Tissue mechanics promote IDH1-dependent HIF1α-tenascin C feedback to regulate glioblastoma aggression.
Miroshnikova YA, Mouw JK, Barnes JM, Pickup MW, Lakins JN, Kim Y, Lobo K, Persson AI, Reis GF, McKnight TR, Holland EC, Phillips JJ, Weaver VM. · 2016
Read Tissue mechanics promote IDH1-dependent HIF1α-tenascin C feedback to regulate glioblastoma aggression. by Miroshnikova YA, Mouw JK, Barnes JM, Pickup MW, Lakins JN, Kim Y, Lobo K, Persson AI, Reis GF, McKnight TR, Holland EC, Phillips JJ, Weaver VM. (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5361403.txt.
Introduction
Yekaterina A. Miroshnikova and co-authors linked tissue stiffness to brain tumor force in Nature Cell Biology in 2016. Patients whose glioma tumors carry a flaw in the IDH1 gene tend to live longer. Such tumors grow in a stiff, dense web that starves them of oxygen and wakes a stress switch, HIF1α. The team asked how stiffness, the switch, and the gene flaw play off each other.
Force of tumors and patient outlook tracked both the switch level and a stiff web rich in a protein called tenascin C. In mouse grafts, the flawed gene curbed tumor force by damping the switch and its web output. Web stiffness then fed back to lift the switch again, a loop that could bypass the gene's guard. In return cases, a small RNA that should hold the switch down had lost its grip. Stiffness alone, they argue, can drive this cancer.
Catalog record
| CREATOR | Miroshnikova YA, Mouw JK, Barnes JM, Pickup MW, Lakins JN, Kim Y, Lobo K, Persson AI, Reis GF, McKnight TR, Holland EC, Phillips JJ, Weaver VM. |
|---|---|
| YEAR | 2016 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Tissue mechanics promote IDH1-dependent HIF1α–tenascin C feedback to regulate glioblastoma aggression By Yekaterina A. Miroshnikova, Janna K. Mouw, J. Matthew Barnes, Michael W. Pickup, Johnathan N. Lakins, Youngmi Kim, Khadjia Lobo, Anders I. Persson, Gerald F. Reis, Tracy R. McKnight, Eric C. Holland, Joanna J. Phillips Published in: Nature cell biology DOI: 10.1038/ncb3429 ABSTRACT Increased overall survival for patients with glioma brain tumours is associated with mutations in the metabolic regulator isocitrate dehydrogenase 1 (IDH1). Gliomas develop within a mechanically challenged microenvironment that is characterized by a dense extracellular matrix (ECM) that compromises vascular integrity to induce hypoxia and activate HIF1α. We found that glioma aggression and patient prognosis correlate with HIF1α levels and the stiffness of a tenascin C (TNC)-enriched ECM. Gain- and loss-of-function xenograft manipulations demonstrated that a mutant IDH1 restricts glioma aggression by reducing HIF1α-dependent TNC expression to decrease ECM stiffness and mechanosignalling. Recurrent IDH1-mutant patient gliomas had a stiffer TNC-enriched ECM that our studies attributed to reduced miR-203 suppression of HIF1α and TNC mediated via a tension-dependent positive feedback loop. Thus, our work suggests that elevated ECM stiffness can independently foster glioblastoma aggression and contribute to glioblastoma recurrence via bypassing the protective activity of IDH1 mutational status. Malignant gliomas are a highly heterogeneous group of brain tumours that exhibit diverse invasiveness, aggressiveness and treatment responsiveness. Diffuse lower-grade gliomas (LGGs, World Health Organization (WHO) grades II and III) demonstrate highly variable clinical behaviour ranging from indolent to rapidly progressive1. LGGs often recur as glioblastomas (GBMs, WHO grade IV glioma), which are inarguably the most advanced and lethal adult brain tumour, with poor treatment responsiveness and a high recurrence rate contributing to poor patient outcome. Mutations in the metabolic enzyme isocitrate dehydrogenase (IDH) characterize the majority of LGGs and a small percentage of GBMs, and define subtypes that associate with better responses to radiation treatment and an improved prognosis compared with gliomas with wild-type (WT) IDH1–4. Oncogenic IDH mutations divert metabolic flux resulting in high levels of (R)-2-hydroxyglutarate, an onco-metabolite implicated in hypoxia-inducible factor 1 alpha (HIF1α) stability, epigenetic modifications and extracellular matrix (ECM) remodelling5. The relationship between IDH mutations and downstream HIF1α signalling remains unclear with conflicting reports suggesting elevated HIF1α protein in IDH-mutant tumours6 or, conversely, blunted hypoxia sensing through HIFα mediated by elevated activity of prolyl 4-hydroxylases7. As tumours are characterized by altered tissue-level and cellular mechanics, including ECM remodelling and stiffening, and a stiffened ECM can compromise blood vessel integrity to induce hypoxia and activate HIF1α (refs 8–10), we investigated the interplay between IDH1 mutation status, ECM mechanics, HIF1α signalling and GBM aggression. RESULTS ECM stiffness associates with IDH1 mutations in primary tumours To determine whether glioma aggression associates with ECM stiffness, we utilized atomic force microscopy (AFM) to quantify the stiffness11 of the ECM in a cohort of fresh-frozen human brain biopsies representing non-tumour gliosis, de novo (primary) LGGs (WHO grades II and III) and de novo primary GBMs (WHO grade IV). We determined that gliotic tissue had the lowest ECM stiffness (Young's modulus, E;10–180 Pa) while LGGs (50–1,400 Pa) and GBMs (70–13,500 Pa) were progressively stiffer (Fig. 1a). Consistent with the elevated ECM stiffness, mechanosignalling increased progressively from gliotic tissue to the stiffer GBMs, as indicated by higher phosphorylated focal adhesion kinase (pFAK residue Tyr397) and myosin light chain 2 (pMLC2 residue Ser19) (Fig. 1b). NanoString nCounter gene expression analysis12 of an outcome-predictive gene cluster13 (aggressiveness binned on a scale of 1–10 with 1 indicating the most aggressive) of human GBM tissue biopsies indicated a significant correlation between the proportion of highly stiff areas within a GBM tissue (E >1,400 Pa) and worst patient prognosis score (Fig. 1c, red). By contrast, those tissues that contained a high proportion of soft ECM regions (E < 200 Pa) had the best patient prognosis score13 (Fig. 1c, blue). Interestingly, despite a strong overall correlation between ECM stiffness and glioma grade, relatively stiff and compliant subsets of patient tumours were evident in both the LGG and GBM samples. To explore this finding, we assessed the IDH1 mutational status in our cohort of patient samples. Mutations in IDH genes characterize the majority of LGGs and a small percentage of GBMs1–4. For both the LGG and GBM samples, we noted that the stiffness of the ECM in the R132H IDH1 tumours was significantly less than the ECM measured in the WT IDH1 tumours (Fig. 1d,e). To this end, the R132H IDH1 GBM samples exhibited mechanical characteristics more concordant with the LGG samples (Fig.1d,e). R132H IDH1 GBM samples also had lower pFAK and pMLC2 levels, reflecting reduced mechanosignalling (Fig. 1f). These data demonstrate that ECM stiffness correlates with glioma aggression and that IDH1 mutational status associates with a softer ECM, regardless of histological grade. TNC modifies ECM stiffness and mechanosignalling To examine the relationship between IDH mutation and ECM stiffness, we used a candidate approach to profile major ECM constituents in GBM, hyaluronic acid (HA), and HA-interacting molecules14,15. Immunohistochemistry revealed that the elevated ECM stiffness in the patients with poorer prognosis was accompanied by a substantial increase in HA expression as well as increased levels of TNC, and indicated that these levels increased progressively from gliosis to LGGs and to the GBMs (Fig. 2a and Supplementary Fig. 2a,b)16. Importantly, ECM stiffness did not correlate with levels or distribution of type I collagen, vasculature or cellularity (Supplementary Fig. 1a–g). Indeed, a survey of the ECM status of the softer ECMs in the IDH1-mutant human GBM biopsies revealed a reduction in the levels of TNC (Fig. 2b). Further, bioinformatics analysis of publicly available messenger RNA expression data1 revealed a significant correlation between IDH1 mutational status and TNC expression in LGG tumours, with stratification of LGG patients by TNC upregulation revealing reduced patient survival (Fig. 2c,d). Similarly, IDH1 mutational status correlated significantly with TNC mRNA expression in GBM17,18 tumours (Fig. 2e). NanoString nCounter gene expression analysis12 of TNC expression (where lower scores correspond to lower expression) in GBM samples yielded a significant correlation between TNC and the tumour aggressiveness score derived via the outcome-predictive gene expression signature (where lower scores correlated with reduced overall survival, Fig. 2f). Consistent with a relationship between IDH1 mutational status, ECM stiffness and glioma aggression, ectopic expression of the R132H IDH1 mutation significantly improved the survival of nude mice bearing xenografts derived from either primary GBM cells or the U87 cell line (Fig. 2g and Supplementary Fig. 2c,d). The resultant tumours were also softer and exhibited reduced mechanosignalling, as revealed by decreased pMLC2 and pFAK (Fig. 2h and Supplementary Fig. 2e,f). In both the primary and U87 xenograft models, we noted that the softer IDH1-mutant-expressing GBM tumours had low to negligible TNC expression (Fig. 2i and Supplementary Fig. 2g). To experimentally test the role of TNC in ECM stiffness and glioma aggression, we knocked down TNC in primary GBM cells (WT IDH1, Supplementary Fig. 2h,i) and assessed the resulting effects on ECM stiffness and tumour aggression. Nude mice injected orthotopically with GBM cells with short hairpin RNA (shRNA)-reduced TNC not only survived longer (Fig. 2j), but the tumours were softer and had lower mechanosignalling, as revealed by reduced pFAK and pMLC2 (Fig. 2k,l). These findings demonstrate that TNC can impact glioma aggression by increasing ECM stiffness and