Fig 1.
Up-regulation of ALK expression in GBMs.
(A) Staining by hematoxylin and eosin (HE) and IHC for ALK in astrocytomas. Note the diffuse distribution of strong cytoplasmic ALK (5A4)-positive cells in grade IV (GBM) tumor, in contrast to the sporadic distribution or absence in grade II and III astrocytomas. Insets show the magnified views of the boxed area. Original magnification, x100 and x400 (inset). (B) IHC score for ALK (5A4) in astrocytomas. G, grade (C) Relationship of ALK (5A4) expression with overall survival and progression-free survival for all grades of astrocytomas. n, number of cases. (D) Relationship of ALK (5A4) expression and isocitrate dehydrogenase 1 (IDH1) gene status with overall survival and progression-free survival for all grades of astrocytomas. n, number of cases. wt, wild type; mt, mutant type.
Fig 2.
Up-regulation of ALK expression in hypervascular areas of GBMs.
(A) Left: staining by hematoxylin and eosin (HE) and IHC for ALK in GBMs. Note the strong cytoplasmic ALK (5A4) positivity in cells around vascular components (a) (indicated by arrows) of GBMs, in contrast to the sporadic staining or absence in tumor cells adjacent to necrotic foci (b) (partitioned by dotted line). Panels (a) and (b) are magnified views of the boxed areas. Original magnification, x100 (left) and x200 (middle and right). Right: IHC score for ALK (5A4) in perivascular (Peri-v) and perinecrotic (Peri-n) areas. (B) Staining by HE (upper and lower left) and double-staining for ALK/CD34 (upper and lower middle) and SMA/CD34 (upper and lower right) in mature vessels (upper) and microvessels (lower) in GBM. Strong ALK (5A4) (brown) positivity is observed in GBM cells surrounding CD34 (red)-positive mature vessels (upper middle), while focal ALK (5A4) immunoreactivity is also evident in CD34-positive microvessels (lower middle). Note the close association between SMA- and CD34-positive cells in both tumor vasculatures (upper and lower left). Insets show magnified views of the boxed areas. Original magnification, x200 and x400 (inset). (C) Left: staining by HE and IHC for ALK and CD34 in GBMs. Note the strong ALK (5A4) immunopositivity in areas with high CD34 immunoreactivity, in contrast to the sporadic staining or absence in the low immunoreactivity lesions. Original magnification, x100. Right: relationship between ALK (5A4) immunointensity and microvascular density as determined by CD34 immunoreactivity in GBMs. The data shown are means±SDs. (D) Western blot analysis of the indicated proteins after CoCl2 treatment with the different doses shown for 24 hours (left) and 50 μM CoCl2 for the time shown (right) in KINGS-1 cells.
Fig 3.
ALK mRNA expression and gene arrangement status in GBMs.
(A) Staining by hematoxylin and eosin (HE) and ISH for ALK mRNA. Note the abundant mRNA signals in tumor cells around vascular components (indicated by arrows), in contrast to the weak signals adjacent to necrotic foci (partitioned by dotted line). Insets show magnified views of the boxed areas. Original magnification, x100 and x400 (inset). (B) Relationship of ALK (5A4) expression between the ISH signal positivity and the IHC score in GBMs. (C) FISH analysis of four GMB cases with high ALK expression. The interphase nuclei of these cases indicate absence of ALK rearrangement, as shown by the merged red and green signals (indicated by arrows).
Fig 4.
Association between ALK expression and neovascular features in GBMs.
(A) Staining by hematoxylin and eosin (HE) and IHC for ALK in vascular co-option. Note the strong ALK (5A4) immunoreactivity in vascular co-option with classic perivascular cuffs (indicated by arrows). Insets show magnified view of the boxed area. Original magnification, x200 and x400 (inset). (B) Staining by HE and IHC for ALK and CD34 in vascular mimicry. Note the diffuse ALK (5A4) immunoreaction and focal CD34 immunoreactivity in the perfused vascular networks containing red blood cells. The lower right panel shows the magnified view of the boxed area in the lower left panel. Original magnification, x100 and x400 (lower right panel). (C) Upper: staining by HE and IHC for ALK. Note the strong ALK (5A4) immunoreactivity in tumor vasculature (indicated by boxes and magnified in the inset), as well as tumor cells around vascular components. Original magnification, x200 and x400 (inset). Middle and lower: staining by HE and ISH for ALK mRNA in GBMs. Note the positive ALK mRNA signals in tumor vasculature which are indicated by boxes in middle right panel and magnified in lower right panels (positive ALK mRNA signal in tumor vasculature are indicated by arrows), as well as tumor cells around vascular components. Original magnification, x200 and x400 (lower left panel).
Table 1.
Relationship between ALK expression and neovascularization in GBMs.
Fig 5.
ALK/Stat3/HIF-1α axis in astrocytoma cells.
Western blot (left) and RT-PCR (right) analyses for the indicated molecules in (A) KS-ALK#4 cells and (B) KINGS-shALK#37 and #46 cells. (C) KS-1 cells were transfected with HIF-1α (left) and VEGF-A (right) reporter constructs, together with either ALK, Stat3C, or HIF-1α. Relative activity was determined based on arbitrary light units of luciferase activity normalized to pRL-TK activity. The activities of the reporter plus the effector relative to that of the reporter plus empty vector are shown as means±SDs. The experiment was performed in duplicate. v, empty vector. Western blot (left) and RT-PCR (right) analyses for the indicated molecules in (D) KS-ALK#4 cells and (E) KINGS-shALK#37 and #46 cells after CoCl2 treatment for 4 hours. (F) Left: staining by hematoxylin and eosin (HE) and IHC for HIF-1α in GBMs. Note the strong HIF-1α immunoreactivity (indicated by arrows) in both perivascular areas and pseudopalisading around necrotic lesion (partitioned by dotted line). Original magnification, x100. Right: IHC score for HIF-1α in perivascular (Peri-v) and perinecrotic (Peri-n) areas of GBMs.
Fig 6.
Association between ALK expression and cell proliferation in astrocytoma cells.
(A) Left: KS-ALK#4 cells and mock cells were seeded at low density. The cell numbers are presented as means±SDs. P0, P2, P4, P6, and P8 indicate 0, 2, 4, 6, and 8 days after cell passage, respectively. Right: Cell Counting Kit-8 (CCK-8) assay for cell proliferation. Cells were seeded at 1x103 cells in 96-well plates. Viable cell numbers were quantitated. Relative absorbance values (P5 or P7 relative to P2) are presented as means±SDs. P2, P5, and P7 indicate 2, 5, and 7 days after cell passage, respectively. This experiment was performed in triplicate using independent samples. (B) Western blot analysis for the indicated proteins in ALK#4 and mock cells after serum stimulation for the times shown. (C) Left: KINGS-shALK#37, #46 cells, and mock cells were seeded at low density. The cell numbers are presented as means±SDs. P0, P3, P6, and P9 indicate 0, 3, 6, and 9 days after cell passage, respectively. Right: CCK-8 assay for cell proliferation to quantitate viable cell numbers as mentioned above. (E) Staining by hematoxylin and eosin (HE) and IHC for pStat3, pAkt, and Ki-67 in GBMs. Note the strong immunoreactivity for these molecules in perivascular lesions (vessels are indicated by arrows), in contrast to the weak immunoreaction in perinecrotic areas (necrotic lesion is partitioned by dotted line). Original magnification, x100. (F) IHC scores for pStat3 and pAkt and Ki-67 labeling indices in perivascular (Pv) and perinecrotic (Pn) lesions.
Table 2.
Correlations among IHC markers investigated in GBM cases.
Fig 7.
Relationship of ALK expression with Sox4 and N-myc in astrocytomas.
(A) Various ALK promoter constructs used in this study. (B) KS-1 cells were transfected with ALK promoter constructs, together with the indicated Sox genes. Relative activity was determined based on arbitrary light units of luciferase activity normalized to pRL-TK activity. The activities of the reporter plus the effector relative to that of the reporter plus empty vector are shown as means±SDs. The experiment was performed in duplicate. (C) KS-1 cells were transfected with ALK promoter constructs, together with either Sox4, c-myc, or N-myc. (D) Various promoter constructs were used for evaluating transcriptional regulation of the ALK promoter by either Sox4 or N-myc. (E) The shortest ALK promoter constructs containing mutations in two putative E-boxes (E1 and E2), along with either Sox4 or N-myc, were transfected into KS-1 cells. (F) The Sox4 (left) and the N-myc (right) promoter constructs, along with either c-myc, N-myc, or Sox4, were transfected into KS-1 cells. (G) Left: staining by hematoxylin and eosin (HE) and IHC for N-myc and c-myc in GBMs. Note the strong immunoreactivity for N-myc, but not c-myc, in tumor cells around vessel (indicated by arrows) but not perinecrotic area (necrotic lesion is partitioned by dotted line). Right: IHC scores for N-myc and c-myc in perivascular (Peri-v) and perinecrotic (Peri-n) lesions.
Fig 8.
Schematic representation of ALK signal networks in modulation of neovascularization and cell proliferation in perivascular GBM cells.