Université Victor Segalen Bordeaux 2, CNRS UMR 5084, 146 Rue Léo Saignat, 33076 Bordeaux-Cedex, France.
Analyst. 2010 Dec;135(12):3052-9. doi: 10.1039/c0an00513d. Epub 2010 Oct 7.
Malignant gliomas are very aggressive tumors, highly angiogenic and invading heterogeneously the surrounding brain parenchyma, making their resection very difficult. To overcome the limits of current diagnostic imaging techniques used for gliomas, we proposed using FTIR imaging, with a spatial resolution from 6 to 10 μm, to provide molecular information for their histological examination, based on discrimination between normal and tumor vasculature. Differentiation between normal and tumor blood vessel spectra by hierarchical cluster analysis was performed on tissue sections obtained from xenografted brain tumors of Rag-gamma mice 28 days after intracranial implantation of glioma cells, as well as for human brain tumors obtained in clinics. Classical pathological examination and immunohistochemistry were performed in parallel to the FTIR spectral imaging of brain tissues. First on the animal model, classification of FTIR spectra of blood vessels could be performed using spectral intervals based on fatty acyl (3050-2800 cm(-1)) and carbohydrate (1180-950 cm(-1)) absorptions, with the formation of two clusters corresponding to healthy and tumor parts of the tissue sections. Further data treatments on these two spectral intervals provided interpretable information about the molecular contents involved in the differentiation between normal and tumor blood vessels, the latter presenting a higher level of fatty acyl chain unsaturation and an unexpected loss of absorption from osidic residues. This classification method was further successfully tested on human glioma tissue sections. These findings demonstrate that FTIR imaging could highlight discriminant molecular markers to distinguish between normal and tumor vasculature, and help to delimitate areas of corresponding tissue.
恶性脑胶质瘤是非常侵袭性的肿瘤，高度血管生成，并异质浸润周围的脑实质，使其切除非常困难。为了克服当前用于脑胶质瘤的诊断成像技术的局限性，我们提出使用 FTIR 成像，其空间分辨率为 6 至 10 μm，根据正常和肿瘤血管的区分，为其组织学检查提供分子信息。通过层次聚类分析对从 Rag-gamma 小鼠颅内植入胶质瘤细胞 28 天后的异种移植脑肿瘤获得的组织切片以及从临床获得的人类脑肿瘤进行了正常和肿瘤血管光谱的区分。对脑组织的 FTIR 光谱成像进行了经典的病理检查和免疫组织化学检查。首先在动物模型上，使用基于脂肪酸酰基（3050-2800 cm(-1)) 和碳水化合物（1180-950 cm(-1)) 吸收的光谱间隔，可以对血管的 FTIR 光谱进行分类，形成两个对应于组织切片的健康和肿瘤部分的聚类。对这两个光谱间隔的进一步数据处理提供了关于区分正常和肿瘤血管的分子含量的可解释信息，后者表现出更高水平的脂肪酸链不饱和性和意想不到的 osidic 残基吸收损失。这种分类方法在人类脑胶质瘤组织切片上也得到了成功的验证。这些发现表明，FTIR 成像可以突出区分正常和肿瘤血管的鉴别分子标记，并有助于划定相应组织的区域。
