CELL LINES AND ANIMAL MODELS

Meningiomas arise from the meningothelial (arachnoidal cap) cells. They constitute approximately 20% of all intracranial tumors. About 85% of meningiomas are benign (grade I), while 10% are atypical (grade II), and 3% to 5% are malignant (anaplastic, grade III). Tumor models are critical to understand the function of genetic alterations in cancer, to evaluate their contribution to tumor initiation and progression, and to assess the toxicity and therapeutic efficacy of established and newly developed treatments. A major obstacle to research in meningioma biology and further therapeutic development has been the limited availability of long-term cell lines and animal models, which recapitulate the human clinical scenario. There has been and remains a continuing challenge to establish and maintain benign meningioma cell lines that retain the original phenotype both in vitro and in vivo similar to the original primary tumor specimen. An additional hindrance has been a lack of animal models with sufficiently high frequencies of spontaneous meningioma development and difficulties in implanting intracranial dural-based tumors to accurately model the human condition. These issues are further compounded by the lack of meningeal specific promoters. The LTAg2B was the first human leptomeningeal cell line available, although immortalization by transfection with an SV40 large T antigen gene construct under the control of the Rous sarcoma virus promoter. This cell line (LTAg2B) maintained the morphology of primary cultures and stained positive by immunohistochemistry for arachnoid-specific markers at early passages.¹ The loss of intermediate filament staining in later passages may reflect a change in the phenotype of the cells owing to their high proliferative rate.

There are a few human and rat meningioma cell lines that have been established from malignant meningiomas.^2–4 These cell lines have been derived from aggressive variants of meningioma, which represent only a small portion of meningiomas. There is a critical need to have additional cell lines that span the entire spectrum of meningioma histologic grades.

Most animal models have been developed by implanting human meningioma cells derived from a primary tumor subcutaneously into athymic mice. Grade I meningiomas grow very slowly with a large amount of variability in this environment.⁵ Recently, a novel genetic model was devised utilizing Cre recombinase technology to specifically inactivate NF2 in arachnoidal cells, resulting in the formation of intracranial meningothelial hyperplasia and meningiomas in roughly 30% of the mice.⁶ This powerful new technology significantly improves on prior models and may open avenues of investigation never before possible in meningioma research. This model is discussed in the latter part of this chapter.

CELL CULTURE

Cell culture provides a tool that is reliable, convenient, and, most important, reproducible for investigations into the biology of human cancers. A valid meningioma culture needs to fulfill the following requirements: (1) It must be composed of meningioma cells; (2) the growth rate of the cells should be predictable, reliable, and reproducible; and (3) the therapeutic response of the cells must be similar to that of human meningiomas. Studies clearly show that early passage cell lines (3–5) derived from benign human meningiomas undergo cellular senescence after only a few passages.^7–12 Therefore any type of biological studies must be done quickly with the limited number of available cells. Usually the studies conducted on these cells are not reproducible because the cells cannot be frozen and then thawed; thus not allowing studies to be repeated in different laboratories under varied conditions. To overcome this problem, three separate studies have used hTERT to immortalize cells.

hTERT and Meningioma Cell Immortalization

The major reason for the difficulty in generating established cell lines from benign meningiomas is when the cells from these tumors are put into culture they senesce. Many pathways are involved in cellular senescence and disruption of these pathways is required for continued cell growth and immortalization.⁹ Low or absent telomerase activity is in part one of the reasons for senescence; progressive shortening of telomerase occurs with each subsequent cell cycle. Telomeres are repetitive DNA sequences at the ends of chromosomal DNA. Telomerase is a ribonucleoprotein polymerase enzyme, composed of a integral RNA component (hTR) and several other proteins, including a reverse transcriptase subunit (hTERT). Telomerase activity is thought to be essential for maintaining chromosome ends by adding repetitive TTAGGG sequences to the telomeres; it is linked to cellular immortality.¹³ Cell cycle checkpoint pathways also control cellular senescence, and successful cell immortalization may also involve disruption of the p53 and/or Rb pathways.¹⁴ In meningiomas studies have shown that telomerase activity is directly related to tumor malignancy.^7,8,15 Benign tumors have low or absent telomerase activity; whereas, the majority of malignant tumors are positive. In addition, hTERT-positive meningiomas have a higher incidence of recurrence, suggesting it use a potential predictor for recurrence.¹⁶ There are three reports, including one from our group, establishing immortalized meningioma cell lines derived from grade I tumors by transduction with hTERT, the catalytic subunit of telomerase.^17–19 After immortalization, the morphologic features of the cell lines remained the same as the corresponding nontransduced cells (Fig. 63-1). On seeding of the primary cells, large nuclei and tight cell junctions were observed. Cells remained positive for cytokeratin, vimentin, and S-100 by immunohistochemistry (Fig. 63-2). There is some controversy as to whether transduction with hTERT alone is sufficient for immortalization of benign meningiomas. Baia and colleagues¹⁷report that immortalization of two benign meningiomas cell lines required disruption of the p53 and pRb pathways in addition to the expression of hTERT. In contrast, Puttmann and colleagues and Cargioli and colleagues^18,19 reported that transduction with hTERT alone is sufficient for immortalization of benign meningioma. This most likely reflects the genetic alterations in the original tumors from which these cell lines were derived. In addition, the doubling time of the cells transduced with both SV40 large T antigen and hTERT was shorter than the cells transduced with hTERT only. Transplantation of the hTERT-immortalized cells subdurally into nude mice resulted in tumors with typical histologic features of meningioma. Both macroscopically and microscopically, the tumors showed features of meningioma (Fig. 63-3A, B). Grossly, the tumors were well-demarcated, firm, nodular masses of tan to white tissue adherent to the dura and leptomeninges on the brain surface. Microscopically, the tumor tissue was densely cellular and was composed of spindled to epithelioid cells with monomorphic round to oval nuclei and moderate amounts of cytoplasm. The tumors showed growth patterns that varied from small nodular meningothelial-like clusters to more fascicular spindled growth. Focal whorls and microcalcifications were present in some tumors. Although nuclei displayed some pleomorphism and had visible nucleoli, other atypical features such as small cells with a high nuclear-to-cytoplasmic ratio, brisk mitotic activity, and sheet-like growth were not observed. One tumor showed focal areas of nuclear condensation and fragmentation consistent with necrosis. Immunohistochemically, the tumors showed weak positive staining for EMA and focal weak staining for S-100 (Fig. 64-3C, D). It should be noted that the histopathologic examination revealed that the cell line transduced with both SV40 large T antigen and hTERT had some atypical changes when compared with the cells transduced with hTERT only. In conclusion, these studies show that genetic manipulation can be used to overcome the senescence of primary benign meningiomas and the cell lines generated can serve as useful model systems for biological studies and evaluation of novel therapies on meningioma.

FIGURE 63-1 Phase-contrast images of meningioma cells in culture. (A) Me3 at P3; (B) Me3TSC at P9; (C) Me10 at P1; and (D) Me10T at P2. Original magnification, ×40.

(From Cargioli TG, Ugur HC, Ramakrishna N, et al. Establishment of an in vivo meningioma model with hTERT. Neurosurgery 2007;60:750–60.)

FIGURE 63-2 Immunofluorescent staining of meningioma cells with cytokeratin, vimentin, and S100 antibodies. Transduced cell lines maintained expression even at later passages.

(From Cargioli TG, Ugur HC, Ramakrishna N, et al. Establishment of an in vivo meningioma model with hTERT. Neurosurgery 2007;60:750–60.)

FIGURE 63-3 A, Photograph showing a grossly visible tumor attached to the cortex in the posterior parietal region of the left hemisphere. The tumor is well demarcated, firm, and white, and conforms macroscopically to meningioma characteristics observed in humans. B, Photomicrograph showing tumor tissue that is densely cellular and composed of spindled to epithelioid cells with monomorphic round to oval nuclei and moderate amounts of cytoplasm. The tumor volume was 2.03 mm³ (hematoxylin and eosin; original magnification, ×200). C, Photomicrograph showing that the tumor is weakly positive for EMA; original magnification, ×400. D, Photomicrograph showing that the tumor is weakly positive for S-100; original magnification, ×400.

(From Cargioli TG, Ugur HC, Ramakrishna N, et al. Establishment of an in vivo meningioma model with hTERT. Neurosurgery 2007;60:750–60.)

MALIGNANT MENINGIOMA CELL LINES

There have been several long-term cell lines, which have been established from malignant meningiomas.²⁰ The most commonly used line was initiated from a tumor of a 61–year-old man with repeated recurrent primary intraosseous malignant meningioma of the skull.² The doubling time of cultured cells was reported to be 62 hours. The mesenchymal tumor marker vimentin was identified by immunohistochemistry. Tumorigenicity in athymic nude mice was observed; with a tumor doubling time of 5 days. The ki-67 labeling index of cultured cells and xeonografted tumors was approximately 36% and 30%, respectively. The tumor retained its ultrastructural features including whorl formation patterns in vivo. This line has been useful to test the therapeutic efficacy of antitumor agents. An additional cell line, KT21-MG, has been established from a human malignant meningioma with amplified c-myc oncogene.³ Ultrastructurally, interdigitation of adjacent cell membranes with desmosomes was seen in the cultured cells that were observed in the original tumor. Immunohistochemical analysis showed the cells were positive for vimentin. The cell line was able to produce tumors in nude mice, proliferate under low serum conditions, anchorage independent growth in soft agar and many chromosomal abnormalities including monosomy of chromosome 22. Further, c-myc was found to be amplified in the original tumor and xenografts.

GENE EXPRESSION PROFILING: TISSUE VERSUS CELL CULTURE

Well established cell lines are commonly used for cell biology studies of cancer. As mentioned earlier, there are very few established cell lines for meningiomas. Most in vitro studies have been performed on primary meningioma cultures of the tumors. Studies in malignant tumors suggest the gene expression profiles of cultured tumor cells are distinct from the corresponding tumor.²¹ In addition, they are affected by culture conditions and number of passages. Sasaki and colleagues²² performed gene expression profiling using GeneChip on original frozen tumor specimens and primary cultures established from the corresponding tumors at two time points from 3 meningiomas (one meningioma of each World Health Organization Grade [WHO] Grades I, II, and III). Among the 12,000 genes examined, 51 genes were up regulated by fivefold or more and 19 genes were down-regulated by twofold or more in primary cultures compared with corresponding frozen tissue. The expression levels of a subset of these genes were confirmed by quantitative real-time transcription polymerase chain reaction. Increased expression of genes for extracellular matrix proteins, cytoskeletal proteins, and cell surface markers were noted in the primary cultures. This study does have several limitations including the small sample size; the original tumor tissue contains fibroblasts, endothelial cell, leukocytes and other cell types, which may confound the results. This article does serve as a warning that cell culture genetic studies may not accurately reflect the genetic milieu of a tumor in vivo. Therefore treatment studies in vitro may not directly correlate with treatment effect in vivo.

MATRIGEL AND XENOGRAFT AUGMENTATION

An in vivo model of meningioma cell growth is essential for a better understanding of meningioma biology and as a model for testing the efficacy of novel therapeutics before initiating clinical trials. The first transplantation of human meningiomas was described in 1945.²³ In these studies, tumors were injected into the eyes of guinea pigs. Successful subcutaneous implantation of human malignant meningioma was first reported in 1977.²⁴ This was followed by a study that reported the successful serial passage of transplanted meningioma that remained histologically unchanged. Implantation of meningioma cells in the subrenal capsule of nude mice has the most successful tumor take.²⁵ The subrenal capsule is a very different microenvironment than the brain and does not allow for serial tumor measurements. Matrigel has enhanced the growth of a number of cell lines in vivo; it is a substance derived from a mouse sarcoma and is a mixture of basement membrane proteins and growth factors. A high tumor take has been demonstrated by subcutaneous meningioma cells mixed with Matrigel at the time of injection. The histology of the xenografted tumor closely resembled that of the original human tumor.²⁶ By immunohistochemistry, the tumors were positive for human fibronectin and collagen IV and negative for human or mouse laminin. About half the tumors were immunohistochemically positive for epithelial membrane antigen. The reason for the enhancement of the growth of the cells in the presence of Matrigel may be the enhancement of growth of the tumor cells and the concentration of cellular and circulating growth factors. Thus far, the majority of preclinical trials have been conducted in xenograft mouse meningioma model.²⁷ Recently, a mouse model of intracranial meningioma that use the novel nonivasive bioluminescent imaging system has been developed.^28,29 However, whereas xenograft models are straightforward and relatively easy to use, their accuracy in predicting the efficacy of an anticancer agent in patients is questionable. This may be secondary to cell line selection, the environment or location of the xenograft transplant, or the lack of de novo tumor development.

MOUSE MODELING IN MENINGIOMA PRECLINICAL AND TRANSLATIONAL RESEARCH

With the identification of molecular pathways involved in tumorigenesis, novel models of cancer have been developed to understand tumorigenesis, evaluate potential therapies, and ultimately impact on patient outcome. The mouse has provided an excellent platform for modeling cancer in a mammalian system. Currently, mouse modeling of cancer is possible through expression of oncogenes, specific genetic mutations, or the inactivation of tumor suppressor genes. These experiments have begun to provide us with an understanding of the molecular pathways involved in tumor initiation and progression. Mouse models recapitulate human tumors by de novo tumor development, preserving the organ microenvironment, and thus may provide an excellent setting to evaluate the efficacy of molecular targeted therapies and identify new potential targets in a system with few additional genetic changes. Another powerful tool that can be easily derived from accurate mouse models of meningioma is represented by primary cells such as arachnoidal cells, which is the purest genetic system to evaluate therapies and perform high-throughput drug screenings. Broadly, genetically engineered mouse models of cancer encompass mice overexpressing a transgene (oncogene or point mutation), knock-in models of genetic point mutations, as well as complete knock-out and conditional knock-out models using the Cre-loxP system.

Individuals with neurofibromatosis type 2 (NF2) are at significantly elevated risk for developing meningiomas,³⁰ suggesting that the NF2 gene might play a central role in regulating arachnoidal cell proliferation as a tumor suppressor gene. Biallelic inactivation of the NF2 gene has been identified in 30% to 70% of sporadic meningiomas, leading to loss of expression of the NF2 gene product, merlin.³¹ In addition, NF2 inactivation is likely an early event in sporadic meningioma pathogenesis and is observed as frequently in grade I meningiomas as it is in high-grade tumors.³²

In an effort to define the role of NF2 loss in meningioma tumorigenesis, a series of mouse models have been generated.

Although cancer prone, heterozygous traditional Nf2 knock-out mice (Nf2^+/-) do not develop meningiomas, but rather die with osteosarcomas and other tumor types not found in humans with NF2. Nf2^-/- mice die early during embryonic development of a failure to initiate gastrulation.^33,34

To examine the consequences of Nf2 loss in selected cell types of adult mice, a conditional knock-out allele (Nf2^flox2) has been generated. It was clearly demonstrated that conditional biallelic Nf2 inactivation is a rate-limiting step in murine Schwann cell tumorigenesis by using the P0 promoter to specifically express the Cre recombinase in Schwann cells.³⁴ Remarkably, meningioma was not observed in these mice, suggesting that Cre recombinase expressed from a Schwann cell-specific promoter does not affect meningioma progenitor cells.

Because of the lack of arachnoidal cell specific promoter, an alternative approach was used for delivery of Cre recombinase by direct injection of a recombinant Cre adenovirus.⁶ This approach also had the advantage of targeting the initiating genetic lesion (in this case, homozygous inactivation of Nf2) to a small population of susceptible arachnoidal cells, which is likely to model human cancer more accurately than when all of the cells in a targeted tissue are mutated. Direct injection of recombinant Cre adenovirus into the cerebrospinal fluid (CSF) circulation in Nf2^flox2 pups allowed targeting of leptomeninges. Beginning at four months of age, 30% of mice with arachnoidal cell Cre-mediated excision of Nf2 exon 2 developed a range of benign meningiomas subtypes histologically similar to the human tumors: transitional, meningothelial and fibroblastic⁶ (Fig. 63-4). In addition, similar to human patients in whom histologically benign meningiomas may invade brain, mouse meningiomas showed a similar pattern with tumors cells infiltrating the adjacent cerebral parenchyma causing reactive astrogliosis. Because of adenovirus-Cre diffusion through the CSF circulation, spinal meningiomas with spinal cord compression were also observed.

FIGURE 63-4 Histologic characterization of murine Nf2^-/- meningiomas after transorbital or subdural adCre injection in Nf2^flox2/flox2 mouse pups. A, Hematoxylin and eosin stained section showing a transitional meningioma en plaque overlying the brain stem. B, C, Transitional meningioma covering the right trigeminal nerve. Some cells have prominent nucleoli, reminiscent of atypical features in human tumors (arrows).

(From Kalamarides M, Stemmer-Rachamimov AO, Takahashi M, et al. Natural history of meningioma development in mice reveals a synergy of Nf2 and p16 (Ink4a) mutations. Brain Pathol 2008;18:62–70.)

These results indicated that Cre-loxP-mediated inactivation of both Nf2^flox2 alleles is the rate-limiting step for meningioma development. In this meningioma model the first lesion associated with murine leptomeningeal tumorigenesis was meningothelial proliferation referring to very small, microscopic lesions composed of meningothelial cells that represent early tumor formation. Additional hemizygosity for p53 did not modify meningioma frequency or malignancy suggesting that, as in humans, Nf2 and p53 mutations do not synergize in murine meningeal tumorigenesis.

So far, only few of the individual genes affected by the chromosomal alterations found in atypical and malignant meningiomas are known. Deletion of the p16^INK4a/p14^ARF locus or monosomy of chromosome 9 was found in the majority of higher-grade meningiomas.^32,35

In the mouse Nf2 meningioma model additional nullizygosity for p16^Ink4a increased the frequency of meningioma and meningothelial proliferation without modifying the tumor grade,³⁶ suggesting that loss of the p14^Arf rather than p16^Ink4a component of the locus is critical for malignant transformation of meningiomas.

In addition, magnetic resonance imaging (MRI) can be used to screen for meningothelial proliferation and meningioma development in mice, a crucial step in the translation in this mouse meningiomas model with low tumor penetrance and relatively slow growth rate into a preclinical model (Figs. 63-5 and 63-6). Monitoring meningioma growth by MRI opens the way to future studies in which therapeutic interventions can be tested as preclinical assessment of the potential for clinical application.

FIGURE 63-5 Murine intracranial meningioma show common MRI features of human meningiomas. Some examples of coronal Gd-DTPA-enhanced T1-weighted MR images after transorbital (A, B) or subdural (C, D) adCre injection. A, Arrowhead indicates lateral round Gd-DTPA enhancement corresponding to a fibroblastic meningioma associated with hydrocephalus characterized by enlarged ventricles (asterisks). B, Arrowhead indicates abnormal, thick Gd-DTPA enhancement corresponding to a histologically confirmed meningothelial meningioma that would be defined as olfactory meningioma in human pathology. A hypointense signal corresponding to an osteoma is indicated by an asterisk. C, Meningioma en plaque nested within areas of meningothelial proliferation revealed by lateral linear thick Gd-DTPA enhancement (arrowhead). D, Arrowhead shows intense homogeneous Gd-DTPA enhancement above right trigeminal nerve corresponding to a histologically confirmed right supra-trigeminal transitional meningioma.

(From Kalamarides M, Stemmer-Rachamimov AO, Takahashi M, et al. Natural history of meningioma development in mice reveals a synergy of Nf2 and p16 (Ink4a) mutations. Brain Pathol 2008;18:62–70.)

FIGURE 63-6 Murine intraspinal meningioma show common MRI features of human spinal meningioma. This subdural-injected adCre;Nf2^flox2/flox2;Ink4a*/* 8-month-old mouse rapidly developed paraparesis. A, Axial T1-weighted image showing a large, histologically confirmed, Gd-DTPA-enhanced fibroblastic meningioma (arrowhead) compressing the spinal cord (asterisk). B, Remarkably, a tailing of the meningioma along the duramater (arrowhead) was present at a lower level (asterisk indicates spinal cord). A dural tail is often noted associated with human meningiomas and known as “dural tail sign.” C, Sagittal view of spinal cord (asterisk) showing the spinal meningioma (arrowhead).

(From Kalamarides M, Stemmer-Rachamimov AO, Takahashi M, et al. Natural history of meningioma development in mice reveals a synergy of Nf2 and p16 (Ink4a) mutations. Brain Pathol 2008;18:62–70.)