Expression of hypoxia-inducible carbonic anhydrases in brain tumors

Abstract

Malignant brain tumors exhibit distinct metabolic characteristics. Despite high levels of lactate, the intracellular pH of brain tumors is more alkaline than normal brain. Additionally, with increasing malignancy, brain tumors display intratumoral hypoxia. Carbonic anhydrase (CA) IX and XII are transmembrane isoenzymes that are induced by tissue hypoxia. They participate in regulation of pH homeostasis by catalyzing the reversible hydration of carbon dioxide. The aim of our study was to investigate whether brain tumors of different histology and grade of malignancy express elevated levels of CA IX and XII as compared to normal brain. We analyzed 120 tissue specimens from brain tumors (primary and metastatic) and normal brain for CA IX and XII expression by immunohistochemistry, Western blot, and in situ hybridization. Whereas normal brain tissue showed minimal levels of CA IX and XII expression, expression in tumors was found to be upregulated with increased level of malignancy. Hemangioblastomas, from patients with von Hippel-Lindau disease, also displayed high levels of CA IX and XII expression. Comparison of CA IX and XII staining with HIF-1alpha staining revealed a similar microanatomical distribution, indicating hypoxia as a major, but not the only, induction factor. The extent of CA IX and XII staining correlated with cell proliferation, as indicated by Ki67 labeling. The results demonstrate that CA IX and XII are upregulated in intrinsic and metastatic brain tumors as compared to normal brain tissue. This may contribute to the management of tumor-specific acid load and provide a therapeutic target.

Fig. 1

Immunohistochemical detection of CA IX and XII in normal brain and gliomas. Sections of normal brain and gliomas of increasing grades of malignancy were analyzed by immunohistochemistry using antibodies against CA IX (A–D) or CA XII (E–H). Staining of normal brain sections (A and E) did not show any CA IX and XII expression. In the low-grade gliomas (B and F), only weak and occasional staining was observed. Anaplastic astrocytomas (C and G) showed distinct patches of increased staining. The strongest staining was observed in the glioblastoma multiforme samples (D and H) and was most evident in perinecrotic areas. Bar = 100 μm.

Fig. 2

Patterns of CA IX expression in glioblastomas. A. Increased CA IX expression (arrows) around areas of micronecrosis (N) was consistently observed in glioblastoma specimens. B. In some glioblastoma specimens, a less intense but more widespread staining, in which almost all of the cells stained positively for CA IX, was detected throughout the entire section. Note the vessels (V) in the middle of the section, which suggests that hypoxic induction is unlikely in this case. The capillary endothelial cells (arrows) are not stained, indicating a tumor-cell-specific expression of CA IX. Bar = 100 μm.

Fig. 3

Immunohistochemical detection of CA IX and XII in brain metastases and hemangioblastomas. Specimens from a brain metastasis (non-small-cell carcinoma of the lung, panels A and B) and a capillary hemangioblastoma from a patient with VHL disease (panels C and D), stained positively for both CA IX (A and C) and CA XII (B and D). Note the perinecrotic pattern observed in the metastatic tumor (A). In the VHL-associated hemangioblastoma, positive staining is observed consistently throughout the sections, even in the absence of necrosis. Bar = 100 μm.

Fig. 4

Semiquantitative analysis of CA staining and tumor type. Immunohistochemical staining (IHS) scores of the CA IX (upper panel) and CA XII (lower panel) sections were ranked according to the semiquantitative rating scale described in Materials and Methods. Histologic classification is indicated on the x-axis (Gr. = glioma grade). The graph depicts the median values of the scoring data, and the bars indicate the 75th percentile of the data distribution. Statistical analysis (as described in Materials and Methods) revealed that CA IX and XII expression differed significantly from normal brain in all but the grade I gliomas. *P < 0.05 in comparison to normal brain.

Fig. 5

Analysis of CA IX and CA XII staining and cell proliferation rate. Ki67-positive cells were counted in a high-power (200×) vision field and are expressed as percentage of the total number of cells per vision field. All histologies were included in this analysis. The Ki67 data were calculated for each CA IX (upper) or XII (lower) staining score group (x-axis). The line within the box depicts the median, and the upper boundary of the box indicates the 75th percentile, the lower boundary the 25th percentile. Whiskers above and below show the 90th and 10th percentiles, respectively. Using a rank-based correlation analysis as described in Materials and Methods, we determined that both CA IX and CA XII correlate with the Ki67 labeling index (CA IX vs. Ki67: correlation coefficient = 0.751, P < 0.01; CA XII vs. Ki67: correlation coefficient = 0.532, P < 0.01).

Fig. 6

Coincidence of CA IX, CA XII, and HIF-1α in a glioblastoma specimen. Patterns of immunohistochemical staining of CA IX (A), CA XII (B), and HIF-1α (C) in adjacent sections were compared. The HIF-1α is weaker and more nuclear compared to CA IX and XII. The distribution shows a partial overlay among the three molecules. The region of greatest coincidence is indicated by the arrows. Bar = 100 μm.

Fig. 7

Western blot analysis of CA IX and XII in specimens of increasing malignancy. Protein extracts (40 μg) of representative samples from normal brain (lanes 1 and 2), low-grade astrocytoma (lane 3), anaplastic astrocytoma (lane 4), and glioblastoma multiforme (lane 5) were separated and transferred to a nitrocellulose membrane. Western blotting for CA IX and CA XII was performed as described in the Materials and Methods section on five samples of each tissue class (normal brain, low-grade astrocytoma, anaplastic astrocytoma, and glioblastoma). Representative results show increased expression for both CA IX and XII in gliomas with higher grade of malignancy, although the extent of CA IX upregulation appears stronger than that of CA XII. Note that the 54/58-kd double band in the CA IX results and the 43/44-kd doublet in the CA XII results occur only in the samples with higher expression (lanes 4 and 5). Relative protein loading is shown by reprobing with an antibody to β-actin.

Fig. 8

Analysis of CA IX and CA XII mRNA expression by in situ hybridization. ³⁵S-radiolabeled riboprobes complementary (antisense) to CA IX (A and B) and CA XII (C and D) were hybridized to normal brain (A and C) and glioblastoma multiforme (B and D) sections. After emulsion autoradiography and counterstaining in cresyl violet, bright field photographs of representative sections were taken. A significant upregulation of CA IX and XII mRNA in the glioblastoma samples was observed. Representative results from eight normal brains and 20 glioblastoma specimens are shown. In accordance with the Western blot results, the extent of CA IX induction appeared to be significantly higher than that of CA XII. Bar = 100 μm.