated tun >r cells aggregate to form large clusters without neur e extension, fl. RA treatment induces mild neurite ingrowth. C a similar level of neurite extension is also observed in cells treated with GDNF. D, treatment of the cells with both CiDNF and RA induces remarkable neurite outgrowth. E. cells treated with I(X)ng/ml NGF. F. cells treated with both RA and NGF. G-J. tumor cells scatter or form small clusters. A small number of line neuntes are spontaneously ex tended (G'l. However, treatment of the cells with GDNF alone (/). RA alone (Hi, or RA + GDNF (/) has no effect on the morphological differentiation.
'
least 2 weeks. The grade of morphological differentiation was assessed by the thickness of outgrowing neuntes; (£•)grade2 + , moderate and gradual enhance number of neuntes, the thickness of the neuritic rods, and the rapidness of the ment; dl) grade 3 + , strong and gradual enhancement: and (?) grade 4 + , strong induction of neurite extension, as compared with those of untreated control and fast enhancement. cells on days 4 and 7 after starting the culture: (s. and NGFI-A. 1 X IO7 factor at all: (hi grade I +. a mild and gradual enhancement in the numbers and cells were suspended in 10 ml of standard medium and cultured in 150-mm 2160
Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. ODNF/NTN-INDUCED DIFFERENTIATION IN NEUROBLASTOMA
Fig. 2. Effect of GDNF and NTN on favorable- prognosis neuroblastoma cells (stage 1) in primary culture. The pictures were taken on day 9 of culture. A. cells were incubated in standard medium: fi. cells were treated with GDNF (50 ng/ml); C. cells were treated with NTN (50 ng/ml). In some primary neu- roblastomas. the tumor cells responded betler to NTN than to GDNF.
GFRa-2. 5'-GAATCCAACTGCAGCTCT-3' and 5'-AAGCAAGCCTGAA- GATGTCC-3' (PCR product size. 280 bp): and GAPDH. 5'-CATCAAGAAG- CM O CM O LO 00 r- r- GTGGTGAAGC-3' and 5'-TCTTACTCCTTGGAGGCCAT-3' (PCR product ' 00
Fig. 3. Expression of c-Rel mRNA in neuroblastoma cell lines and primary neuroblas- reading frame. The full-length cDNA was obtained by the rapid amplification tomas. Northern blot analysis was performed as described in "Materials and Methods." of cDNA end (5'-RACE; Clontech Laboratories. Inc.. Palo Alto), using human Neuroblastoma cell lines tend to express higher levels of c-flcf mRNA compared with kidney Marathon RACE library (Clontech Laboratories. Inc.) as a template. primary neuroblastomas. Cases 32 and 12 express c-Rel. albeit at a very low level. As a Statistical Analysis. For statistical analysis, patients were divided into control for the amount of RNA. the same filter was rehybridi/.ed with GAPDH. groups based on age ( Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. GDNF/NTN-INDUCED DIFFERENTIATION IN NEUROBLASTOMA NTN seemed more marked than that produced by GDNF. The neurite Stage 4 extension induced by NTN was also markedly enhanced by the treatment of cells with 5 /¿MRA. Expression of c-Ret in Neuroblastomas. Previously, we and an GDNF other group reported that the Ret proto-oncogene is frequently ex pressed in many neuroblastomas independent of disease stages (8, 37). NTN The expression of c-Ret mRNA was examined by either Northern analysis or RT-PCR. All 37 primary neuroblastomas expressed Ret c-Ret mRNA by the RT-PCR procedure (Table 1; Fig. 3; Fig. 5). Ret mRNA was expressed in 8 of 10 neuroblastoma cell lines by Northern GFRa-1 analysis and in 9 of 10 neuroblastoma cell lines by the RT-PCR procedure (data not shown). The levels of c-Ret expression in the primary neuroblastomas were generally lower than those of the neu GFRi,-2 roblastoma cell lines (Fig. 3). Expression of GFRat-1 and GFRa-2 in Neuroblastomas. To obtain a human GFRa-1 probe, we screened a human adult substantia nigra cDNA library. The amino acid sequence of the gene we cloned Response to GDNF (GenBank accession number U95847) deleted five amino acids in the Fig. 5. mRNA expression of CDNF, NTN. c-Rel. CFRa-1, and GFRa-2 in primary coding frame with some amino acid substitutions, compared to the neuroblastomas detected by the RT-PCR procedure. The PCR products were electrophore- recently reported human GFRa-1. Fig. 4 shows an amino acid align sed on 8% polyacrylamide gels or 3% agarose gels and stained with ethidium bromide. ment of the novel isoform of human GFRa-1 (GFRa-i-15deï) with The expression of GAPDH is shown as a control. For CFRa-l. the upper band (PCR product size, 376 bp) reflects the expression of GFRa-1, and the lower band (product size, the previously reported human GFRa-1. 361 bp) reflects the expression of GFRa-l-15tlel. The RT-PCR method was applied to examine the levels of expres sion of GFRa-1 mRNA and GFRa-l-15del mRNA. The former was RESULTS expressed in 34 of 37 (92%) tumors, and the latter was expressed in all 37 (100%) tumors (Table 1; Fig. 5). GFRa-2 mRNA was also Effects of GDNF and RA on Primary Neuroblastomas. A total expressed in 35 of 37 (95%) primary neuroblastomas. In 10 neuro of 38 fresh neuroblastoma tissues obtained by surgery before treat blastoma cell lines, GFRa-1, GFRa-I-l5del, and GFRa-2 were ex- ment were processed directly for primary culture in vitro as described in "Materials and Methods." The clinical features of the patients are shown in Table 1. The untreated control cells usually formed clusters, RA and most cells died within a week or so (Fig. IA). In some cases, spontaneous neurite outgrowth was observed. The exogenous GDNF (100 ng/ml) promoted cell survival and induced mild to strong dif ferentiation in 34 of 38 cases (89%; Table 1; Fig. 1C). The cells -7.0kb treated with GDNF adhered tightly to the collagen-coated dish and - survived for more than 4 weeks. The GDNF-induced neurite extension -6.0 usually started on day 2-3 and was enhanced thereafter by increasing c-Ret -4.5 in number, length, and thickness. The morphological differentiation -3.9 was irreversible, because depriving the medium of GDNF did not cause the withdrawal of neuntes and led the cells to die after several weeks (data not shown). The dose-response study showed that 1 ng/ml GFRa-1 -10.0 GDNF could induce neurite outgrowth from the tumor cells, reaching a maximum at 10 ng/ml. NGF also induced remarkable morphological differentiation in 32 of 38 cases (84%; Fig. 1£;Table 1). As compared with the neurite outgrowth induced by treatment with NGF, that GAPDH m* induced by GDNF was less extensive. The treatment of the primary neuroblastoma cells with 5 /J.MRA induced relatively weak neurite outgrowth in 19 of 38 (50%) B 10 20 25 30 35 40 (cycles) tumors (1+-3+ in Table 1; Fig. Iß).When the cells were treated with both 5 /IM RA and 1-100 ng/ml GDNF simultaneously, a - GFRa-2 striking neurite extension was induced (Fig. ID). Depriving the medium of both RA and GDNF led the cells to die gradually without reversing the effect. RA treatment also enhanced the effect of NGF (Fig. IF). - GFRa-2 Effect of NTN on Primary Neuroblastomas. The effect of NTN RA(+) (100 ng/ml) was compared with that of GDNF (100 ng/ml) in a separate series of 17 primary neuroblastomas (seven stage 1, two stage Fig. 6. RA induces mRNA expression of c-Kei. but not of GFRa-I or GFRa-2, in a favorable-prognosis neuroblastoma (case 16, stage 1) in primary culture. A. for the 2, two stage 4s, four stage 3, and two stage 4 neuroblastomas). Fifteen expression of c-Ret and GFRa-1. cells were treated with 5 /IM RA in 5% CO2/95% air at of 17 primary neuroblastomas responded to NTN by extending neu- 37°Cfor 3 days. The control untreated cells «erecultured for 1 day in the absence of RA rites. The pattern of induction of neurite outgrowth by NTN was very and harvested before starting to die. c-Rel expression is up-regulated by the RA treatment, whereas the expression level of GFRa-1 is unchanged. As a control for the amounts of similar to that induced by GDNF (Fig. 2). In some neuroblastomas, RNA, the same filter was rehybridb.ed with GAPDH. B. to examine the expression of however, the induction of morphological differentiation produced by GFRa-2, RT-PCR was performed as described in "Materials and Methods." 2162 Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. GDNF/NTN-INDUCED DIFFERENTIATION IN NEUROBLASTOMA to NGF was significantly associated with younger age (P < 0.(XM)1), GDNF non-stage 4 tumors (P < 0.001 ), a single copy N-wvc (P < 0.0004). aneuploidy (P < 0.0066), and a high TRK-A expression (P < 0.0001 ). Similarly, good responsiveness to RA + GDNF was well correlated with younger age (P < 0.0039), non-stage 4 tumors {P < 0.0023). a single copy N-mvc (P < 0.027), and a high TRK-A expression c-Fos (P < 0.0062), but not with aneuploidy (P < 0.09). The morphological differentiation in response to GDNF alone or to RA showed signifi cant correlation with none of the above-mentioned prognostic indica tors. DISCUSSION NGFI-A The role of the c-Ret proto-oncogene in neuroblastomas is unclear; however, c-Ret is frequently expressed in both primary neuroblasto mas and neuroblastoma cell lines (8. 37, 38). The recent discovery that a neurotrophic factor. GDNF. is a ligand for a heterotetrameric re ceptor complex of c-Ret and GFRa-1 has allowed us to further investigate the role of Ret (23-26). We found that neuroblastomas in GAPDH primary culture frequently respond to GDNF or NTN by extending neurites. This responsiveness is correlated with the expression of both c-Ret and GFRa-1 and/or GFRa-2 and is enhanced by treatment with Fig. 7. Induction of the immediate-early genes c-Fos and NGFI-A after treatment with RA. This is especially prominent in favorable-prognosis groups of GDNF in the primary culture of neuroblastoma (case 16. stage 1). Cells (1 X IO7) were neuroblastomas. The GDNF stimulation induced the expression of incubated in the presence of 5 ¡J.MRAfor 3 days and then treated with or without UHI immediate-early genes c-Fos and NGFI-A. RA treatment induced ng/ml GDNF for 40 min before harvesting. As a control for the amounts of RNA. the same filter was rehybridized with GAPDH. c-Ret expression, but not GFRa-1 or GFRa-2 expression. In addition. most primary neuroblastomas expressed GFRa-2. and the pattern of the NTN effect on primary tumor cells was very similar to or even pressed by 50% (5 of 10), 70% (7 of 10), and 90% (9 of 10) of the cell stronger than that of GDNF. Thus, like NGF, GDNF and NTN may lines, respectively. play an important role in regulating the differentiation and/or survival Expression of GDNF and NTN in Neuroblastomas. To examine of neuroblastoma cells in concert with RA or possibly with RA-like whether or not GDNF and/or NTN are expressed in the primary substances in vivo. neuroblastoma tissues. RT-PCR was performed. Representative cases The most striking difference between the NGF/TRK-A signaling are shown in Fig. 5; 24 of 37 (65%) tumors showed a positive signal and the GDNF(NTN)/Ret/GFRa-1 (GFRa-2) signaling in neuroblas for GDNF, and 28 of 37 (76%) showed a positive signal for NTN. toma is the expression pattern of their receptor molecules. The ex GDNF was expressed in 4 of 10 neuroblastoma cell lines, and NTN pression of TRK-A is very high in favorable-prognosis tumors, was expressed in all neuroblastoma cell lines tested. whereas it is very low in advanced tumors, especially in tumors with RA Induces Expression of c-Ret but not of GFRa-1 and N-ÕWVCamplification(5-7). In contrast, the expression of c-Ret and GFRa-2. We next tested whether the RA-induced enhancement of GFRa-l or GFRa-2 is ubiquitous among all stages of neuroblastomas. neunte outgrowth is due to the induction of Ret and/or GFRa-1/ The second difference is that the induction of neurite outgrowth by GFRa-2 or some other mechanisms. The stage 1 tumor cells in GDNF or NTN alone is rather weak compared to that induced by primary culture, whose response to GDNF was enhanced by the RA NGF. However, when the cells are treated with both RA and GDNF treatment, were harvested after treatment with 5 JU.MRA for 3 days. or NTN, the grade of induction of morphological differentiation was The expression of Ret mRNA was dramatically increased, whereas similar to that induced by NGF. In addition, such enhanced respon that of GFRa-1 and GFRa-2 was unchanged (Fig. 6). The induction siveness is significantly prominent in favorable-prognosis neuroblas of c-Ret by RA treatment was also observed in representative neuro tomas. These results suggest that favorable-prognosis tumors have an blastoma cell lines in a dose-dependent and time-dependent manner intrinsic signaling machinery for cell differentiation and survival that (data not shown). may be shared between NGF and GDNF or NTN. The present result Induction of c-Fos and NGFI-A. The primary cells from a stage has also indicated that one of the mechanisms for the enhancement of 1 neuroblastoma were treated in the presence or absence of 5 JLIMRA neurite outgrowth by RA may be an increase in the expression of Ret for 3 days at 37°C.Exogenous GDNF at a concentration of 100 ng/ml receptor molecules. It is unclear why aggressive neuroblastoma cells, induced the expression of the immediate-early genes c-Fos and as compared with favorable-prognosis tumor cells, have less potential NGFI-A (Fig. 7). to differentiate in response to RA and GDNF. Clinical Significance. Because the follow-up duration of the pa The RT-PCR study has revealed that both GDNF and NTN mes tients with neuroblastoma studied here was too short (2-12 months), sages are present in the neuroblastoma tissues. Based on previous the clinical significance of the effect of GDNF with or without RA on studies, GDNF may be produced by stromal cells such as fibroblasts neuroblastomas was evaluated in relation to age (<12 months versus and Schwann cells (20). A similar situation is found for NGF in a 12 months), stage (1, 2, 3, and 4s versus stage 4), N-mvr amplifi neuroblastomas (42). This would suggest that an increase in stromal cation (<10 copies versus z\0 copies), DNA ploidy (diploidy + tet- components accelerates tumor cell differentiation by supplying more raploidy versus aneuploidy), and expression of TRK-A (high versus NGF and GDNF/NTN. This is in contrast to brain-derived neurotro low) according to the results shown in Table 1. Consistent with the phic factor (BDNF) and NT-4. because they are produced from the previous reports (5-7), the level of TRK-A expression was signifi neuroblastoma cells in an autocrine or paracrine manner (9). GDNF cantly decreased in stage 4 tumors (P < 0.001 ). Good responsiveness and NTN may also have an autocrine loop in neuroblastomas. because 2163 Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. ODNF/NTN-INDUCF.D DIFFERENTIATION IN NF.UROBLASTOMA most neuroblastoma cell lines gave positive signals for these factors 14. Li, Y-S., Milner, P. G., Chauhanm, A. K.. Watson, M. A., Hoffman. R. M., Kodnewr, by RT-PCR. C. M.. Milbrandt. J.. and Deuel, T. F. Cloning and expression of a developmentally regulated protein that induces mitogenic and neurite outgrowth activity. Science We have cloned a novel GFRa-1 isoform with a 15-amino acid (Washington DC). 250: 1690-1694, 1990. deletion that seems to be a product of alternative splicing. The 15. Lin, L-F. H., Doherty, D. H., Lile, J. D., Bektesh, S., and Collins. F. GDNF: a glial GFRa-l-I5del was expressed in parallel with GFRa-l in most pri cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science (Washington DC), 260: 1130-1132, 1993. mary neuroblastomas. Because the effects of GDNF and NTN are 16. Kolzbauer, P. T., Lampe, P. A., Heuckeroth. R. O.. Golden, J. P.. Creedon. D. J., almost the same in primary neuroblastomas and neuroblastoma cell Johnson. E. M.. Jr.. and Milbrandt, J. Ncurturin, a relative of glial-cell-line-derived neurotrophic factor. Nature (Lond.), 384: 467-470, 1996. lines, it is possible that both factors share a Ret tyrosine kinase Tomac, A., Lindqvist, E., Lin, L-F. H., Ogren, S. O., Young, D., Hoffer, B. J., and receptor, and that isoforms of the GFRa-1 receptor family define a Olson. L. Protection and repair of the nigrostriatal dopaminergic system by GDNF in ligand specificity for GDNF family molecules (43). vivo. Nature (Lond.), 373: 335-339, 1995. 18. Beck, K. D.. Vallerde. J.. Alexi. T.. Poulsen. K.. Moffat. B.. Vandlen. R. A., Rosenthal. A., and Hefti. F. Mesencephalic dopaminergic neurons protected by GDNF from axotomy-induced degeneration in the adult brain. Nature (Lond.). 373: ACKNOWLEDGMENTS 339-341, 1995. Springer. J. E., Mu, X.. Bergmann. L. W., and Trojanowski. J. Q. Expression of We thunk the following institutions in Japan for providing surgical samples GDNF mRNA in rat and human nervous tissue. Exp. Neurol.. 127: 167-170. 1994. and clinical data: First Department of Surgery. Hokkaido University; Depart Henderson, C. E., Phillips, H. S., Pollock, R. A., Davies, A. M., Lemeulle, C., ment of Pediatrics. National Sapporo Hospital; Department of Pediatrie Sur Armanini. M. P.. Simpson, L. C., Moffet. B.. Vandlen, R. A.. Koliatsos, V. E., and gery, Tohoku University: Department of Pediatrie Surgery, Iwate Prefectural Rosenthal, A. GDNF: a potent survival factor for motoneurons present in peripheral Central Hospital; Department of Surgery. Gunma Children's Medical Center; nerve and muscle. Science (Washington DC). 266: 1062-1064. 1994. Springer. J. E.. Mu. X.. Bergmann, L. W., and Trojanowski, J. Q. Peripheral Department of Pediatrie Surgery, Jichi Medical University; Department of expression and biological activities of GDNF. a new neurotrophic factor for avian and Hematology and Oncology. Saitama Children's Medical Center; Department mammalian peripheral neurons. J. Cell Biol.. 130: 137-148. 1995. of Pediatrie Surgery, Tsukuba University; Department of Pediatrics, Juntendo 22. Buj-Bello. A.. Buchman. V. L., Horton. A.. Rosenthal. A., and Davies. A. M. GDNF University; Department of Surgery. Kiyose Metropolitan Children's Hospital; is an age-specific survival factor for sensory and autonomie neurons. Neuron, 15: Departments of Surgery and Pathology. Chiba Children's Hospital; Depart 821-828. 1995. 23. Durbec. P.. Marcos-Gutierrez. C. V.. Kikenny. C., Grigoriou, M., Wartiowaara. K., ment of Pediatrie Surgery, Matsudo Municipal Hospital; Department of Pedi Suvanto. P.. Smith. D.. Ponder. B.. Costantini. F., Saaram, M., Sariola, H., and atrie Surgery, Chiba University: Department of Pediatrie Surgery, Kimitsu Pachnis, V. GDNF signalling through the Ret receptor tyrosine kinase. Nature Central Hospital; Department of Pediatrie Surgery, Niigata University; De (Lond.), 381: 789-793. 1996. Trupp. M.. Arenas. E.. Fainzilber. M.. Nilsson. A-S., Sieber, B-A.. Grigoriou, M.. partment of Pediatrie Surgery, Niigata Municipal Hospital: Department of 24. Kikenny. C.. Salazar-Grueso, E., Pachnis, V., Arumäe.U., Sariola, H.. Saarma, M., Pediatrics. Aichi Medical University; Department of Pediatrics, Kyoto Prefec and Ibáñez,C.F. Functional receptor for GDNF encoded by c-rel proto-oncogene. tural Medical University; Department of Pediatrie Surgery. Osaka City General Nature (Lond.), 381: 785-789. 1996. Medicai Center; Department of Pediatrics, Osaka Medical University; Tumor 25, Jing, S.. Wen. D.. Yu. Y., Hoist, P. L.. Luo. Y.. Fang, M.. Tamir. R.. Antonio. L., Hu. Board. Hyogo Children's Hospital: Department of Pediatrie Surgery. Kum- Z.. Cupples, R., Louis, J-C. Hu, S., Altrock. B. W., and Fox, G. M. GDNF-induced activation of the Ret protein tyrosine kinase is mediated by GDNFR-a. a novel amoto University: and the Departments of Pediatrics and Pediatrie Surgery, receptor for GDNF. Cell, 85: 1113-1124, 1996. Kagoshima University. We also thank Jeffrey Milbrandt and Hideki Enomoto 26. Treanor. J. J. S.. Goodman. L.. de Sauvage, F.. Stone, D. M., Poulsen, K. T.. Beck, (Departments of Pathology and Internal Medicine. Washington University, St. C. D., Gray, C., Armanini, M. P., Pollock. R. A.. Hefti, F., Phillips, H. S.. Goddard, Louis. MO) and Garrett M. Brodeur (Division of Oncology. Children's Hos A.. Moore, M. W.. Buj-Bello. A., Davies, A. M., Asai, N., Takahashi, M., Valden, R., pital of Philadelphia. Philadelphia. PA) for reading the manuscript. Henderson. C. E.. and Rosenthal. A. Characterization of a multicomponent receptor for GDNF. Nature (Lond.), 382: 80-83. 1996. 27. Baloh. R. H.. Tansey. M. G.. Golden. J. P.. Creedon. D. J., Heuckeroth. R. O., Keck, C. L.. Zimonjic, D. B., Popescu. N. C., Johnson, E. M.. Jr.. and Milbrandt. J. TmR2. REFERENCES a novel receptor that mediates neurturin and GDNF signaling through Ret. Neuron, 19: 793-782, 1997. 1. Bolande, R. P. The neurocristopathies: a unifying concept of disease arising in neural crest maldevelopment. Hum. Pathol.. J: 409-429, 1974. Klein, R. D.. Sherman. D., Ho, W-H., Stone, D.. Bennett, G. L., Moffat. B., Vandlen, R.. Simmons, L., Gu, Q., Hongo. J-A.. Devaux, B.. Poulsen. K., Armanini. M., 2. Brodeur. G. M., and Nakagawara, A. Molecular basis of clinical heterogeneity in neuroblastoma. Am. J. Pediatr. Hematol. Oncol.. 14: 111-116, 1992. Nozaki, C.. Asai, N., Goddard, A., Phillips, H.. Henderson, C. E., Takahashi. M., and 3. Chao. M. V. Neurotrophin receptors: a window into neuronal differenliation. Neuron. Rosenthal, A. A GPI-linked protein that interacts with Ret to form a candidate 9: 583-593. 1992. neurturin receptor. Nature (Lond.). 387: 717-721. 1997. 4. Snider. W. D. Functions of the neurotrophins during nervous system development: 29. Buj-Bello. A.. Adu. J.. Piñón.L.G. P.. Horton, A.. Thompson, J.. Rosenthal. A., what the knockouts are teaching us. Cell. 77: 627-638. 1994. Chinchetru. M.. Buchman. V. L.. and Davies, A. M. Neurturin responsiveness 5. Nakagawara, A., Arima. M.. A/ar. C. G.. Scavarda, N. J.. and Brodeur. G. M. Inverse requires a GPI-linked receptor and the Ret receptor tyrosine kinase. Nature (Lond.). relationship hclween trk expression and N-mvr amplification in human neuroblasto 387: 721-724. 1997. mas. Cancer Res., 52: 1364-1368. 1992. 30. Sanicola. M., Hession. C., Worley. D.. Carmillo. P., Ehrenfels. C., Walus, L., 6. Nakagawara. A.. Arima-Nakagawara. M.. Scavarda. N. J.. Azar, C. G-, Cantor, A. B., Robinson. S.. Jaworski, G.. Wei, H.. Tizard, R., Whitty, A., Pepinsky, R. B.. and Cate, and Brodeur. G. Association between high levels of expression of the TRK gene and R. L. Glial cell line-derived neurotrophic factor-dependent RET activation can be favorable outcome in human neuroblastoma. N. Engl. J. Med., 32H: 847-854. 1993. mediated by two different cell-surface accessory proteins. Proc. Nati. Acad. Sci. USA, 7. Su/.uki. T.. Bogenmann. E.. Shimada. H.. Stram, D., and Seeger, R. C. Lack of 94: 6238-6243. 1997. high-affinity nerve growth factor receptors in aggressive neuroblastomas. J. Nati. Suvanto, P.. Wartiovaara. K.. Lindahl. M.. Arumäe.U-, Moshnyakov, M.. Horelli- Cancer Inst., «5.-377-384. 1993. Kuitunen, N.. Airaksinen. M. S.. Palotie. A., Sariola, H.. and Saarma. M. Cloning. 8. Borrello. M. G., Bongarzone. I.. Pienotti. M. A.. Luksch. R.. Gasparini. M.. Collini. mRNA distribution and chromosomal localisation of the gene for glial cell line- P.. Piloni. S.. Rizzetti, M. G.. Mondellini. P.. De Bernardi. B.. Martino. D. D., derived neurotrophic factor receptor ß,a homologue to GDNFR-a. Hum. Mol. Garaventa. A.. Brisigotti, M., and Tonini, G. P. trk and ret proto-oncogene expression Genet.. 6: 1267-1273, 1997. in human neuroblastoma specimens: high frequency of trk expression in non-ad Durbec. P. L.. Larsson-Blomberg, L. B., Schuchardt, A.. Costantini. F.. and Pachnis, vanced stages. Int. J. Cancer, 54: 540-545, 1993. V. Common origin and developmental dependence on c-ret subsets of enteric and 9. Nakagawara, A.. Azar. C. G.. Scavarda. N. J., and Brodeur. G. M. Expression and function of TRK-B and BDNF in human neuroblastomas. Mol. Cell. Biol., 14: sympathetic neuroblasts. Development (Camb.). /22: 349-358. 1996. 759-767, 1994. 33. Mulligan. L., Kwok, J., Healey. C.. Eisdon, M., Eng, C., Gardner, E., Love, D.. Mole, 10. Sendtner, M., Carroll, P., Holtmann. B.. Hughes. R. A., and Thoenen. H. Ciliary S.. Moore, J.. Papi, L., Ponder. M.. Telenius, H., Tunnacliffe, A., and Ponder, A. neurotrophic factor. J. Neurobiol., 25: 1436-1453. 1994. Germ-line mutations of the ret proto-oncogene in multiple endocrine neoplasia type 11. Eckenstein. F. P. Fibrohlast growth factors in the nervous system. J. Neurobiol., 25: 2A. Nature (Lond.), 363: 458-460, 1993. 1467-1480. 1994. 34. Hofstra. R.. Landsvater. R.. Ceccherini, I., Stülp.R.. Stelwagen. T., Luo, Y.. Pasini, 12. Murphy, M.. Reid. K. Brown. M. A., and Bartlett, Y. A. Involvement of leukemia B.. Höppener.J.. van Amstel. H.. Romeo. G.. Lips. C., and Buys, C. A mutation in inhibitory factor and nerve growth factor in the development of dorsal root ganglion the ret proto-oncogene associated with multiple endocrine neoplasia type 2B and neurons. Development (Camb.). 117: 1173-1182. 1993. sporadic medullary thyroid carcinoma. Nature (Lond.), 367: 375-376, 1994. 13. Nakagawara. A., Milbrandt, J., Muramatsu. T.. Deuel. T. F.. Zhao, H., Cnaan. A., and 35. Edery, P.. Lyonnet, S., Mulligan, L.. Pelet. A.. Dow. E.. Abel, L., Holder. S.. Brodeur. G. M. Differential expression of pleiotrophin and midkine in advanced Nihoul-Fekete. C.. Ponder. B.. and Munnich. A. Mutations of the ret proto-oncogene neurohlastomas. Cancer Res., 55; 1792-1797. 1995. in Hirschsprung's disease. Nature (Lond.), 367: 378-380, 1994. 2164 Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. GDNF/NTN-INDUCED DIFFERENTIATION IN NEUROBLASTOMA 36. Romeo. G., Patrizia, R., Luo, Y., Barone. V., Seri, M., Ceccherini. I., Pasini, B.. tional Criteria for Neuroblastoma diagnosis, staging, and response lo treatment. Bocciardi, R.. Lerone. M.. Kääriäinen.H.,and Maartucciello. G. Point mulalions J. Clin. Oncol., //: 1466-1477. 1993. affecting the tyrosine kinase domain of the ret proto-oncogene in Hirschsprung's 40. Azar. C. G.. Scavarda. N. J.. Reynolds, C. P., and Brodeur. G. M. Multiple delects of disease. Nature (Lond.), 367: 377-378. 1994. the nerve growth factor receptor in human neuroblastomas. Cell Growth Differ., I: 37. Nagao. M.. Ishizaka, Y.. Nakagawara. A.. Kohno. K.. Kuwano. M., Tahira. T., Itoh, 421-428. 1990. F.. Ikeda. I., and Sugimura. T. Expression of ret proto-oncogene in human neuro- 41. Chonuvynski. P., and Sacchi, N. Single step method of RNA isolation by acid blastomas. Jpn. J. Cancer Res.. 81: 309-312, 1990. guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem.. 762.- 156- 38. Ikeda, I.. Ishizaka. Y., Tahira. T., Suzuki. T., Onda, M., Sugimura. T., and Nagao, M. 159. 1987. Specific expression of the ret proto-oncogene in human neuroblastoma cell lines. 42. Nakagawara. A., and Brodeur, G. M. Role of ncurotrophins and their receptors in Oncogene, 5: 1291-1296. 1990. human neuroblastomas: a primary culture study. Eur. J. Cancer, in press. 1998. 39. Brodeur, G. M., Pritchard. J.. Berthold, F.. Carlsen, N. L. T.. Castel, V.. Castleberry. 43. Creedon. D. J.. Tansey. M. G.. Baloh. R. H.. Osbome. P. A.. Lampe, P. A., Hahrner, R. P.. Bernardi. B. D.. Evans, A. E.. Favrot. M.. Hedborg, F.. Kaneko. M., Kemshead, T. J., Heuckeroth. R. ().. Milbrandt, J.. and Johnson, E:.M. Neurturin shares receptors J.. Lampert, F., Lee, R. E. J., Look, T.. Pearson, A. D. J.. Philip. T.. Roald. B.. and signal Iransduction pathways with glial cell line-derived neurotrophic factor in Sawada. T., Seeger. R. C.. Tsuchida. Y., and Voûte.P. A. Revisions of the Interna- sympathetic neurons. Proc. Nati. Acad. Sci. USA. 94: 7018-7023. 1997. 2165 Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research. Glial Cell Line-derived Neurotrophic Factor/Neurturin-induced Differentiation and Its Enhancement by Retinoic Acid in Primary Human Neuroblastomas Expressing c- Ret, GFRα-1, and GFRα-2 Tomoro Hishiki, Yoshinori Nimura, Eriko Isogai, et al. Cancer Res 1998;58:2158-2165. Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/58/10/2158 E-mail alerts Sign up to receive free email-alerts related to this article or journal. Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected]. Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/58/10/2158. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site. Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1998 American Association for Cancer Research.