Ultrastructural Characterization of Primary Cilia in Pathologically Characterized Human Glioblastoma Multiforme (GBM) Tumors

Joanna J Moser; Marvin J Fritzler; Jerome B Rattner


BMC Clin Pathol. 2014;14(40) 

In This Article


Primary cilia are non-motile sensory cytoplasmic organelles that have been implicated in signal transduction, cell to cell communication, left and right pattern embryonic development, sensation of fluid flow, regulation of calcium levels, mechanosensation, growth factor signaling and cell cycle progression.[1,2] They are present in the central nervous system and depletion of primary cilia in pro-opiomelanocortin hypothalamic neurons have induced hyperphagia.[3,4] Central nervous system primary cilia are key organelles required for Sonic hedgehog signalling (Shh)[5–8] where components Patched, Smoothened, Suppressor of fused and Gli transcription factors concentrate in the primary cilium.[9–11] It is currently thought that an intact primary cilium is required to enable proper Shh pathway function.[12] Subventricular zone astrocytes extend their primary cilium into the cerebral spinal fluid (CSF) suggesting they play a role in sensing ion concentration, pH, osmolarity, and changes in protein or glucose levels.[13] It is possible that astrocyte primary cilia can sense concentrations of neurotransmitters, growth factors, hormones, osmolarity, ions, pH and fluid flow in the extracellular space and relay homeostatic information (or lack thereof) to the centrosome.

Defects in the formation and/or function of primary cilia underlie a variety of human diseases that impact neurological development and are broadly referred to as ciliopathies and include diseases such as Alström, Bardet-Biedl, Joubert, Meckel-Gruber and oral-facial-digital type 1 syndromes. Common neurological phenotypes include obesity, ataxia and mental retardation.[14] The expression and function of primary cilia has become a focus of attention in a number of normal and malignant cells and tissues but have not been characterized in human glioblastoma tissue samples. Given that primary cilia are linked to cell cycle regulation and progression, several studies have suggested that primary cilia may play a role in tumor formation.[15,16]

Previously our group undertook a comparative investigation of primary cilia in cultured primary human astrocytes and compared them to those found in five human astrocytoma/glioblastoma cell lines.[17] We demonstrated that the primary cilium region in cultured astrocyte cells is structurally complex, with ciliogenesis progressing through five distinct stages (Figure 1), and included foci for endocytosis-based signalling.[17]

Figure 1.

Primary ciliogenesis progresses through five morphologically distinct stages in human astrocytes. Key characteristics of each stage are indicated with arrows. Paired lateral vesicles are prominent at the distal end of the basal body in Stage 1. Distal appendages are triangular in appearance and reside at the distal end of the basal body (Stage 2). The paired lateral vesicles fuse to become a vesicular hat and become stretched by the outgrowth of the primary cilium and can be seen progressing through stages 2 through 4. Stage 5 shows a mature primary cilium with a surrounding cilium-pit. Used with permission from Moser et al. BMC Cancer 2009, 9:448, Figure 2A © BioMed Central.

Further, we documented that in each of the five astrocytoma/glioblastoma cell lines studied (U-87 MG, T98G, U-251 MG, U-373 MG, U-138 MG), fully formed primary cilia are either expressed at a very low level, are completely absent or do not proceed through all the stages of ciliogenesis.[17] In addition, we noted several defects in the structure of astrocytoma/glioblastoma centrioles, including abnormal length and appendage architecture, that were not observed in primary human astrocytes.[17] We concluded that aberrant ciliogenesis is common in cells derived from astrocytomas/glioblastomas and that this deficiency likely contributes to the phenotype of these malignant cells. These initial studies in astrocytoma/glioblastoma cell lines indicate that defects in primary cilium ciliogenesis do occur in glioblastoma cells and provided the impetus for this current study which characterizes the morphology of primary cilia and documents ciliogenesis defects in molecularly characterized human glioblastoma multiforme (GBM) tumors. Glioblastomas, although relatively uncommon with an annual incidence rate of 3–4 cases per 100,000 people, have disproportionately high morbidity and mortality rates with median survival pegged at 12–15 months.[18,19] Primary glioblastomas typically occur in patients older than 50 years of age and are characterized by epidermal growth factor receptor (EGFR) amplification and mutations, loss of heterozygosity of chromosome 10q and other abnormalities as reviewed in Wen and Kesari (2008).[20] One of the most common defects in growth factor signalling involves EGFR[21] and amplification occurs almost exclusively in glioblastomas with 40-50% of patients containing EGFR amplification.[20] Isocitrate dehydrogenase (IDH) mutations are a strong predictor of a more favourable prognosis and a highly selective molecular marker for secondary glioblastomas.[22] Mutations of genes encoding IDH1 and IDH2, as compared to no mutations, are associated with younger age and a better prognosis in adults with gliomas.[23] O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation silences the MGMT gene, decreasing DNA repair activity and increases the susceptibility of tumor cells to chemotherapeutic agents.[20] Recently it was shown that MGMT promoter methylation was associated with better overall survival in patients with GBM regardless of therapeutic intervention.[24] Given this burden of disease, it is important to determine the degree to which ciliogenesis is compromised in glioblastoma tumors as this information will inform the identity of altered mechanisms which may become targets for the development of future treatments.

Our hypothesis was that primary cilia in human GBM cells would be completely absent or show defects in the early stages of ciliogenesis. Our primary objective was to examine primary cilia expression and structure in human GBM tissue samples at both the light and ultrastructural level.