SEARCH

SEARCH BY CITATION

Keywords:

  • neuroblastoma;
  • calcium-sensing receptor;
  • parathyroid hormone-related protein;
  • differentiation;
  • retinoic acid;
  • development

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

BACKGROUND:

Differentiated histopathology is a favorable prognostic factor in neuroblastic tumors, and molecular pathways underlying neuroblastoma differentiation can be modulated pharmacologically. The calcium-sensing receptor (CaR) and parathyroid hormone-related protein (PTHrP) regulate differentiation processes in some cellular contexts. CaR is up-regulated when neural stem cells are specified to the oligodendrocyte lineage and regulates PTHrP production in astrocytes. The objective of the current study was to assess whether CaR and PTHrP participate in neuroblastoma differentiation pathways.

METHODS:

CaR and PTHrP messenger RNA (mRNA) and protein expression were analyzed in neuroblastic tumors, and correlation with prognostic factors was assessed. CaR and PTHrP expression levels were analyzed in neuroblastoma cell lines treated with all-trans-retinoic acid or 5-bromo-2′-deoxyuridine (BrdU).

RESULTS:

CaR expression was correlated with favorable histology, age at diagnosis <1 year, low clinical stage, and low clinical risk. CaR was absent in undifferentiated neuroblasts and was expressed in differentiating neuroblasts. CaR and PTHrP were highly expressed in ganglion and in Schwann-like cells. PTHrP mRNA levels were higher in ganglioneuroblastomas and ganglioneuromas than in neuroblastomas (P < .0001). Both genes were up-regulated in neuroblastomas with treatment-induced maturation features. CaR, but not PTHrP, was up-regulated at early phases of in vitro neuronal differentiation induction. Substrate-adherent, non-neuronal cell lines displayed the highest PTHrP levels among the neuroblastoma cell lines examined. The up-regulation of PTHrP and of 2 glial differentiation markers was observed in 2 cell lines that were treated with BrdU, whereas CaR was induced in only 1 cell line.

CONCLUSIONS:

CaR and PTHrP were expressed in differentiated, favorable neuroblastic tumors, and both genes were up-regulated by inducing differentiation. Cancer 2009. © 2009 American Cancer Society.

Neuroblastic tumors arise from the peripheral nervous system and are composed mainly of 2 neural crest-derived cell types: neuroblastic/ganglion cells and Schwann-like cells. In 1999, the International Neuroblastoma Pathology Classification (INPC) was proposed as a prognostically significant classification of neuroblastic tumors,1 and 4 categories were established: neuroblastoma (Schwannian stroma-poor); ganglioneuroblastoma, intermixed (Schwannian-stroma rich); ganglioneuroma (Schwannian-stroma dominant); and ganglioneuroblastoma, nodular. Neuroblastic maturation toward ganglion cells and a high proportion of Schwannian stroma were recognized among favorable histologic features and were correlated with better overall survival.2

Genetic alterations in neuroblastic tumors, such as amplification in the v-myc myelocytomatosis viral-related oncogene MYCN and recurrent chromosomal abnormalities, are critical for the pathogenesis of these tumors, but they hardly are reversible. Conversely, the molecular pathways responsible for neuroblastoma differentiation can be modulated pharmacologically, and this provides an opportunity to design novel therapeutic tools. In fact, the differentiating agent retinoic acid induces differentiation in neuroblastoma cell lines3 and increases the event-free survival of patients with neuroblastoma.4

Neuroblastoma cell lines are used as in vitro models. Some are composed of neuroblastic (N-type), substrate-adherent non-neuronal (S-type), or intermediate (I-type) cells that can be distinguished by their morphology, immunophenotype, and biochemical features.5 The treatment of neuroblastoma cell lines with retinoic acid induces neuronal differentiation,6, 7 and 5-bromo-2′-deoxyuridine (BrdU) produces differentiation along the glial-like lineage.7, 8

The calcium-sensing receptor (CaR) is a G-protein–coupled receptor with the primary function of detecting extracellular calcium fluctuations and regulating parathyroid hormone secretion accordingly.9 The human CaR gene contains 6 coding exons and 2 alternatively transcribed first uncoding exons, 1A and 1B.10 One of its splicing variants lacks exon 5.11 Expression of CaR has been observed in organs involved in calcium homeostasis but also in other tissues. The role of CaR in many of these organs and tissues remains unclear, although several lines of evidence support the possibility that it is necessary for keratinocyte differentiation.12 In the developing central nervous system, CaR is regulated with a pattern that parallels that of myelin basic protein.13 In addition, up-regulation of CaR occurs in neural stem cells when they are specified to the oligodendrocyte lineage, the counterpart of Schwann cells in the central nervous system.14

CaR expression also has been described in several neoplasias, but its activation determines opposite cell fates in different tumors. In prostate carcinoma cell lines, CaR activation facilitates proliferation and metastatic progression,15 whereas CaR expression in colon carcinoma correlates with differentiated histology.16

In many normal and neoplastic cells, CaR activation results in up-regulation and increased secretion of the parathyroid hormone-related protein (PTHrP),17 which exerts autocrine and paracrine functions that modulate cell growth and differentiation.17, 18 In cancer, PTHrP activity has been associated mostly with hypercalcemia of malignancy19 and bone metastases.20 However, a large prospective analysis concluded that PTHrP immunoreactivity in breast cancer predicted an improved prognosis.21 The current study was undertaken to test the hypothesis that CaR and PTHrP may participate in the differentiation processes of neuroblastic tumors.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

Patients and Samples

Sixty snap-frozen tumors from 48 patients who were diagnosed with neuroblastic tumors at Hospital Sant Joan de Deu (Barcelona, Spain) and at other Spanish institutions were selected: Forty-six specimens were obtained at diagnosis, 11 were postchemotherapy specimens, and 3 specimens were obtained at the time of disease recurrence. Tumors were classified according to INPC criteria. Also, age at diagnosis, clinical stage (International Neuroblastoma Staging System), MYCN amplification status, clinical risk (high risk was defined as stage 4 and age >1 year or MYCN amplification), and time to follow-up were recorded. Tumors were designated as MYCN amplified if they had ≥10 MYCN copies.

Two tissue arrays were constructed using duplicate 1-mm cores from 64 formalin-fixed, paraffin-embedded tumors. Thirteen tumors that were not included in the arrays also were analyzed. Samples were derived from 64 primary tumors, from 11 postchemotherapy samples, and from 2 tumors at the time of disease recurrence. All human specimens were obtained under the approval of the institutional review boards of the authors' institutions.

RNA Isolation, Complementary DNA Synthesis, and Polymerase Chain Reaction

Total RNA was isolated with TriReagent (Sigma Chemical Company, St. Louis, Mo). First-strand complementary DNA (cDNA) was synthesized with random primers and M-MLV (Promega, Madison, Wis). Amplification of CaR was performed as described11 to coamplify the full-length transcript (649 base pairs) and the splicing variant that lacked exon 5 (419 base pairs). Amplification of the housekeeping gene TATA-binding protein (TBP) also was performed with the forward primer 5′-CAGCCGTTCAGCAGTCAAC-3′ and the reverse primer 5′-TGTTGGTGGGTGAGCACAAG-3′. The 2 CaR amplicon sequences were analyzed after excision and purification in a 3100-Avant Sequence Detection System (Applied Biosystems, Foster City, Calif). CaR regions exon 1B (5′-AAGACCGTGACCTTGGCATA-3′) to exon 3 (5′-GCTGGGCTGCTGTTTATCTC-3′) and exon 6 (5′-GGACCAGGAAAGGGATCATT-3′) to exon 7 (5′-TGGCATAGGCTGGAATGAAGG-3′) were amplified under the same conditions.

Quantitative Real-Time Polymerase Chain Reaction

CaR, PTHrP, S100-β, peripheral myelin protein 22 (PMP22), and TBP messenger RNA (mRNA) analysis, using the ΔΔCT relative quantification method, were performed on an ABI Prism 7000 Sequence Detection System with TaqMan Assay-on-Demand Gene Expression products, according to the manufacturer's protocols (Applied Biosystems). All experiments included no template controls and were repeated twice independently. Transcript levels were normalized to TBP expression. PTHrP levels in tumors were measured relative to those in adrenal gland. Transcript levels in differentiation assays were measured relative to those in untreated cells, except for CaR levels, which were measured relative to the levels in the U2OS osteosarcoma cell line.

Immunohistochemistry

Two-micrometer-thick sections were deparaffinized; rehydrated; heated in 10 mmol/L citrate buffer, pH 6.0; and subsequently incubated with 3% nonfat milk and 3% H2O2. Rabbit polyclonal anti-CaR antibody (Affinity BioReagents, Cambridge, United Kingdom) at 1:200 dilution or rabbit polyclonal anti-PTHrP antibody (34-53; Merck, La Jolla, Calif) at 1:100 dilution was applied for 1 hour at 37°C. Reactions were developed with the Novolink Polymer Detection System (Novocastra Laboratories, Newcastle upon Tyne, United Kingdom). Counterstaining was performed with hematoxylin. Kidney and peripheral nerve ganglia were used as positive controls. Negative controls were processed in the absence of primary antibody. The results were assessed by a pathologist (N.T.). Evaluation included cellular type, percentage of positive cells (cutoff value, >10%), and intensity of staining (0, absent staining; 1, faint staining; 2, staining clearly above background; 3, strong staining). Tumors were classified into 3 groups as follows: Group 1, undifferentiated neuroblastomas; Group 2, poorly differentiated, differentiating neuroblastomas; and Group 3, ganglioneuroblastomas and ganglioneuromas. The mean staining value and the standard deviation were calculated for each group.

Concordance between mRNA and protein expression results was assessed. CaR results were considered concordant when tumors in which CaR amplification was observed in reverse transcriptase-polymerase chain reaction (RT-PCR) had immunohistochemical analysis with stain values from 1 to 3. For PTHrP analysis, results were considered concordant when tumors with quantitative RT-PCR (qRT-PCR) expression levels higher than the median level (>4.98) had stain values of 3 and when tumors with qRT-PCR ≤4.98 had stain values from 0 to 2.

In Vitro Differentiation Assays

Twelve neuroblastoma cell lines were used to assess levels of expression under standard culture conditions: 4 N-type cell lines (LAN-1, LAI-55N, SH-SY5Y, and BE[2]-M17V), 4 I-type cell lines (SK-N-BE[2]-C, SK-N-ER, SK-N-JD, and SK-N-LP), and 4 S-type cell lines (LAI-5S, SMS-KCN, SH-EP1, and SK-N-AS). The cell lines kindly were provided by Dr. B. Spengler (Fordham University, New York, NY) and Dr. N.K.V. Cheung (Memorial Sloan-Kettering Cancer Center, New York, NY). SK-N-AS cells were purchased from the European Collection of Cell Cultures. Cells were grown in RPMI-1640 supplemented with 10% fetal bovine serum (Invitrogen, Barcelona, Spain), 2 mM L-glutamine, penicillin (100 U/mL), and streptomycin (100 μg/mL) at 37°C in a 5%CO2 atmosphere.

Two I-type cell lines, SK-N-BE(2)-C (MYCN amplified) and SK-N-ER (MYCN nonamplified), were used for differentiation assays. Cells were seeded, 10 μmol/L all-trans-retinoic acid (Sigma Chemical Company) were added to media the next day, and the cells were collected in TriReagent 4 days later. Cells cultured without retinoic acid were seeded, processed, and collected at the same time. Experiments with duplicate samples were performed 5 times.

SK-N-BE(2)-C and SK-N-ER cells also were treated with 10 μmol/L BrdU (Sigma Chemical Company) for 15 days. Drug was renewed every 3 days, and cells were harvested in TriReagent on Days 3, 6, 9, 12, and 15. Cells cultured without BrdU were seeded, processed, and collected simultaneously. Experiments with duplicate samples were performed 3 times each.

Senescence Analysis

Cytochemical detection of senescence-associated β-galactosidase was performed with the Senescence Cells Histochemical Staining Kit (Sigma Chemical Company) according to the manufacturer's instructions.

Statistical Analysis

The chi-square test was used to correlate the categorical variables. Comparison of means was performed with Kruskal-Wallis and Mann-Whitney U tests. The time elapsed from diagnosis to an event or to the end of follow-up was used to compute event-free survival and overall survival probabilities according to the method of Kaplan and Meier.22 The log-rank statistic was used to compare the event-free and overall survival probabilities between groups.23 The prognostic significance of variables was assessed by using Cox proportional models.24P < .05 was considered significant. Cohort sizes varied between analysis because of either missing clinical data or tissue availability for the factors under study.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

CaR Messenger RNA Expression in Neuroblastic Tumors Correlates With Favorable Histology, Age at Diagnosis <1 year, Low Tumor Stage, and Low Clinical Risk

CaR expression initially was analyzed in 4 non-neoplastic tissues: kidney, adrenal gland, bone marrow mononuclear cells, and placenta. The full-length transcript and the spliced form that lacked exon 5 were coamplified in all samples except for placenta cDNA, in which only the full-length product was detected (Fig. 1A).

thumbnail image

Figure 1. Calcium-sensing receptor (CaR) messenger RNA (mRNA) expression in normal tissues and in neuroblastic tumors. (A) Reverse transcriptase-polymerase chain reaction (RT-PCR) of CaR and the housekeeping gene TATA-binding protein (TBP) in non-neoplastic tissues (bp indicates base pairs). Lane 1, molecular weight marker (MWM); lane 2, kidney; lane 3, adrenal gland; lane 4, bone marrow mononuclear cells; lane 5, placenta; lane 6, nontemplate control (NTC). (B) RT-PCR of CaR and TBP in NB neuroblastic tumors. Lane 1, MWM; lane 2, kidney (positive control); lane 3, undifferentiated neuroblastoma (Undiff NB); lane 4, differentiating neuroblastoma (diff NB); lanes 5 and 6, ganglioneuroblastoma (GNB); lane 7, ganglioneuroma (GN); lane 8, NTC. (C) The percentage of neuroblastic tumors with CaR mRNA expression at diagnosis (histologic categories) and at second-look surgery.

Download figure to PowerPoint

CaR expression was assessed next in 46 primary neuroblastic tumors. CaR transcripts were not detected in the 8 undifferentiated neuroblastomas that were examined at diagnosis (0%). The full-length product was amplified in 6 of 15 (40%) poorly differentiated and differentiating neuroblastomas. The spliced form that lacked exon 5 was coamplified in a differentiating neuroblastoma (Fig. 1B). Also, full-length transcripts were detected in 5 of 9 ganglioneuroblastomas and in 5 of 9 ganglioneuromas (55.5%). The spliced form was amplified simultaneously in 2 intermixed ganglioneuroblastomas. Among 3 samples that were obtained at the time of disease recurrence, 2 were undifferentiated neuroblastomas in which CaR transcripts were not observed, and the other was a differentiating neuroblastoma in which full-length transcripts were detected.

Undifferentiated neuroblastoma was the only histologic group that contained no tumors with CaR expression (P < .001) (Fig. 1C, Table 1). CaR mRNA was correlated significantly with age at diagnosis <1 year; clinical stages 1, 2, and 4S; low clinical risk; and favorable histology (Table 1). No correlation was observed with MYCN amplification status. Patients with CaR-positive tumors had better event-free and overall survival than patients with CaR-negative tumors, but the differences between groups were not statistically significant.

Table 1. Correlation of Calcium-Sensing Receptor and Parathyroid Hormone-related Protein Messenger RNA Expression With Prognostic Factors in Neuroblastic Tumors
CharacteristicNo. of PatientsCaR mRNA Expression: No. (%)No. of PatientsPTHrP mRNA Levels (Arbitrary Units)
  1. CaR indicates the calcium-sensing receptor gene; mRNA, messenger RNA; PTHrP, the parathyroid hormone-related protein gene; NS, not significant; MYCN, v-myc myelocytomatosis viral-related oncogene.

Age at diagnosis, y    
 <186 (75)9 
 ≥1296 (20.7)27 
 P .0077 NS
Clinical stage    
 1, 2, 4S179 (52.9)18 
 3, 4193 (15.8)16 
 P .018 NS
Risk group    
 High risk152 (13.3)11 
 Low risk2311 (47.8)24 
 P .0285 NS
Histologic category    
 Undifferentiated neuroblastoma80 (0)93.79
 Poorly differentiated, differentiating neuroblastoma156 (40)123.07
 Ganglioneuroblastoma, ganglioneuroma1810 (55.6)1531.7
 P .0274 <.0001
Shimada classification    
 Favorable histology2515 (78.9)1923.82
 Unfavorable histology164 (21.1)165.71
 P .0284 .0417
MYCN status    
 Amplified5 6NS
 Nonamplified36 30 
 P NS NS

To confirm that lack of CaR cDNA amplification was caused by an absence of expression and not by a new splicing form that lacked exon 4 or 6, additional PCR reactions were performed to amplify regions that spanned exons 1B through 3 and exons 6 through 7. Amplicons of the expected sizes were detected in kidney, in adrenal gland, and in 2 tumors with CaR expression but not in 2 neuroblastomas in which CaR transcripts were not amplified previously (data not shown).

CaR Is Expressed in Neuronal and Glial Lineages of Poorly Differentiated and Differentiating Neuroblastomas, Ganglioneuroblastomas, and Ganglioneuromas

Immunohistochemical analysis of CaR was performed in 64 primary neuroblastic tumors. Undifferentiated neuroblasts mostly were negative (Fig. 2A). Instead, even the slightest degree of neuroblast differentiation displayed CaR immunostaining (Fig. 2B). Cytoplasmic reactivity with plasma membrane stain reinforcement was observed in ganglion cells (Fig. 2C). The neuropil was stained faintly, and Schwann cells were strongly positive (Fig. 2D). Undifferentiated, CaR-negative neuroblasts and positive ganglion and Schwann cells coexisted in ganglioneuroblastomas. CaR mean staining values were higher in poorly differentiated and differentiating neuroblastomas than in undifferentiated neuroblastomas and were significantly higher in ganglioneuroblastomas and ganglioneuromas than in undifferentiated neuroblastomas (P < .0001) (Fig. 2F). High CaR staining values were correlated with favorable histology (P = .0095) and clinical stages 1, 2 and 4S (P = .031). CaR mRNA and protein expression was analyzed in 31 cases (24 primary tumors, 6 postchemotherapy samples, and 1 specimen obtained at the time of disease recurrence), and the results were concordant in 26 of 31 cases (83.8%).

thumbnail image

Figure 2. Calcium-sensing receptor (CaR) immunostaining in neuroblastic tumors and normal tissues. (A) Poorly differentiated neuroblastoma (NB). (B) Differentiating NB. (C) Ganglion cells in ganglioneuroma (GN). (D) Schwannian stroma in ganglioneuroblastoma (GNB). (E) Peripheral nerve ganglia. (F) Mean CaR immunostaining value in histologic categories (asterisk indicates P < .0001). (G) CaR staining of 2 samples obtained from the same patient: undifferentiated (Undiff) NB at diagnosis and a specimen with morphologic features indicative of differentiation at second-look surgery (original magnification, ×400).

Download figure to PowerPoint

PTHrP Messenger RNA Levels Are Higher in Ganglioneuroblastomas and Ganglioneuromas Than in Neuroblastomas

PTHrP was expressed in the 36 primary neuroblastic tumors analyzed by qRT-PCR. Significantly higher levels were observed in ganglioneuroblastomas and ganglioneuromas than in neuroblastomas (P < .0001) (Table 1, Fig. 3A). Similar PTHrP expression was observed in undifferentiated, poorly differentiated, and differentiating neuroblastomas. No association was detected between PTHrP levels and other prognostic variables.

thumbnail image

Figure 3. Parathyroid hormone-related protein (PTHrP) messenger RNA (mRNA) and protein expression in neuroblastic tumors. (A) PTHrP mRNA analyzed by quantitative real-time polymerase chain reaction was greater in ganglioneuroblastomas (GNB) and ganglioneuromas (GN) (n = 15) than in neuroblastomas (NB) (n = 21); P < .0001. (B) Mean PTHrP immunostaining values in histologic categories (asterisk indicates P < .0001). (C-E) PTHrP immunostaining in (C) undifferentiated neuroblasts, (D) ganglion cells in GNB, and (E) Schwannian stroma in GN (original magnification, ×400).

Download figure to PowerPoint

It is noteworthy that PTHrP levels higher than the median value (4.98 relative units [RU]) were correlated with better overall survival. The survival rate of patients with PTHrP >4.98 RU was 100% with a mean follow-up of 51.96 months, and the survival rate of patients with PTHrP ≤4.98 RU was 61.52% with a mean follow-up of 33.27 months (95% confidence interval, 25.29%-84.23%; P = .022). However, the PTHrP level was not an independent predictor of clinical outcome in multivariate analysis (data not shown).

PTHrP Is Highly Expressed in Ganglion and Schwann Cells of Ganglioneuroblastomas and Ganglioneuromas

PTHrP immunostaining was low or absent in undifferentiated neuroblasts (Fig. 3C). Strong cytoplasmic reactivity was present in most ganglion cells (Fig. 3D) and in Schwannian stroma (Fig. 3E). In some cases, nuclei and cytoplasm were stained simultaneously (Fig. 3D). PTHrP staining was correlated with favorable histology (P = .0048), and undifferentiated neuroblastoma was the histologic group with the lowest PTHrP mean staining value (P < .0001) (Fig. 3B). In agreement with PTHrP mRNA analysis, the PTHrP values observed in undifferentiated, poorly differentiated, and differentiating neuroblastomas were not significantly different (Fig. 3B). Two undifferentiated neuroblastoma specimens that were obtained at the time of disease recurrence displayed low PTHrP expression. PTHrP mRNA and protein expression was analyzed in 20 cases (15 primary tumors, 4 postchemotherapy samples, and 1 tumor obtained at disease recurrence), and the results were concordant in 16 of 20 cases (80%).

CaR and PTHrP Are Highly Expressed in Neuroblastomas With Treatment-induced Differentiation Features

CaR transcripts were detected in 5 of 11 (45.4%) second-look samples (Fig. 1C). Morphologic features indicative of differentiation (increased cellular and cytoplasm size, ganglion cell-like morphology) and strong CaR immunoreactivity were present in their corresponding paraffin-embedded tissues (Fig. 2G). Three of these 5 patients presented with undifferentiated, CaR-negative neuroblastomas at diagnosis. The other 2 patients received treatment before their first specimen was procured. Six second-look samples without CaR expression were obtained from patients who presented with undifferentiated or poorly differentiated neuroblastomas at diagnosis and displayed minimal or no morphologic changes after treatment. Mean PTHrP mRNA levels in postchemotherapy samples (16.25 RU ± 17.97 RU) were higher than the median value in the diagnostic specimens (4.98 RU), and strong PTHrP immunostaining was observed in samples that had treatment-induced maturation features.

CaR and PTHrP Are Up-regulated Upon in Vitro Differentiation Induction

CaR and PTHrP expression initially was analyzed in 12 untreated neuroblastoma cell lines. CaR mRNA was not detected in any of the cell lines. PTHrP was expressed in all cell lines, and higher levels were detected in S-type than in N-type and I-type cell lines (Fig. 4A).

thumbnail image

Figure 4. Calcium-sensing receptor (CaR) and parathyroid hormone-related protein (PTHrP) regulation with retinoic acid treatment. (A) PTHrP levels in N-type, I-type, and S-type neuroblastoma (NB) cell lines, as assessed by quantitative real-time polymerase chain reaction (PCR). SK-N-ER cells were cultured in the presence or absence of 10 μM all-trans-retinoic acid (ATRA) for 4 days, and (B) PTHrP and (C) CaR messenger RNA levels were analyzed (asterisk indicates P < .0001). (D) Reverse transcriptase-PCR of TATA-binding protein (TBP) and CaR in kidney (positive control) and in SK-N-ER cells exposed or not exposed to retinoic acid. MWM indicates molecular weight marker; NTC, nontemplate control; bp, base pairs.

Download figure to PowerPoint

To evaluate CaR and PTHrP regulation during early phases of neuronal differentiation, 2 I-type cell lines were exposed to retinoic acid for 4 days, and the PTHrP levels did not change (Fig. 4B). Conversely, CaR expression was induced in SK-N-ER cells (Fig. 4C), and the full-length CaR transcript was detected (Fig. 4D). However, CaR was not up-regulated in SK-N-BE(2)-C cells that were exposed to the same treatment.

The SK-N-ER and SK-N-BE(2)-C cell lines were treated with BrdU for 15 days. CaR (Fig. 5A), PTHrP (Fig. 5B), and PMP22 (Fig. 5C) were up-regulated significantly in SK-N-ER cells. A 2-fold to 5-fold increase in S100-β mRNA also was observed (data not shown). The CaR full-length amplicon was the only 1 detected from Day 3 to Day 12, when, concurrent with high CaR mRNA levels, the spliced form that lacked exon 5 was coamplified (Fig. 5D).

thumbnail image

Figure 5. Calcium-sensing receptor (CaR), parathyroid hormone-related protein (PTHrP), and peripheral myelin protein 22 (PMP22) regulation with 5-bromo-2′-deoxyuridine (BrdU) treatment. SK-N-ER cells were grown in the presence or absence of 10 μM BrdU for 15 days. Cells were harvested on the days indicated, and (A) CaR, (B) PTHrP, and (C) PMP22 messenger RNA levels were analyzed (asterisks indicate P = .037). (D) The full-length CaR transcript (649 base pairs [bp]) and the spliced form that lacked exon 5 (419 bp) also were analyzed by reverse transcriptase-polymerase chain reaction. MWM indicates molecular weight marker; NTC, nontemplate control; TBP, TATA-binding protein.

Download figure to PowerPoint

Significant up-regulation of PTHrP, S100-β, and PMP22 also was induced by BrdU treatment in SK-N-BE(2)-C cells. However, CaR expression was not detected at any time point. A senescence test performed at Day 15 of BrdU exposure revealed a maximum of 15% senescent cells.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

The current data provide the first evidence to our knowledge of CaR and PTHrP expression in neuroblastic tumors. CaR was expressed in approximately 50% of poorly differentiated and differentiating neuroblastomas, ganglioneuroblastomas, and ganglioneuromas. CaR expression was correlated with good prognostic variables in neuroblastic tumors, such as age at diagnosis <1 year, low clinical stage, and low clinical risk. In addition, higher PTHrP mRNA levels were observed in ganglioneuroblastomas and ganglioneuromas than in neuroblastomas, and both genes were up-regulated upon treatment-induced tumor differentiation.

The correlation of CaR and PTHrP expression in primary neuroblastic tumors with differentiated histologic features is not a surprising finding. Although CaR activation reportedly was involved in proliferation and skeletal metastases in prostate carcinoma,15 CaR expression was correlated with differentiated histology in colon carcinoma.16 In the adult and developing brain, CaR is expressed in neurons and in the 3 glial components: astrocytes, oligodendrocytes, and microglia.25CaR is up-regulated in neural stem cells when they are specified to the oligodendrocyte lineage and, upon prolonged activation, it promotes the maturation of oligodendrocyte precursors.14 Moreover, CaR regulates PTHrP secretion in astrocytes and astrocytomas.26 We observed that CaR and PTHrP were up-regulated during neuroblastoma differentiation toward neuronal and glial lineages, but our results indicate that they may play different roles in these pathways.

First, CaR and PTHrP were highly expressed in ganglion and Schwann cells, but CaR already was present in differentiating neuroblasts. To assess whether CaR participates in neuronal differentiation at earlier stages than PTHrP, the regulation of both genes was evaluated in neuroblastoma cell lines that were exposed briefly to retinoic acid. CaR was not expressed in untreated cell lines, in keeping with the undifferentiated nature of neuroblastomas from which they are derived. Up-regulation of CaR, but not PTHrP, was observed in SK-N-ER cells after a short exposure to retinoic acid. However, CaR expression was not induced in the MYCN-amplified SK-N-BE(2)-C cell line. Whether MYCN-associated retinoid resistance27 may account for this difference remains to be elucidated.

Second, CaR was expressed in 50% of ganglioneuroblastomas and ganglioneuromas, but PTHrP levels higher than the median value were detected in almost 100% of these tumors. Moreover, untreated S-type cell lines, which display some features of Schwann-like cells, expressed higher PTHrP levels than N-type and I-type cell lines. When I-type cells were exposed to BrdU, S100-β, a Schwann-cell marker, and PMP22, which encodes a major protein component of peripheral nerve myelin, were up-regulated. This finding and a low number of senescent cells indicated differentiation induction along the glial-like lineage. Concurrently, PTHrP was up-regulated in the 2 cell lines that were examined, whereas CaR expression was induced only in the MYCN-nonamplified cell line. Together, our results suggest that CaR participates mainly at early phases of neuroblast differentiation and when neuroblastic tumors reach a certain degree of differentiation and are composed mostly of Schwannian stroma, PTHrP production may be regulated by stimuli other than CaR activation.

We have analyzed the expression of 2 alternative splicing forms of CaR that are regulated differently and that play different roles during keratinocyte differentiation11 and in growth plate chondrocytes.28 In keratinocytes, only the full-length CaR is able to mediate calcium-stimulated inositol phosphate production, and the spliced form that lacks exon 5 moderately reduces the inositol phosphate response of the full-length CaR.11 Conversely, in growth plate chondrocytes, the spliced form that lacks exon 5 is able to function and even compensates for the full-length absence in CaR−/− mice.28 In our in vitro differentiation assays, the full-length transcript was the only mRNA present at early time points of BrdU treatment, when CaR expression levels were low or moderate. Then, the spliced form that lacks exon 5 was coamplified when the highest levels of CaR were reached, on Day 12. This peak of expression was concurrent with that of 2 glial differentiation markers, indicating that both CaR transcripts coexisted only at advanced stages of the differentiation process. Further investigations with a functional approach are needed to analyze whether both CaR transcripts and their encoded proteins have distinct roles in neuroblastoma.

In summary, we report for the first time that CaR and PTHrP are expressed in differentiated, favorable neuroblastic tumors. Both genes are up-regulated in treatment-induced, differentiated tumors and upon in vitro differentiation induction. A precise picture of CaR and PTHrP involvement in the neuroblastoma differentiation processes is necessary to evaluate whether their pharmacologic modulation could represent a new differentiating therapy for neuroblastoma.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

Supported by the Spanish Ministry of Health (RITSI G03/089, FIS2004/PI041259), the Spanish Association Against Cancer, and the Catalan government (2005SGR00605).

The authors made no disclosures.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References