Expression of ZIC family genes in meningiomas and other brain tumors
© Aruga et al; licensee BioMed Central Ltd. 2010
Received: 15 September 2009
Accepted: 3 March 2010
Published: 3 March 2010
Zic zinc finger proteins are present in the developing rodent meninges and are required for cell proliferation and differentiation of meningeal progenitors. Although human ZIC genes are known to be molecular markers for medulloblastomas, their expression in meningioma has not been addressed to date.
We examined the mRNA and protein expression of human ZIC1, ZIC2, ZIC3, ZIC4 and ZIC5 genes in meningiomas in comparison to other brain tumors, using RT-PCR, analysis of published microarray data, and immunostaining.
ZIC1, ZIC2 and ZIC5 transcript levels in meningiomas were higher than those in whole brain or normal dura mater, whereas all five ZIC genes were abundantly expressed in medulloblastomas. The expression level of ZIC1 in public microarray data was greater in meningiomas classified as World Health Organization Grade II (atypical) than those classified as Grade I (benign). Immunoscreening using anti-ZIC antibodies revealed that 23 out of 23 meningioma cases were ZIC1/2/3/5-immunopositive. By comparison, nuclear staining by the anti-ZIC4 antibody was not observed in any meningioma case, but was strongly detected in all four medulloblastomas. ZIC-positive meningiomas included meningothelial, fibrous, transitional, and psammomatous histological subtypes. In normal meninges, ZIC-like immunoreactivities were detected in vimentin-expressing arachnoid cells both in human and mouse.
ZIC1, ZIC2, and ZIC5 are novel molecular markers for meningiomas whereas ZIC4 expression is highly selective for medulloblastomas. The pattern of ZIC expression in both of these tumor types may reflect the properties of the tissues from which the tumors are derived.
Meningiomas are primary central nervous system tumors derived from arachnoidal (meningothelial) cells [reviewed in [1–3]]. Meningiomas are the most common type of benign intracranial brain tumor, with an annual incidence of approximately 2.3 to 6 per 100,000 persons . They are classified into three World Health Organization (WHO) grades: benign (Grade I), atypical (Grade II), and malignant (Grade III)  based on the degree of anaplasia, number of mitoses, and presence of necrosis [1, 3]. Grade I, II and III meningiomas account for approximately 80%, 5% to 20%, and 1% to 2% of all meningiomas, respectively . Further classification is possible based on histopathological types; for example, Grade I tumors include meningiothelial, fibrous, transitional (mixed), psammomatous, angiomatous, and secretory meningiomas . Clinicopathological examination of meningioma has benefited from the discovery of several molecular markers such as vimentin and epithelial membrane antigen (EMA) . The presence of these molecules in meningiomas may partly reflect the cellular properties of arachnoidal cap cells from which meningiomas are thought to derive.
Recently, Inoue et al.  revealed that developing meningeal cells in mouse produce Zic family zinc finger proteins. Zic proteins are known to play critical roles in animal development [reviewed in ]. In humans, mutations in ZIC genes are associated with congenital anomalies such as holoprosencephaly (medial forebrain dysgenesis), heterotaxy (left-right axis disturbance), and Dandy-Walker malformation (cerebellar dysgenesis) [reviewed in [5, 6]]. In meningeal cell development, mouse Zic proteins are present in the primitive meninx (meningeal cell precursors), and a deficiency of Zic2 or Zic1/Zic3 results in impaired proliferation and differentiation of meningeal precursors . These findings led us to hypothesize that ZIC proteins are present in meningiomas.
Previous studies have reported immunoreactivities to ZIC proteins in medulloblastoma, another type of brain tumor. ZIC1 is predominantly expressed in medulloblastoma [7, 8], and ZIC2 expression is down-regulated in medulloblastoma compared to its mRNA level in normal cerebellum . In normal tissue, ZIC1 is preferentially expressed in cerebellar granule neurons and their progenitors . The abundance of the ZIC1 protein in medulloblastoma is considered to reflect the properties of the cell of origin, cerebellar granule neuron. However, the expression of the other ZIC genes (ZIC2, ZIC3, ZIC4, and ZIC5) has not been investigated in medulloblastoma or other brain tumors. A comparison of the expression of ZIC family members is required. But current studies are limited by the structural similarities and the cross-reactivity of the antibodies .
These observations led us to investigate the expression profiles of ZIC genes in brain tumors, focusing on meningiomas. We first examined the levels of ZIC1-5 mRNAs in various brain tumors. Then ZIC-like immunoreactivity was examined in meningiomas and other brain tumors. Our results indicate that the expression of ZIC1, ZIC2 and ZIC5 is a conserved feature of meningioma.
Human materials and animals
All sampling procedures for human brain tumors were performed according to the Ethics Guidelines for Human Genome/Gene Analysis Research published by the Japanese Ministry of Education, Culture, Sports, Science and Technology http://www.lifescience.mext.go.jp/files/pdf/40_213.pdf. C57BL/6J mice were purchased from Nihon SLC (Shizuoka, Japan). Animal experiments were approved by the Animal Experiment Committee of the RIKEN Brain Science Institute.
RNA isolation and RT-PCR analysis
RNA samples were prepared from human brain tumors using TRIZOL (Invitrogen, Carlsbad, CA, USA). Total RNA from human whole brain, liver, lung, kidney, and kidney tumor was purchased from Clontech (Mountain View, CA, USA). All RNA samples were treated with RNase-free DNase I (Promega, Madison, WI, USA) before the reverse transcription reaction. Reverse transcription and PCR were performed using ThermoScript reverse transcriptase (Invitrogen) and TaKaRa Ex Taq Hot Start Version (TAKARA BIO, Shiga Japan), respectively. The accession numbers, primer sequences, the size of PCR products, and number of PCR cycles were as follows: ZIC1 (NM_003412), 5'-GGCCCGGAGCAGAGTAAT-3' and 5'-AGCCCTCAAACTCGCACTT-3' (229 bp, 26 cycles); ZIC2 (NM_007129), 5'-CCCTTCAAGGCCAAATACAA-3' and 5'-TGCATGTGCTTCTTCCTGTC-3' (218 bp, 26 cycles); ZIC3 (NM_003413), 5'-GCAAGTCTTTCAAGGCGAAG-3' and 5'-CATGCATGTGCTTCTTACGG-3' (225 bp, 28 cycles); ZIC4 (NM_032153), 5'-GCCCTTCAAAGCCAAATACA-3' and 5'-GCCCTCGAACTCGCATC-3' (172 bp, 28 cycles); ZIC5 (NM_033132), 5'-TCTGCTTCTGGGAGGACTGT-3' and 5'-GGGAATGTTTCTTCCGATCA-3' (252 bp, 28 cycles); and ACTB (NM_001101), 5'-CAACCGCGAGAAGATGACC-3' and 5'-TCCAGGGCGACATAGCACA-3' (324 bp, 22 cycles). Each PCR cycle consisted of 1 min at 94°C, 30 s at 62.5°C, and 1 min at 72°C. RT-PCR product was harvested during the log-linear phase of the amplification curve at the PCR cycles indicated above. The PCR products were separated by electrophoresis in a 2.0% agarose gel and quantified by densitometry of ethidium bromide-stained bands using ImageJ version 1.33 http://rsb.info.nih.gov/ij/. The transcript amounts were normalized to ACTB. The results are indicated as relative values to the transcript amount in whole brain total RNA. The means of the three independent RT-PCR experiments are indicated. The absence of amplified products in RNA samples that had not been reverse transcribed was confirmed at the cycles indicated (data not shown).
Analysis of public microarray data
The meningioma microarray results of Keller et al.  were obtained from the Gene Expression Omnibus (GEO) repository ftp://ftp.ncbi.nih.gov/pub/geo/DATA/supplementary/series/GSE12530/GSE12530_RAW.tar; the data included 24 meningioma and two dura mater controls profiled on GE Healthcare/Amersham Biosciences CodeLink Human Whole Genome Bioarrays. Following probes (ID_REF) were subjected for the analysis: ZIC1, 228079; ZIC2, 56068; ZIC3, 387012; ZIC4, 456009; ZIC5, 80028. For each gene, expression level in meningioma was indicated relative to the expression level in normal dura mater.
Protein immunoblots and immunohistochemistry
Human ZIC1-5 expression plasmid vectors were constructed in pcDNA3.1 vector (Invitrogen) that had been modified to contain three tandem hemagglutinin (HA) epitope tag-encoding sequences. The open reading frame sequences were obtained by PCR amplification of a human brain cDNA (Clontech) and a BAC clone (RP11-1148D4, BACPAC Resources Center, CHORI, Oakland, CA, USA) and subsequent reconstruction of the sequence-verified fragments.
The expression vectors were transfected into NIH3T3 cells or 293T cells using TransIT-LT1 transfection reagent (Mirus, Madison, WI, USA). For protein immunoblot, the transfected cells were lysed in sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer. The proteins were separated by SDS-PAGE and transferred to polyvinylidene di?uoride membranes. The membranes were first incubated with 5% skim milk in phosphate-buffered saline (PBS) containing 0.1% Tween20 (PBST) for 1 h to block non-specific binding, and then incubated with either rat monoclonal anti-HA tag antibody (3F10, 1:4000), rabbit polyclonal anti-Zic2 antibody [CXY2, ], mouse monoclonal anti-Zic1 antibody [ZC26, ], or rabbit polyclonal anti-ZIC4 antibody (ProteinTech Group, Chicago, IL, USA) diluted in the blocking buffer, at 4°C overnight. Membranes were then washed in PBST for 1 h, and incubated for 1 h with horseradish peroxidase-conjugated anti-rabbit, anti-rat, or anti-mouse antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). After re-washing with PBST, the bound antibodies were detected using enhanced chemiluminescence Western blot detection reagent (GE Healthcare, Buckinghamshire, UK).
Immunofluorescence staining was performed as described previously [4, 13]. Immunohistochemical staining of human brain tumor specimen was performed as described  and briefly described below. Formaldehyde-fixed, paraffin embedded sections (3 to 6 μm) were prepared using a standard procedure . The antigen retrieval was performed by autoclaving the dewaxed, hydrated tissue specimens in 10 mM sodium citrate (pH 6.5) at 121°C for 5 min. The sections were further incubated in 0.3% hydrogen peroxide for 10 min to inactivate the endogenous peroxidase-like activity. Blocking of non-specific binding was performed by immersing the specimens in a PBST containing 1% skim milk and 2% normal goat serum for 30 min at room temperature. The primary antibodies used in the immunohistochemical staining were anti-ZIC (CXY2, ZC26, and anti-ZIC4) anti-vimentin (Nichirei Biosciences, Tokyo, Japan), anti-EMA (Dako, Glostrup, Denmark), and anti-chondroitin sulfate proteoglycan (CS56, Sigma-Aldrich, St. Louis, MO, USA).
The sections were incubated at 4°C overnight in blocking buffer containing appropriately diluted primary antibodies (CXY2, 1:2000; ZC26, 1:500; anti-ZIC4, 1:400; anti-vimentin, 1:300; anti-EMA, 1:4). The bound primary antibodies were detected by immunoperoxidase reaction using VECTASTAIN Elite ABC kit (Vector Laboratories, Burlingame, CA, USA) and 3,3'-Diaminobenzidine (DAB) as a substrate. The double labeling was carried out by sequential primary antibody incubation and detection using two different chromogens, DAB and DAB-Nickel. Digital images were obtained using the NanoZoomer Digital Pathology C9600 (Hamamatsu Photonics, Shizuoka, Japan) image scanner and analyzed using the NDPViewer software (Hamamatsu Photonics).
Expression of ZIC1-5in various types of brain tumors
Clinicopathological information on the tumors examined in the RT-PCR analysis
Right frontal lobe
Right frontal lobe
Left frontal lobe
Bilateral frontal and parietal lobes
Left frontal lobe
Right parietal lobe
Right temporal lobe
Left cerebellopontine angle
Left frontal lobe (intraventricular)
Fuhrman grade II Renal cell carcinoma
Expression of ZIC1-5in meningioma with different histological grades
Immunoscreening of ZIC proteins in brain tumor specimens
ZIC-like immunoreactivities in brain tumors
Grade I Meningioma
Grade II Meningioma**
Grade III Meningioma**
Primitive neuroectodermal tumor
Atypical teratoid rhabdoid tumor
We performed immunohistochemical staining of other types of brain and intracranial tumors as references (Table 2). In glioblastoma multiforme, no ZIC-like signals were detected, with the exception of one of the four cases that showed a weak signal for ZC26. Two out of four hemangioblastomas stained weakly with CXY2. A primitive neuroectodermal tumor and an atypical teratoid rhabdoid tumor were CXY2-immunopositive.
ZIC proteins can be detected in normal arachnoid cells
ZIC expression in meningioma and normal meningeal cells
The distribution of ZIC-like immunopositive signals in meningiomas may reflect the property of the arachnoid cells from which meningiomas arise. Since Zic proteins play an essential role in the proliferation of meningeal cell progenitors , we consider it possible that ZIC proteins are involved in the proliferation of meningioma cells. This is consistent with our observation that ZIC1, ZIC2 and ZIC5 mRNA levels are higher in meningiomas than in normal brain tissues. We suggest that studies to elucidate the involvement of ZIC genes in the meningioma tumor cell proliferation would be beneficial.
Antibodies against EMA, vimentin and Ki-67 have been used to provide histopathological differential diagnoses of meningioma and to estimate its malignancy grade ; however, no molecular marker that is specific for meningiomas has been described to date. Although additional studies in more cases of various brain tumors are needed, detection of ZIC proteins in meningioma will be helpful for diagnoses in extraaxial brain tumors.
In terms of meningeal expression of ZIC genes, a previous in situ hybridization study  indicates that mouse Zic1 and Zic2 are strongly expressed in the embryonic meningeal cells and its precursors. Our study confirms these results and clarifies that the expression of Zic genes continues during development in the arachnoid cell lineage in mouse. We detected immunoreactive Zic/ZIC proteins in meningeal cells in both mouse and human adult brains. Although the physiological role of ZIC proteins in the adult meningeal cells has not been clarified, ZIC proteins might be involved in the maintenance of cell properties of differentiated arachnoid cells, analogous to their role in meningeal cell differentiation in mouse development.
Differential expression of human ZIC genes in tumors
A major finding of this study is the differential expression of members of the human ZIC gene family in various brain tumor types. RT-PCR analysis revealed that ZIC4 expression is highly enhanced in medulloblastoma, in a sharp contrast to the expression levels in whole brain, while ZIC1, ZIC2 and ZIC5 are expressed both in the meningioma and medulloblastoma. The results of the immunoscreening are consistent with the RT-PCR results in that ZIC4-like nuclear staining was almost limited to the medulloblastoma cases. Although we detected low amounts of ZIC4 transcript in one of the three-meningioma cases, the immunoreactive ZIC4 protein level in this sample was below the limit of detection (data not shown). The distinctive expression profile of ZIC4 highlights its usefulness as a molecular marker for brain tumor pathology.
Interestingly, in a cohort of patients with small cell lung carcinoma, one of the most aggressive tumors known, autoantibodies to ZIC2 are present in 28% of patients, and the presence of these autoantibodies is associated with less aggressive clinical parameters such as a better response to initial therapy . Autoantibodies in small lung cell carcinoma patients show cross-reactivity among ZIC1, ZIC2, and ZIC4 proteins , suggesting that immunoreactivities of the autoantibodies are directed primarily against the conserved zinc finger domains of ZIC . Furthermore, ZIC autoantibodies have been detected in patients with both small cell lung carcinoma and paraneoplastic neurological syndrome, although it is not clear whether the immune response to ZIC proteins is pathogenically related to the development of the neurological syndrome . In small cell lung carcinoma, it is important that the expression of ZIC proteins, and the usability of ZIC autoantibodies as a diagnostic tool, are further clarified. Therefore, comprehensive studies that examine each of the five members of the ZIC family proteins, and their corresponding autoantibodies are required.
Our findings, together with those of previous studies [8–10], demonstrate the importance of ZIC proteins in clinical oncology. Future studies will require improved ZIC protein- or anti-ZIC antibody-detection systems. We suggest that the anti-ZIC antibodies and the full-length ZIC1-5 expression vectors described in this study will be useful for this purpose.
The main results can be summarized as follows:
1) ZIC1, ZIC2, and ZIC5 mRNA levels are high in meningiomas.
2) All ZIC genes are abundantly expressed in medulloblastomas.
3) ZIC1 mRNA levels are higher in Grade II than in Grade I meningiomas.
4) ZIC1/2/3/5-like immunopositive signals are detected in most meningiomas irrespective of their histological type.
5) ZIC1/2/3/5-like immunopositive signals are detected in the arachnoid cell lineage.
6) ZIC1/2/3/5 and ZIC4-like immunopositive signals are detected in both human cerebellar granule neurons and medulloblastomas.
epithelial membrane antigen
Zinc finger protein of the cerebellum.
We thank Ms. Chihiro Saruta (Laboratory for Molecular Neuogenesis, BSI) for advice on the use of the digital scanner, and the Support Unit for Bio-material Analysis, RIKEN BSI Research Resources Center (Mr. Taizen Morishita) for the monoclonal antibody purification. This study was funded by RIKEN BSI and supported by a Grant-in-Aid for Scientific Research (16390086) to JA.
- Perry A, Louis DN, Scheithauser BW, Banerjee R, Margraf L: Meningiomas. World Health Organization Classification of tumours Pathology and genetics of tumours of the nervous system. Edited by: Louis DN, Ohgaki H, Wiestler OD. 2007, Lyon: IARC Press, 164-172. 4Google Scholar
- Riemenschneider MJ, Perry A, Reifenberger G: Histological classification and molecular genetics of meningiomas. Lancet Neurol. 2006, 5: 1045-1054. 10.1016/S1474-4422(06)70625-1.View ArticlePubMedGoogle Scholar
- Commins DL, Atkinson RD, Burnett ME: Review of meningioma histopathology. Neurosurg Focus. 2007, 23: E3-10.3171/FOC-07/10/E3.View ArticlePubMedGoogle Scholar
- Inoue T, Ogawa M, Mikoshiba K, Aruga J: Zic deficiency in the cortical marginal zone and meninges results in cortical lamination defects resembling those in type II lissencephaly. J Neurosci. 2008, 28: 4712-4725. 10.1523/JNEUROSCI.5735-07.2008.View ArticlePubMedGoogle Scholar
- Aruga J: The role of Zic genes in neural development. Mol Cell Neurosci. 2004, 26: 205-221. 10.1016/j.mcn.2004.01.004.View ArticlePubMedGoogle Scholar
- Grinberg I, Millen KJ: The ZIC gene family in development and disease. Clin Genet. 2005, 67: 290-296. 10.1111/j.1399-0004.2005.00418.x.View ArticlePubMedGoogle Scholar
- Michiels EM, Oussoren E, Van Groenigen M, Pauws E, Bossuyt PM, Voute PA, Baas F: Genes differentially expressed in medulloblastoma and fetal brain. Physiol Genomics. 1999, 1: 83-91.PubMedGoogle Scholar
- Yokota N, Aruga J, Takai S, Yamada K, Hamazaki M, Iwase T, Sugimura H, Mikoshiba K: Predominant expression of human ZIC in cerebellar granule cell lineage and medulloblastoma. Cancer Res. 1996, 56: 377-383.PubMedGoogle Scholar
- Pfister S, Schlaeger C, Mendrzyk F, Wittmann A, Benner A, Kulozik A, Scheurlen W, Radlwimmer B, Lichter P: Array-based profiling of reference-independent methylation status (aPRIMES) identifies frequent promoter methylation and consecutive downregulation of ZIC2 in pediatric medulloblastoma. Nucleic Acids Res. 2007, 35: e51-10.1093/nar/gkm094.View ArticlePubMedPubMed CentralGoogle Scholar
- Bataller L, Wade DF, Fuller GN, Rosenfeld MR, Dalmau J: Cerebellar degeneration and autoimmunity to zinc-finger proteins of the cerebellum. Neurology. 2002, 59: 1985-1987.View ArticlePubMedGoogle Scholar
- Keller A, Ludwig N, Backes C, Romeike BF, Comtesse N, Henn W, Steudel WI, Mawrin C, Lenhof HP, Meese E: Genome wide expression profiling identifies specific deregulated pathways in meningioma. Int J Cancer. 2009, 124: 346-351. 10.1002/ijc.23942.View ArticlePubMedGoogle Scholar
- Inoue T, Ota M, Mikoshiba K, Aruga J: Zic2 and Zic3 synergistically control neurulation and segmentation of paraxial mesoderm in mouse embryo. Dev Biol. 2007, 306: 669-684. 10.1016/j.ydbio.2007.04.003.View ArticlePubMedGoogle Scholar
- Aruga J, Inoue T, Hoshino J, Mikoshiba K: Zic2 controls cerebellar development in cooperation with Zic1. J Neurosci. 2002, 22: 218-225.PubMedGoogle Scholar
- Vural B, Chen LC, Saip P, Chen YT, Ustuner Z, Gonen M, Simpson AJ, Old LJ, Ozbek U, Gure AO: Frequency of SOX Group B (SOX1, 2, 3) and ZIC2 antibodies in Turkish patients with small cell lung carcinoma and their correlation with clinical parameters. Cancer. 2005, 103: 2575-2583. 10.1002/cncr.21088.View ArticlePubMedGoogle Scholar
- Sabater L, Bataller L, Suarez-Calvet M, Saiz A, Dalmau J, Graus F: ZIC antibodies in paraneoplastic cerebellar degeneration and small cell lung cancer. J Neuroimmunol. 2008, 201-202: 163-165. 10.1016/j.jneuroim.2008.01.018.View ArticlePubMedPubMed CentralGoogle Scholar
- Graus F, Saiz A, Dalmau J: Antibodies and neuronal autoimmune disorders of the CNS. J Neurol. 2009.Google Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2407/10/79/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.