Abstract
Peroxisome proliferator-activated receptor γ (PPARγ) is a member of the nuclear receptor superfamily of ligand-dependent transcription factors that has been shown to play a major role in adipocyte and monocyte/macrophage differentiation. Recent work has also suggested a role for PPARγ in cell cycle control and/or differentiation of other cell types including breast and lung cancer cells. Using reverse transcription-PCR, we now show for the first time that human neuroblastoma (nb) cells express PPARβ and -γ, but not -α. Using the LA-N-5 nb cell line, we have determined that the natural PPARγ ligand 15-deoxy-δ prostaglandin J2, as well as the synthetic PPARγ agonist GW1929, can stimulate the differentiation of nb cells, as evidenced by the inhibition of cell proliferation, neurite outgrowth, increased acetylcholinesterase activity, and the reduction of N-myc expression. We have also demonstrated that PPARγ is expressed in primary nb and, furthermore, that the expression of this receptor correlates with the maturational stage of the nb cells. Taken together, these studies have implicated a role for PPARγ in peripheral nerve cell biology and suggest that the PPARγ signaling pathway is involved in the regulation of nb cell growth and differentiation.
INTRODUCTION
nb3 is a tumor of the sympathetic peripheral nervous system originating in cells derived from the neural crest. It is the most common extracranial solid tumor in children and comprises up to 50% of malignancies among infants (1) . Although a highly aggressive disease, nb has the highest rate of spontaneous regression of any type of cancer. Furthermore, there is well-documented evidence that a small percentage of nb regressions are preceded by apparent maturational processes within the cancerous tissue (2) . It is these unique features of this disease, the relatively high rates of spontaneous regression and the well-documented occurrences of in vivo tumor-cell maturation, that have prompted a search for agents that can induce the differentiation of this cell type and might have clinical utility. In this regard, recent trials have demonstrated that retinoic acid is an effective compound for prolonging the remission time, increasing the survival, and reducing the recurrence of nb when used in the setting of minimal residual disease (3) .
PPARs compose a subfamily of the nuclear hormone receptors that heterodimerizes with the retinoid X receptor to function as a transcriptional regulator. There are three distinct PPARs that have been described, termed α, β, and γ, each encoded by a separate gene and showing a distinct tissue distribution pattern. Of these three PPAR isoforms, PPARγ has been implicated in the regulation of critical aspects of development and homeostasis, including adipocyte differentiation, glucose metabolism, cell cycle control, and macrophage development and function (4, 5, 6, 7, 8, 9) . PPARγ has been shown to be expressed in certain central nervous system neurons, although it is most highly expressed in white adipose tissue (10) . Both natural and synthetic ligands for PPARγ have been reported to stimulate adipocyte and tumor cell differentiation and to inhibit the growth and/or induce apoptosis of breast, prostate, and lung cancer cells (6 , 11 , 12) . A role for PPARγ in nerve cell development or function has, to our knowledge, not been described. We now show that human nb cells express PPARγ. Our data indicates that PPARγ ligands can stimulate the differentiation of nb cells and that the expression of this receptor correlates with the maturational stage of primary nb tissue. These results implicate a role for PPARγ in nb cell biology and provide the first evidence that the PPARγ signaling pathway may be involved in processes related to peripheral neuronal differentiation.
MATERIALS AND METHODS
Cell Culture.
The LA-N-5 human nb cell line was grown in RPMI 1640 (Cellgro, Herndon, VA) supplemented with 10% heat-inactivated fetal bovine serum, HEPES, 50 IU/ml penicillin/streptomycin, and 1μ g amphotericin (complete medium), as described previously (13) . 15d-PGJ2 was purchased from Alexis Biochemical (San Diego, CA). GW1929 was synthesized by the Medicinal Chemistry Department at Glaxo Wellcome Research and Development and was a generous gift from Dr. T. M. Willson at that institution (Glaxo Wellcome Co., Research Triangle Park, NC). The MCF-7 cell line was purchased from American Type Culture Collection (Rockville, MD). All other chemicals were purchased from Sigma (St. Louis, MO) unless otherwise indicated.
RT-PCR.
Total RNA was prepared from cultured nb cells by using TRI Reagent (Sigma). To amplify a 474-bp PPARγ cDNA fragment, the sequences of PCR primers for sense (5′-TCTCTCCGTAATGGAAGACC-3′), and antisense (5′- GCATTATGAGACATCCCCAC-3′) were synthesized according to the published data (14) . The RT-PCR was carried out as described (15) . The samples were first denatured at 95°C for 3 min before 32 PCR cycles, each with temperature variations as follows: (a) 95°C for 1 min; (b) 60°C for 1 min; and (c) 72°C for 1 min. After the last cycle, an additional extension incubation of 9 min at 72°C was performed. Analysis of amplicons was visualized on 1% agarose gel containing 0.2 μg/μl ethidium bromide. A 100-bp ladder (Life Technologies, Inc., Rockville, MD) was used as a size standard.
Acetylcholinesterase Activity.
Specific AChE activity was measured as a biochemical index of the relative state of differentiation of treated and control LA-N-5 cells. For measurement of AChE activity as described (13) , cells were grown in 12-well plates for 6 days in the presence or absence of indicated concentrations of 15d-PGJ2 or GW1929. After washing twice with PBS, cells were collected and centrifuged, then ice-cold 10 mm sodium phosphate buffer (pH 7) containing 0.5% Triton X-100 was added, and the cells were sonicated for 20 s. AChE activity was determined photometrically by following the hydrolysis of acetylthiocholine as described previously (13) . Protein concentrations were determined with a Sigma Bicinchoninic acid protein assay kit using BSA as the standard. Results in nmol/hr/mg protein, as mean ± SE of triplicate wells in a typical experiment, were expressed as a percentage of control. All experiments were repeated at least three times.
Cell Proliferation Assay.
LA-N-5 cell proliferation was assessed on the basis of the ability of the cells to stain with SRB (16) . Cells were grown in 24-well culture plates for 6 days in the presence or absence of indicated concentrations of 15d-PGJ2 or GW1929. The untreated control wells were ≥90% of confluence before harvest. The culture medium was removed, and the cells were washed three times with PBS. Trichloroacetic acid was then added (final concentration, 10%) for fixation at 4°C. After 1 h fixation, plates were washed five times with tap water. Then the plates were air-dried, and 0.4% SRB in 1% acetic acid was added for 30 min. Unbound SRB was removed by washing with 1% acetic acid four times. After air-drying, SRB dye within cells was dissolved for 5 min with 10 mm unbuffered Tris base (pH 10.5). The absorbance of the extracted SRB dye, representing protein content, was measured with a spectrophotometer at 540 nm.
Immunohistochemistry.
Immunohistochemistry was performed on formalin-fixed, paraffin-embedded primary human ganglioneuroblastomas or nb with ganglionic differentiation using an affinity-purified, rabbit polyclonal anti-PPARγ antibody “Ben” from the laboratory of Dr. M. Greene (17) . This antibody recognizes the epitope of 18 amino acids spanning the first and second exons of PPARγ, “DYKYDLKLQEYQSAIKVE.” Paraffin-embedded sections of neoplasm were deparaffinized, rehydrated, and heat-treated for antigen retrieval by boiling the sections in 10 mm citrate buffer for 5 min and then cooling to room temperature. Sections were then blocked for peroxidases in 3% H2O2 in methanol for 30 min, blocked with nonspecific goat serum for 15 min, and incubated with primary antibody Ben (or appropriate negative control) for 2 h at room temperature in a humidity chamber. Detection used the Vector Elite ABC kit with DAB staining and hematoxylin counterstaining.
Northern Blot Analysis.
LA-N-5 cells were cultured for 8 days in the presence or absence of 15d-PGJ2 or GW1929. Total RNA (30 μg), extracted by using TRI reagent as described (18) , was separated by electrophoresis in a denaturing formaldehyde agarose gel and blotted onto Magnacharge nylon transfer membrane (Micron Separations, Inc., Westborough, MA). The RNA was cross-linked to the membranes by irradiation for 1 min under UV light and baked for 5 min at 70°C. A random priming kit was used for labeling N-myc and GAPDH cDNA probes (Promega Corp., Madison, WI). The blots were hybridized overnight at 42°C with both 32P-labeled N-myc and GAPDH cDNA at the same time. After washing with 2 ×SSC-0.1% SDS and 0.1 ×SSC-0.1% SDS twice each, membranes were exposed for 18–72 h at −70°C to Kodak XAR-5 film with intensifying screens.
Statistical Analysis.
We used one-way ANOVA and Student’s t test for statistical analysis of data (comparison between two groups of data sets). Asterisks shown in the figures indicate significant differences of experimental groups in comparison with the corresponding control conditions (P ≤ 0.05).
RESULTS
Expression of PPAR in Human nb Cells.
The presence of PPARα, -β, and -γ was assessed in LA-N-5 nb cells by RT-PCR. Positive controls consisted of mouse liver for PPARα and MCF-7 cells for PPARβ and -γ (19, 20, 21) . As seen in Fig. 1⇓ , single bands with the predicted sizes for PPARβ and PPARγ product (484 and 474 bp, respectively) were detected in LA-N-5 cells. In contrast, these cells were found to be negative for PPARα (492 bp). The same RT-PCR reaction with double distilled H2O instead of total RNA as a negative control did not make any PCR products (Fig. 1⇓ , Lane 2). Negative results were also obtained with total RNA in the absence of reverse transcriptase as a reverse transcriptase-negative control (not shown).
Detection of PPARs in LA-N-5 human nb cells. Cells were shown to express PPARβ and PPARγ, but not PPARα mRNA as determined by RT-PCR. Positive controls included mouse liver for PPARα and MCF-7 breast cancer cells for PPARβ and -γ.
Morphological Differentiation.
Because PPARγ was found to be the predominant PPAR isoform found in LA-N-5 cells, we assessed the effects of the natural PPARγ ligand 15d-PGJ2 (22) and the synthetic PPARγ ligand GW1929 (23) on cell morphology and other functional activity (below). Fig. 2⇓ demonstrates the ability of 15d-PGJ2 to induce neurite outgrowth from these cells. Similar activity was observed with GW1929 (not shown). This effect first became apparent after ∼2 days of continuous exposure in culture, with maximal increases at 10 days in the formation of neurites occurring at concentrations of 10 μm for both PPARγ ligands. Concentrations <1 μm produced no noticeable morphological effects (data not shown). No decrease in the percentage of viable cells was observed in treated cells as compared with the control cells.
Morphology of LA-N-5 nb cells in the absence and presence of 15d-PGJ2. Cultures were treated for 10 days with vehicle (A) or with 10 μm 15d-PGJ2 (B). Photographed under phase contrast (×200).
Proliferation.
Fig. 3⇓ shows the dose-dependent effects of 15d-PGJ2 and GW1929 treatment on LA-N-5 cell proliferation using the SRB protein stain assay. Results indicated that >30% inhibition was achieved at 15d-PGJ2 concentrations of ≥10 μm, whereas little effect was seen at <2.5 μm after 6 days of treatment. In the case of GW1929, nearly 50% growth inhibition was observed at concentrations of 5 μm, whereas maximal effects of ∼60% inhibition were seen at 20 μm. Doses of GW1929 ≤1 μm had no significant effect on cell growth.
Dose-response curves showing the effects of the PPARγ ligands PGJ2 (A) and GW1929 (B) on the growth of LA-N-5 nb cells. Cells were treated for 6 days with the indicated concentrations of compound or solvent control. Values represent the mean ± SE of at least three independent experiments, each done in triplicate samples.
Acetylcholinesterase Activity.
Concomitant with neurite outgrowth, AChE activity increases in LA-N-5 cells induced to differentiate with a variety of agents (13 , 24) . To assess whether this biochemical index of nb differentiation was associated with 15d-PGJ2 and GW1929-induced neurite outgrowth, AChE activity was measured in cells treated with various concentrations of 15d-PGJ2 and GW1929 for 6 days. As can be seen in Fig. 4⇓ , AChE was significantly increased at 5 μm 15d-PGJ2, with maximal increases occurring at 10 μm (Fig. 4⇓ A). AChE activity was also significantly increased at 20μ m GW1929 (Fig. 4⇓ B). Although the absolute values of AChE varied from one experiment to another, the pattern was consistent and the results shown in Fig. 4⇓ are representative.
Dose-response curves showing the effects of PGJ2 (A) and GW1929 (B) on specific AChE activity in LA-N-5 cells after 6 days of culturing. Each point represents the mean ± SE of three replicate cultures.
N-myc Expression.
Expression of the N-myc oncoprotein has been shown to correlate with the malignant potential of primary human nb cells in vivo and with clinical prognosis negatively related to N-myc expression (25) . In LA-N-5 cells and other nb cell lines, N-myc has been shown to rapidly decrease during differentiation induced by a variety of pharmacological agents (24 , 26) . As such, N-myc expression is considered a molecular correlate of the differentiation stage of nb cells. As seen in Fig. 5⇓ , treatment of LA-N-5 cells for 8 days with either 15d-PGJ2 or GW1929 reduced the level of N-myc mRNA. As a positive control in these experiments, RA was used as a known differentiation inducer. Densitometric scanning indicated an ∼50% decrease in mRNA levels of N-myc induced by RA, whereas PGJ2 and GW1929 induced decreases of 58% and 42%, respectively. All values were normalized to GAPDH to compensate for any slight differences in loading (Fig. 5⇓ B).
Effect of PPARγ ligands on N-myc mRNA levels in LA-N-5 cells. Total cellular RNA (30 μg) was isolated from cells cultured for 8 days in the absence (Con) or presence of 10 μm PPARγ ligands (15d-PGJ2 or GW1929) or 5 μm RA (as a positive control for N-myc down-regulation), then subjected to Northern blot analysis using 32P-labeled probes for N-myc and GAPDH (A). Densitometric scanning indicated an ∼50% decrease in mRNA levels of N-myc induced by RA, whereas 15d-PGJ2 and GW1929 induced decreases of 58% and 42%, respectively (B). All values were normalized to GAPDH to compensate for any slight differences in loading.
PPARγ Expression in Primary nb Tissue.
Immunohistochemical detection of PPARγ protein was examined in sections from three different human primary nb and one ganglioneuroma. The revised Shimada/Joshi diagnoses consisted of the following: (a) case 1, 15-month-old female, abdominal mass, nb, stroma poor, minimal ganglionic differentiation, low mitotic-karyorrhectic index (27) with focal ganglionic differentiation, favorable histology; (b) case 2, 3-month-old male, mediastinal mass, nb, stroma poor, low mitotic-karyorrhectic index, favorable histology; (c) case 3, 3-year-old male, paraspinal mass ganglioneuroblastoma, stroma rich, intermixed subtype, favorable histology; and (d) case 4, 12-year-old female, chest wall mass, ganglioneuroma.
Three of four cases showed neuroblastic cells positive for PPARγ (Table 1)⇓ . Two cases (Figs. 6⇓ and 7⇓ , cases 3 and 4, respectively) showed evidence of strong nuclear staining for PPARγ only in selected neuroblasts with ganglionic differentiation. The amount of signal was quite striking (see Figs. 6, a and b⇓ , and 7⇓ ) and was occasionally both nuclear and cytoplasmic. Primitive neuroblasts showed only weak to negative PPARγ expression by antibody staining. The third positive case (case 1) again showed selective primitive neuroblasts with weak, but definite, nuclear positivity and no evidence of PPARγ in multinucleated neuroblasts (data not shown). RNA was not well-enough preserved in these four tumor specimens to assess for PPARγ mRNA expression. However, the presence of PPARγ mRNA detected by RT-PCR in both the LA-N-5 (see above) and the SHSY5Y (data not shown) human nb cell lines suggests that there may be weak expression in the more primitive neuroblasts. We also note that some of the schwannian stromal nuclei appear weakly positive (Fig. 7)⇓ , and we saw frequent PPARγ positivity in normal peripheral nerve schwann cell nuclei (not shown). Some of the accompanying lymphocytic infiltrate showed nuclear PPARγ expression as might be expected in activated T or B lymphocytes (17 , 28) .
PPARγ expression in primary tissue
PPARγ expression in primary nb (case 3). a, positive and negative ganglioneuroblasts, with more primitive neuroblasts appearing weak to negative for PPARγ expression; ×400. b, ganglioneuroblasts with strong nuclear and cytoplasmic PPARγ expression adjacent to a ganglioneuroblast with only nuclear expression; ×1000.
PPARγ expression in primary ganglioneuroma (case 4). Strong nuclear staining for PPARγ in ganglion cell (a); weak to negative staining in less mature ganglioneuroblasts. Occasional nuclear positivity in lymphocytes (b) and stromal cells (c); ×200.
DISCUSSION
The expression of PPARγ has been reported in several organs and tissues, such as liver, adrenal gland, spleen, skeletal muscle, monocyte/macrophages, some leukemias, and at high levels in adipose tissue (29, 30, 31, 32) . Although PPARγ is expressed at low levels in normal colonic and breast ductal epithelium, it is significantly increased in both colon and breast carcinoma (33, 34, 35) . The expression of this receptor in nb has heretofore not been reported. In this study, we demonstrated that the LA-N-5 human nb cell line expresses readily detectable levels of PPARγ. The cells were also positive for PPARβ, albeit at lower levels than PPARγ, but were negative for PPARα. Recently, there has been significant progress in the identification of natural and synthetic ligands for PPARγ. Endogenous ligands that have been identified include antidiabetic thiazolidinediones (8) , polyunsaturated fatty acids (36) , 15d-PGJ2 (7) , and components of oxidized low-density lipoprotein such as 9-hydroxyoctadecadienoic and 13-hydroxyeicosatetraenoic acids (37) . More recently, synthetic ligands for PPARγ such as GW1929 have been described (23) . In the present work, we have shown that 15d-PGJ2 and GW1929 induce differentiation of LA-N-5 cells as assessed by dose-dependent growth inhibition, neurite outgrowth, increased AChE activity, and reduction of N-myc expression. 15d-PGJ2 has been shown previously to promote the differentiation of breast, lung, and prostate cancer cells (6 , 11 , 12) . Although the mechanism(s) of this activity was not determined, the results have suggested the presence of peroxisome proliferator response elements in the promoter of some critical oncogenes that may be playing a role in the malignant process (38 , 39) . In the case of nb, analysis of the full length N-myc promoter has revealed no significant sequence similarities with known peroxisome proliferator response element motifs, suggesting that direct binding of PPARγ to this promoter may not occur. In our previous studies of RA-induced down-regulation of N-myc, we have determined that the RA action may be mediated through interference with factors necessary for basal promoter activity such as those binding to E2F and Sp1 sites (40) . These findings are consistent with the ability of many members of the steroid hormone receptor family to inhibit gene expression through protein-protein interactions with other basal transcription factors such as NFκB and AP-1 (41) . Future studies will determine whether down-regulation of N-myc by PPARγ ligands may also be mediated through this type of DNA-independent mechanism.
Neuroblastoma tumors (using the term “neuroblastoma” in the clinical sense) have been shown to be a heterogeneous mix in terms of expression of N-myc, TrkB, neuron-specific enolase, glial fibrillary acid protein, catecholamines, and many other significant proteins (42) . This heterogeneity ultimately results in a mixture of cellular phenotypes and arrays reflecting different levels of differentiation derived from the multipotent neural crest-derived neuroblast. Such phenotypes include primitive, large, and multinucleated neuroblasts, the formation of Homer-Wright rosettes and neuropil, schwannian stromal cells, intermediate cells, pheochromocytoma cells, arrangements of neuroblasts in nesting patterns, and varying degrees of ganglionic differentiation. In this regard, a pattern of primitive neuroblasts admixed with cells showing schwannian stromal or ganglionic differentiation appears to be a more important predictor of a better prognosis than the sheer volume of differentiated cells (42) . Maturation of nb to ganglioneuroma can occur in spontaneous regression of disease (43) as well as with a variety of chemotherapeutic treatments, including retinoids, nerve growth factor, and cyclic nucleotides (44) . Infants <1 year of age with International Neuroblastoma Staging System 4S nb are most likely to undergo spontaneous regression. Differentiation therapy generally aims to recapitulate this phenomenon.
Our data indicated that in primary nb tumors, the level of PPARγ protein appears to correlate with the maturational stage of the cell, such that cells with a well-differentiated phenotype are capable of expressing higher amounts of PPARγ. Whether or not this association may be at least partly attributable to the functional consequence of aberrant PPARγ expression in immature nb cells, or simply reflects phenotypic differences in expression between neural cells at distinct levels of differentiation, is unknown. Initial studies have shown that differentiation of nb cell lines with retinoic acid causes an increase in PPARγ expression.4 It has also been demonstrated that normal postmitotic peripheral ganglionic cells and central nervous system neurons can show abundant levels of PPARγ (31 , 45) .
Taken together, our in vitro findings demonstrating the ability of PPARγ ligands to induce the differentiation of nb cells, along with the finding that the maturational level of nb tissue correlates with PPARγ expression, suggest that signaling through this nuclear receptor may play a significant role in nb cell biology. These findings also support future investigations into the prognostic significance of PPARγ expression in neuroblastoma and whether therapeutic administration of ligands for this receptor may have clinical potential for the treatment of this disease.
Footnotes
-
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 18 U.S.C. Section 1734 solely to indicate this fact.
-
↵1 This work was supported by NIH Grants CA43503, HD5276 (to N. S.), and NS34432 (to R. K. W.).
-
↵2 To whom requests for reprints should be addressed, at Department of Obstetrics and Gynecology, Emory University School of Medicine, 1639 Pierce Drive, Atlanta, Georgia, 30322. Phone: (404) 727-9155; Fax: (404) 727-8615; E-mail: Nsidell{at}emory.edu
-
↵3 The abbreviations used are: nb, neuroblastoma; PPAR, peroxisome proliferator-activated receptors; 15d-PGJ2, 15-deoxy-δ prostaglandin J2; RT-PCR, reverse transcription-PCR; AChE, acetylcholinesterase; SRB, sulfornodamine; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RA, retinoic acid.
-
↵4 N. Sidell, S. W. Han, and M. E. Green, unpublished data.
- Received July 18, 2000.
- Revision received October 12, 2000.
- Accepted October 13, 2000.
References
Volume 7, Issue 1
