CHL1 and Nrcam Are Primarily Expressed in Low Grade Pediatric
Total Page:16
File Type:pdf, Size:1020Kb
Open Med. 2019; 14: 920-927 Research Article Robin Wachowiak, Steffi Mayer, Anne Suttkus, Illya Martynov, Martin Lacher, Nathaniel Melling, Jakob R. Izbicki, Michael Tachezy CHL1 and NrCAM are primarily expressed in low grade pediatric neuroblastoma https://doi.org/10.1515/med-2019-0109 Keywords: CHL1; NrCAM; Neuroblastoma; Immunohisto- received November 7, 2018; accepted October 19, 2019 chemistry; Tumor markers; Neuropathology Abstract: Background. Neural cell adhesion molecules like close homolog of L1 protein (CHL1) and neuronal glia related cell adhesion molecule (NrCAM) play an impor- tant role in development and regeneration of the central 1 Introduction nervous system. However, they are also associated with Neuroblastoma is an embryonic malignancy deriving cancerogenesis and progression in adult malignancies, from neural crest cells that undergo rapid differentia- thus gain increasing importance in cancer research. We tion during fetal development. As the transition from therefore studied the expression of CHL1 and NrCAM normal to malignant tissue can occur in multiple steps, according to the course of disease in children with neu- its phenotype is highly heterogeneous [1]. Although pro- roblastoma. gress has been made in the treatment of neuroblastoma, Methods. CHL1 and NrCAM expression levels were histo- the outcome of children at high risk remains poor with a logically assessed by tissue microarrays from surgically long-term survival as low as 50 % [2]. Different parameters resected neuroblastoma specimens of 56 children. Expres- such as age, stage and chromosomal aberrations have an sion of both markers was correlated to demographics as impact on prognosis. Still, there is an ongoing need for well as clinical data including metastatic dissemination tumor markers, which allow a better determination of the and survival. individual prognosis and therapeutic monitoring espe- cially in high risk patients. Results. CHL1 was expressed in 9% and NrCAM in 51% of Recently, cell adhesion molecules (CAMs) have been neuroblastoma tissue samples. Expression of CHL1 was implicated in the processes of malignant transformation higher in patients with low Hughes grade 1a/b (p=0.01). and progression [3]. In particular, members of the immu- NrCAM was more often detected in patients with a low noglobulin superfamily L1, including transmembrane International Staging System (INSS) score 1/2 (p=0.04). proteins L1CAM, close homologue of L1 (CHL1), neuron Conclusion. CHL1 and NrCAM expression was associated glia related CAM (NrCAM) and neurofascin have been with low-grade pediatric neuroblastoma. These adhesion identified as important factors [4, 5]. Various studies molecules may play a role in early tumor development of emphasized the importance of L1CAMs in neuronal migra- neuroblastoma. tion and survival, axon outgrowth, synaptic plasticity, and regeneration, resulting in a regular development and homoeostasis of the central nervous system (CNS) [6]. Alterations in the expression and function of CAMs, *Corresponding author: Robin Wachowiak,Department of Pediatric especially L1, CHL1 and NrCAM, have been associated Surgery, University Hospital Leipzig, Liebigstrasse 20 A, 04103 Leip- with pathological processes leading to different malig- zig, Germany, Phone: 0049 3419726400, Fax: 0049 3419726249, nancies including melanoma, prostate and colon cancer E-Mail: [email protected] [7-12]. CHL1 can be down- or upregulated in human cancer Steffi Mayer, Anne Suttkus, Illya Martynov, Martin Lacher, De- partment of Pediatric Surgery, University Hospital Leipzig, Leipzig, genesis [13]. An increased expression of NrCAM has been 04103, Germany correlated to metastatic development and tumorigenesis Nathaniel Melling, Jakob R. Izbicki, Michael Tachezy, Department in different human cancers [7, 14, 15]. Moreover, overex- of General, Visceral and Thoracic Surgery, University Medical Center pression of NrCAM elevates cancer cell motility and inva- Hamburg Eppendorf, Hamburg, 20246, Germany siveness in vitro and has been linked to a poor prognosis Open Access. © 2019 Robin Wachowiak et al., published by De Gruyter. This work is licensed under the Creative Commons Attribution 4.0 License. CHL1 and NrCAM in pediatric neuroblastoma 921 in adult patients [7, 14]. Likewise, overexpression of L1 in paraffin block using a semi-automated tissue arrayer. Sub- adults correlates to tumor progression and metastatic dis- sequently, 5 µm slides of the complete tissue microarray semination in glioma, melanoma, ovarial and colon carci- (TMA) were cut using the paraffin sectioning aid system nomas [12]. In contrast, an increased expression of NrCAM (Instrumentics, Hackensack, NJ, USA). and L1 in gene array analyses has been associated with a favorable outcome in pediatric neuroblastoma [16, 17]. Taken together, members of the immunoglobulin 2.3 Immunohistochemistry superfamily L1, which share a similar structure with a 35-45% homology, might serve as interesting prognos- For immunohistochemistry, 5-µm sections were placed tic markers in neuroblastoma. The aim of the study was on precoated slides (3-triethoxysilylpropylamin; Merck, to investigate members of the L1 family with regards to Darmstadt, Germany), deparaffinized and exposed to their diagnostic and prognostic potential in this pediatric heat-induced antigen retrieval for 5 minutes in an auto- tumor. We therefore determined the expression of CHL1 clave at 121°C in Tris-EDTA-Citrate buffer, pH 7.8. After- and NrCAM by immunohistochemistry in a neuroblas- wards, the primary antibody either specific for CHL1 (goat, toma tissue microarray and correlated it to the individual polyclonal antibody: AF2126, R&D Systems, MN, USA) or course of disease. NrCAM (goat anti-human NrCAM antibody: AF2034, R&D Systems, MN, USA,) was applied at 37°C and pH 9.1 for 60 minutes. Bound antibodies were then visualized using the EnVision Kit (Dako, Glostrup, Denmark) according to the 2 Material and Methods manufacturer’s directions. Unaffected pancreatic tissue served as positive and lymphoid as negative controls. 2.1 Study design The study was approved by the Ethics Committee of 2.4 Quantification of staining intensities the Chamber of Physicians in Hamburg, Germany. The research related to human has been complied with all Staining intensities of positive tumor cells were assessed relevant national regulations, institutional policies and as described recently [19]. In brief, lack of staining was in accordance to the tenets of the Helsinki Declaration. defined negative, while weak, moderate and strong stain- Written informed consent was obtained from all parents ing were defined positive. Labeled sections were ana- for investigation of resected neuroblastoma tissue lyzed by two independent investigators (RW and MT) that samples. Pediatric patients who underwent surgical treat- were blinded to the patient’s identity or clinical status. A ment of neuroblastoma at the University Medical Center pathologist was involved in cases of discrepancy to reach Hamburg Eppendorf between November 1999 and October a consensus. 2004 were included. No preselection was performed and none of the children was pretreated. Clinical and patho- logical data included the International Neuroblastoma 2.5 Statistical Analysis Staging System (INSS), histological grade (according to Hughes), N-myc amplification, loss of heterozygosity of Statistical analysis was performed by SPSS for Windows chromosome 1p (LOH 1p), age at diagnosis, sex, metastatic version 11.5 (SPSS, IBM Corporation, NY, USA) and Graph- dissemination and event free as well as overall survival. Pad Prism (Version 7.04, GraphPad Software, Inc., San Diego, CA, USA). Chi square test was used to compare cat- egorical variables, Fischer’s exact test to compare odds 2.2 Tissue Microarray between two groups and Kruskal-Wallis test for compar- isons of continuous variables. Categorical variables are Pediatric neuroblastoma tissues were fixed in 4% buff- expressed as frequency and percentage; continuous var- ered formalin and embedded in paraffin as described iables are represented as medians with maximum and previously [18]. Hematoxylin-eosin stained sections were minimum or as means with standard deviation. Kaplan- cut from primary tumor blocks, containing representative Meier survival curves were analyzed using the log-rank tumor regions. Afterwards, tissue cylinders with a diam- test. Significance level was set as p<0.05. eter of 600 µm were used to stamp out selected sections of the original donor block. These were arrayed on a new 922 R. Wachowiak et al. 3 Results 3.2 Expression of CHL1 Immunohistochemistry revealed a heterogonous staining 3.1 Study population pattern of CHL1 inside cancerous lesions with predomi- nant membranous expression. Expression of CHL1 was A total of 56 children (24 female and 32 male) who under- detected in five (9%) samples (Fig. 1A /1B). Each CHL1 went surgical resection of neuroblastoma were included positive tumor was classified Hughes grade 1, which dif- in the study. Median age at the time of diagnosis was 30 fered significantly from CHL1 negative samples (p=0.01). months. The tumor was localized adrenally in 34 (61%) In CHL1 negative samples, Hughes grade 1 (n=17; 33%), 2 and non-adrenally in 22 (39%) patients. 28 patients (50%) (n=15; 29%) and 3 (n=19; 37%) were equally distributed had metastatic dissemination at the time of operation. (p>0.05). Comparing CHL1 positive and negative tumor Median follow-up of all patients was 72 months. During samples,