New models for MPNST: establishment and comprehensive characterization of two tumor cell lines

Abstract

Background: Malignant peripheral nerve sheath tumors (MPNSTs) are rare, invasive, and aggressive soft tissue sarcomas arising from peripheral nerves. They may occur sporadically or in association with Neurofibromatosis type 1 (NF1), in which they are the leading cause of mortality. Currently, there are no effective therapies other than surgery. Therefore, tumor-derived cell lines are essential for testing new therapeutic strategies, especially when used in parallel with in vivo models. In this study, we present two new MPNST cell lines and two patient-derived orthotopic xenograft (PDOX) models from a sporadic (SP-10) and an NF1-related (NF1-18B) MPNST patient to increase the number of available preclinical models for in vitro and in vivo drug testing.

Methods: The cell lines were isolated and extensively characterized genetically (tumor suppressor gene mutation status, DNA content), phenotypically (cell morphology, marker expression), and functionally (proliferation rate, colony formation capacity, migration rate, tumorigenic ability). We validated the models by comparing the genomic (copy number variation profile) and histological characteristics of the cell lines and PDOX tumors with their corresponding patient tumors.

Results: The new cell lines and PDOXs tumors exhibited similar genomic copy number variation profiles, histological patterns, and marker expressions as the patient tumors, validating them as faithful models. Interestingly, the NF1-18B cell model presented two cell subpopulations with different ploidy states (one < 3n and the other 4n) and functional features in vitro. NF1-18B 4n, along with SP-10 cell lines, exhibited in vitro functional hallmarks of MPNSTs, including high proliferation and migration rates and colony forming ability. However, only the SP-10 model exhibited aggressive tumorigenicity in athymic mice. In contrast, the NF1-18B < 3n showed a low migration rate and did not form colonies or aggregates in vitro.

Conclusions: The newly established MPNST cell lines, along with their corresponding PDOX models, serve as valuable tools for both in vitro and in vivo testing of novel therapeutic agents. Notably, the SP-10 cell line model represents one of the few documented cases isolated from a genuine "classic" MPNST.

Keywords: Cellular model; Malignant peripheral nerve sheath tumors; NF1; Neurofibromatosis type 1; Sporadic; and patient-derived xenograft.

Fig. 1

Phenotypic characterization of newly established MPNST cell lines. (A) DNA content analysis of the tumor cell lines, plotted as number of cells versus amount of DNA, and performed at low and high passages. The NF1-18B cell line shows aneuploidy below 3n at low passages and it is tetraploid (4n) at high passages. (B) Representative morphological images of NF1-18B (“<3n” and “4n”) and SP-10 cell lines at low and high confluence. (C) Representative immunofluorescence images of SOX9, S100B, and H3K27me3 markers. DAPI nuclear staining is shown in blue. Fibroblasts were used as positive control for expression of SOX9, SP-01 melanoma cell line was used as positive control for S100B, and NF1-09 MPNST cell line was used as positive control for H3K27me3 marker (Creus-Bachiller et al., 27). C+: Positive control. The magnification of the images in B and C is 20X. The scale bar is 150 μm

Fig. 2

Functional characterization of the newly established MPNST cell lines. A. Cell growth curves of the three cell lines (NF1-18B “< 3n”, NF1-18B “4n”, and SP-10) plotted as MTT absorbance values every 24 h. Growth curves were generated using the mean values (N = 6) ± standard deviation (SD, error bars). Population doubling time values are calculated in hours. B. Representative images of the 2-D colonies generated by the cell lines in a single well of a 12-well plate. Images were taken without amplification. C. Representative images of cell aggregates obtained in the hanging drop assay. Images were taken at 24 and 48 hours at 4X magnification. The scale bar is 50 µm. D. Representative images of the wound healing assay. Images of wound closure were taken at 0, 12, and 24 hours at 100X magnification. The scale bar is 200 µm. E. Representative images of Hematoxilin & Eosin (H&E) and immunostaining for Ki67 and vimentin in orthotopic cell line-derived tumors. NF1-18B “4n” did not generate tumors in mice. Images were taken at 4X magnification. The scale bar is 50 µm. F. Relative tumor volume growth over two weeks in mice orthotopically and subcutaneously engrafted with a cell line derived SP-10 tumor. Growth curves were generated using the mean values (N = 3) ± standard deviation (SD, error bars)

Fig. 3

Patient-derived cell lines and PDOX recapitulate the major genomic features of primary SP-10 (A) and NF1-18B (B) tumors. Copy number (CN) variation profile represented by BAF and LRR of patient tumors, PDOX tumors, and cell lines. In the case of the NF1-18B cell line, the CN profile of the two NF1-18B cell line subpopulations (“< 3n” and “4n”) is shown. CN variation is represented by a colored line below the LRR: gray for 2n region; yellow to red for chromosomal gain regions (> 2n); and green to dark green for chromosomal loss regions (< 2). LOH regions are shown in blue. Major genomic differences between the primary tumors and the models are highlighted in purple. A calculation of the mean ploidy of the sample is shown on the right side. BAF: B-allele frequency; LRR: Log-R ratio; LOH: loss of heterozygosity

Fig. 4

Cell lines and PDOX tumors recapitulate the main histological features of primary tumors. (A) Representative images of Hematoxilin & Eosin (H&E) staining of primary and PDOX tumors. (B) Representative images of immunostaining of SOX10, S100B, H3K27me3, Ki67, and Vimentin in patient primary tumors, PDOX tumors, and cell lines. SP-01 melanoma cell line was used as positive control for S100B and SOX10, and NF1-09 MPNST cell line was used as positive control for H3K27me3 marker (Creus-Bachiller et al., 2023). Images were taken at 4X magnification. The scale bar is 50 μm. C+: Positive control