Aberrant activation of Hedgehog/Gli1 pathway on angiogenesis in gliomas
Correspondence Address: Source of Support: This study was supported by the Youth
Foundation of Tongji University (No. 2010KJ008),, Conflict of Interest: None DOI: 10.4103/0028-3886.105192
Source of Support: This study was supported by the Youth Foundation of Tongji University (No. 2010KJ008),, Conflict of Interest: None
Background: Hedgehog/Gli1 (HH/Gli1) pathway plays an important role in the patterning and development of the central nervous system during embryogenesis. Recent data have shown its potential involvement in a subset of human gliomas and inhibition of the pathway resulted in tumor suppression in both in vitro and in vivo studies. The underlying mechanisms of tumor suppression, however, remain to be fully elucidated. Materials and Methods: Gli1 expression was investigated in 60 surgically resected glioma tissues (World Health Organization (WHO) III-IV). Results: Gli1 was expressed in 43 gliomas with high Gli1 expression in nine cases, moderate expression in 21 cases, and low expression in 13 cases. Additionally, microvessel counts were higher in Gli1 positive gliomas than those in Gli1 negative gliomas. Gli1 expression in gliomas was positively correlated with microvessel density (MVD). To explore the molecular mechanisms of the phenotypic changes, we performed quantitative real-time polymerase chain reaction (PCR) and Western blot analysis to monitor the changes of a series of genes, which play critical roles in the regulation of glioma angiogenesis. In conclusion, HH/Gli1 pathway inhibition resulted in down-regulation of vascular endothelial growth factor (VEGF), matrix metalloproteinase 2 (MMP2), and matrix metalloproteinase 9 (MMP9) expressions, whereas this pathway activation led to up-regulation of VEGF, MMP2, and MMP9 expressions. These molecular changes of the HH/Gli1 pathway inhibited by indirect drug approach were consistent with Gli1 RNA-interference (RNAi) in glioma cell lines. Conclusion: Our findings demonstrated that the aberrantly active HH/Gli1 pathway contributed to angiogenesis in part through induction of VEGF, MMP2, and MMP9.
Keywords: Angiogenesis, gliomas, Hedgehog/Gli1 pathway, microvessel counts
As an important determinant in normal embryonic development, Hedgehog/Gli1 (HH/Gli1) pathway plays a crucial role in tissue patterning, cell differentiation, and cell proliferation.  In mammals, there are three Hedgehog ligand proteins: Sonic Hedgehog (SHH), desert Hedgehog (DHH), and Indian Hedgehog (IHH). In the absence of ligands, the pathway is inactive. The transmembrane protein receptor Patched (Ptch1) inhibits Smoothened (Smo) from the activation of down-stream signaling components and Gli1-mediated transcription of target genes. Activation of HH/Gli1 pathway is initiated through three ligands bound to Ptch1, which alleviates the Ptch1-mediated suppression of Smo. The translocation of the active form of the zinc finger transcriptional factor Gli1 to the nucleus activates transcription and expression of target genes such as Gli1 and Ptch1.  Apart from its crucial roles in embryogenesis, HH/Gli1 pathway has attracted considerable interest in the field of cancer research. Recent publications have implicated aberrant pathway activation in the growth and maintenance of common malignancies such as basal cell carcinoma, lung, esophageal, pancreatic, prostate, and biliary cancers, as well as gliomas. Specific targeting of HH/Gli1 pathway may therefore offer a highly effective therapeutic strategy for treating a variety of lethal tumors. , However, these studies mainly focus on the functions of HH/Gli1 pathway in regulating transcription and expression of the genes controlling the cell cycle, cell apoptosis, and multidrug-resistance. ,, Effects of this pathway on angiogenesis-related gene regulation have also been illustrated. ,,
Intense angiogenesis is one of the basic biological features of gliomas.  Studies have shown that occurrence, development, and even relapse of gliomas are closely correlated with angiogenesis. Therefore angiogenesis is well recognized as an essential mechanism for glioma growth.  Angiogenesis is a complex multistep process, usually involving in breakdown of the vascular membrane and extracellular matrix (ECM), proliferation, and migration of endothelial cells. Over secreting vascular endothelial growth factor (VEGF), matrix metalloproteinase 2 (MMP2), and matrix metalloproteinase 9 (MMP9) play an important role during this process. ,,,, In addition, low oxygen levels increase messenger RNA (mRNA) transcription of VEGF, MMP2, and MMP9 in glioma cells by the increase of the stability of hypoxia-inducible factor-1 alpha (HIF-1 α). ,,, However, the mechanisms involving in the process are not well known.
In this study, Gli1 and CD31 expressions was first investigated in gliomas. Additionally, the association between HH/Gli1 pathway and microvessel density (MVD) in gliomas was studied. Finally, we explored the effect of activation and inhibition of HH/Gli1 pathway on regulation of angiogenesis-related gene expressions in glioma cell lines by using RNA-interference (RNAi) and indirect drug treatment. Our results indicated that HH/Gli1 pathway was abnormally activated in gliomas and active levels of the pathway were positive related to tumor MVD. Moreover, HH/Gli1 pathway activated could promote VEGF, MMP2, and MMP9 expression, and meanwhile the pathway inhibited conversely could contribute to expressions of these angiogenesis-related genes suppressed.
Polyclonal antibodies to GAPDH (FL-335), Gli1 (H-300), Ptch1 (H-267), CD31 (H-300), VEGF (EE02), MMP2 (8B4), and MMP9 (H-129) were obtained from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Recombineant Human Sonic Hedgehog (SHH, 100-45) and Cyclopamine were purchased from Peprotech (Rocky Hill, NJ, USA) and Biomol (Plymouth Meeting, PA, USA), respectively. SHH was dissolved in PBS, with the exception of Cyclopamine, which was dissolved in 100% DMSO (D2650; Sigma, St. Louis, MO, USA) and stored at −20°C until use. All agents were diluted with an appropriate buffer before each experiment.
Patient selection and immunohistochemistry
Glioma specimens were collected from 60 patients and all patients submitted written informed consent prior to surgery. Every specimen was histopathologically confirmed either grade III or IV gliomas (World Health Organization the central nervous system (WHO CNS) histopathological system). The study group comprised of 34 males and 26 females. The mean (S.E.M.) age was 53.2 ± 1.5 years (range: 32-66 years). Surgical specimens were fixed in 10% formalin, embedded in paraffin. Each glioma was immunohistochemically examined for Gli1 and CD31 staining. Specimens were cryosectioned serially to 4 μm thickness. Single-antibody detection was performed as previously described.  Heat-mediated antigen retrieval was carried out in a steamer at 90°C in 0.01 M sodium citrate (pH 6.0) for 30 min followed by 3% H 2 O 2 in methanol solution for 20 min blocking at room temperature. Primary rabbit polyclonal anti-CD31 (1:200) and anti-Gli1 (1:200) were applied at 4°C overnight. Secondary goat anti-rabbit immunoglobulin (Invitrogen, NY, USA) was incubated for 2 h at room temperature. Sections as negative controls were processed with tris-buffered saline rather than primary antibody. Immunocomplexes were visualized by brown pigmentation via a standard 3, 30-diaminobenzidine (DAB) protocol.
For Gli1 counting, a total of 100 cancer cells were counted in each section. Only tumor cells with nuclear Gli1 immunoreactivities were considered for scoring purposes. The number of tumor cells showing nuclear staining of Gli1 in relation to the total number of tumor cells was expressed as a percentage (normalized to the Gli1 nuclear staining percentage). Gliomas with positive or negative Gli1 staining were classified. The MVD was calculated by the method described previously.  Briefly, each slide was examined for the most vascular area and three pictures were randomly taken at ×200 magnification. Any CD31 positive cells separated from other stained cells and not thought to be contiguous or branching from other vessels were counted. The results from all three pictures for each slide were averaged for the resulting MVD (in 0.7 mm 2 ).
Cell lines culture and immunofluorescence staining
Four human glioma cell lines (A172, SHG44, U251, and U87) were used. Cells were maintained in DMEM/F12 (Gibco, Invitrogen, NY, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco), 100 U/ml penicillin and 100 μg/ml streptomycin (Gibco).
A total of 1 × 10 5 cells of A172 and U251 were pretreated with 200 ng/ml SHH or PBS for 48 h, and then were incubated on cover glasses coated by polylysine at 37°C overnight. Cover glasses were fixed in 4% paraformaldehyde for 30 min at room temperature, and permeated in PBS with 5% horse serum (Gibco), 3% BSA (Sigma) and 0.3% Triton X-100 (Sigma) for 10 min. The samples were incubated with primary polyclonal rabbit anti-Gli1 antibody (dilution 1:200) at 4°C overnight followed by a fluorescein isothiocyanate (FITC)-labeled secondary antibody for 1 h at room temperature. Nuclei were stained with DAPI. Finally, the slides were mounted and examined by confocal microscopy.
RNA interference and cell transfection
The sequences for the Gli1 short-hairpin (shRNA) target is CTCCACAGGCATACAGGAT, and the scramble sequence used is TTCAGACCCGTCAACAAA. Annealed shRNA was cloned into the lentiviral vector directly. We designed shRNA as: 5′-CGCGTCCCCCTCCACA GGCATACAGGATTTCAAGAGAATCCTGTATG CCTGTGGAGTTTTTGGAAAT-3′. Single-stranded small-hairpin DNA samples were synthesized and annealed. Double-stranded DNA was subcloned into the MluI/ClaI site of PLVTHM Lentiviral vector; a lentiviral vector containing scrambled shRNA was used as a control. Replication-defective lentiviral particles pseudotyped with VSV-G envelope were produced by 3-plasmid transient transfection of 293T cells as previously described  with 20 μg of one of the gene transfer constructs (Lenti-shGLi1, Lentil-EGFP), 10 μg of psPAX2, and 5 μg of pMDG with a calcium phosphate transfection kit (Gibco-BRL). The transfection medium was replaced with fresh culture medium after 12 h. Viral supernatants were harvested 65 h after transfection and filtered through 0.45 mm filters (Nalgene, Rochester, NY, USA). Supernatants were either used immediately for transductions or stored in frozen aliquots at −80°C. Titers of viral supernatants were determined by infected HeLa cells with serial dilutions of viral supernatants 48 h after infection, followed by fluorescent activated cell sorter (FACS) analysis of EGFP-positive cells.
Target cells from the U87 and SHG44 cell lines were transferred with Lentivirus, involving the shGli1 and control shRNA separately. A total of 1 × 10 6 cells were plated in 25 cm 2 culture flasks overnight and LV-shGli1 and LV-controls were added to the cell culture medium for 48 h incubation (MOI = 10). Transduction efficiency was detected by analysis of FACS and Western blot.
RNA extraction and reverse transcription polymerase chain reaction
Total RNA was extracted from 1 × 10 6 cells. Reverse transcription was performed with the First Strand cDNA Synthesis Kit-ReverTra Ace-a-(FSK-100; Toyobo, Japan), according to the manufacturer's protocol. For cDNA synthesis, approximately 1 μg total RNA was reverse-transcribed into cDNA using the Oligo dT primer. Amplification conditions comprised an initial denaturation for 3 min at 94°C followed by 30 cycles of 94°C for 40 s, 55°C for 1 min, and 72°C for 1 min and 72°C for 5 min. Primer information was as follows: Gli1 (5′-TTCCTACCAGAGTCCCAAGT-3′ and 5′-CCCTATGTGAAGCCCTATTT-3′), Ptch1 (5′-CCACGACAAAGCCGACTACAT-3′ and 5′-GCTGCAGATGGTCCTTACTTTTTC-3′), Smo (5′-CCTTTGGCTTTGTGCTCATTACCTT-3′ and 5′-CGTCACTCTGCCCAGTCAACCT-3′), VEGF (5′-CCCACTGAGGAGTCCAACAT-3′ and 5′-CATTTACACGTCTGCGGATCT-3′), MMP2 (5′-CCCACTGAGGAGTCCAACAT-3′ and 5′-CATTTACACGTCTGCGGATCT-3′), MMP9 (5′-TCCCTGGAGACCTGAGAACC-3′ and 5′-GGCAAGTCTTCCGAGTAGTTT-3′), and β-actin (5′-CTGACTGACTACCTCATGAAGATC-3′ and 5′-GTGAGAAGGTCGGAAGGAAGG-3′). RT-PCR products were separated on ethidium bromide stained 1.6% agarose gels. Semi-quantitative analysis was done with Furi Gel Image Analysis System (FR-2000; Shanghai FURI Science and Technology Co. Ltd., Shanghai, China).
Quantitative real-time polymerase chain reaction
Quantitative real-time PCR was applied to evaluate the mRNA expression change of Gli1, Ptch1, Smo, VEGF, MMP2, and MMP9 in U87, SHG44, U251, and A172 cells treated with either SHH, Cyclopamine, or transduction of LV-shGli1. Concentration of these reagents was used as described in the "Results". A 20 μl reaction including 2 μl of complementary DNA template, 2 μl of forward and reverse primer, 6 μl DEPC H 2 O, and 10 μl SYBR Green Mix (QPK-201; Toyobo) was conducted by ABI Prism 7500 Sequence Detection System (Applied Biosystems, NY, USA). PCRs of each template were performed in duplicate in a 96-well plate. The thermal cycling conditions included an initial denaturation step at 95°C 10 min and 50 cycles at 95°C for 15 s and at 60°C for 1 min. The relative fold change 2−DDCt method was used to determine the relative quantitative gene expression compared with β-actin. The transcription level of target genes observed in calibrating samples was treated as the basal level and given the value 1.0.
Total cell lysates from the cell lines were prepared in sodium dodecyl sulfate lysis buffer containing proteinase inhibitors. Protein concentrations were determined using a protein assay kit (Bio-Rad, Hercules, CA, USA) according to the manufacturer's instructions. Quantified protein lysates were separated in 7.5% or 12.5% sodium dodecyl sulfate-polyacrylamide gel and electrophoretically transferred to a polyvinylidene difluoride membrane (Amersham Biosciences, Piscataway, NJ, USA). The membranes were blocked with 5% non-fat milk. The primary antibodies and dilutions were polyclonal rabbit anti-GAPDH (1:200), anti-Gli1 IgG (1:2000), anti-Ptch1 IgG (1:200), anti-VEGF IgG (1:1000), anti-MMP2 IgG (1:1000), and anti-MMP9 (1:1000). After washing, membranes were incubated with anti-rabbit antibody (Santa Cruz Biotechnology Inc.) and then detected by the enhanced chemiluminescence plus system (Amersham Biosciences).
Data were expressed as means ± S.E.M., and the differences between groups were evaluated using the Student's t-test. Correlation between continuous variables was tested using the Spearman correlation coefficients (r). All data were analyzed by using Statistical Product and Service Solutions (SPSS) 16.0 (SPSS Inc, Chicago, IL, USA). P < 0.05 was considered statistically significant.
Active degree of the HH/Gli1 pathway is positively correlated with MVD of glioma patients
To analyze the relationship of the HH/Gli1 pathway and angiognenesis in gliomas, we first tested for the expression of HH/Gli1 pathway components and CD31 in glioma resections. As nuclear staining of Gli1 was reported to be a primary marker of the HH/Gli1 pathway activity, , Sixty specimens of gliomas were classified into two groups according to positive or negative Gli1 expression. As noted in [Table 1], 43 of 60 specimens (71.7%) had Gli1 expression, and 17 (28.3%) were negative for Gli1 expression. Secondly, angiogenesis differences were analyzed in those two groups. CD31 is a sensitive indicator for vascular endothelial cells in histological tissue sections. The degree of tumor angiogenesis was generally evaluated by both calculation of positive CD31 staining cells per unit area and MVD counting under the microscope. Results showed that MVD in Gli1 positive group (n0 = 43) was significantly higher than in Gli1 negative group (n = 17; P < 0.01). MVD in Gli1 positive gliomas was 27.4 ± 4.6 and 20.9 ± 3.5 in negative samples [Table 1].
To extend these finding, we used the percentage of nuclear staining of Gli1 as the criterion of degree of HH/Gli1 pathway activated to divide 43 glioma samples with Gli1 positive staining into three groups [Table 2]: Low (1% to <30% tumor cells stained), moderate (≥30% to <60% tumor cell stained), and high expression (≥60% tumor cell stained). Analysis of MVD among three groups indicated that the increase of MVD was significantly correlated to up-regulation of active degree of HH/Gli1 pathway ( r = 0.756, P < 0.01) [Table 2]. That suggests a positive association between active degree of HH/Gli1 pathway and MVD in gliomas [Figure 1].
Status of HH/Gli1 pathway and expression of VEGF, MMP2 and MMP9 mRNA in glioma cell lines
To investigate constitutively active HH/Gli1 pathway and the role of Gli1 in regulating glioma angiogenesis-related genes in glioma cell lines, glioma cell lines with varying levels of HH pathway activity were chosen [Figure 2]a. Among four lines, Gli1 was highly expressed in U87 and SHG44 cell lines. Gli1 expression was modest in U251 and relative low in A172. Besides, mRNA expression of angiogenesis-related genes was determined by RT-PCR. The result demonstrated that U87, SHG-44, U251, and A172 had comparable levels of VEGF, MMP2, and MMP9 to internal control GAPDH [Figure 2]b.
Up-regulation of HH/Gli1 pathway activation promotes expressions of VEGF, MMP2, and MMP9
As the down-stream mediator of HH/Gli1 pathway, Gli1 status could regulate the pathway activity. A172 and U251 with relatively low baseline levels were chosen for activating tests. We used exogenous SHH, HH/Gli1 pathway activator drug targeting Ptch1, to up-regulate this pathway activity and applied cell immunostaining and quantitative real-time PCR to analyze the SHH-induced the HH/Gli1 pathway activation qualitatively and quantitatively. Results of immunofluorescence-based confocal microscopy confirmed that Gli1 nuclear translocation markedly increased after exogenous SHH were used in A172 and U251 for 48 h [Figure 3]a. Quantitative real-time PCR measurements showed that Gli1, PTCH1, and Smo mRNA expressions of A172 and U251 increased significantly compared with controls. The value in A172 was 19.5, 3.9, and 2.3, respectively, and similar changes were also observed in U251 (data not shown) [Figure 3]b. The results indicated that HH/Gli1 pathway in A172 and U251 was activated under exogenous SHH. More importantly, we found that mRNA of VEGF, MMP2, and MMP9 increased to, 6.5 ± 0.7, 8.4 ± 0.5, and 2.8 ± 0.4 compared with PBS in A172, and to 11.6 ± 1.4, 10 ± 0.5, and 1.4 ± 0.1 compared with PBS in U251, accompanied with HH/Gli1 pathway activation. The Western blot analysis revealed a significant increase of VEGF, MMP2, and MMP9 in A172 and U251 [Figure 3]c. Accordingly we speculated that SHH induced up-regulation of VEGF, MMP2, and MMP9 expressions through HH/Gli1 pathway activation. Nevertheless, further tests need to be performed to confirm this causative relationship.
Dose dependent suppression of VEGF, MMP2,and MMP9 is induced by Cyclopamine
In this study, we used Cyclopamine, a kind of HH/Gli1 pathway inhibitory drug targeting Smo, to investigate changes of VEGF, MMP2, and MMP9 expressions after HH/Gli1 pathway blockade. Different concentrations of Cyclopamine were used. Quantitative real-time PCR analysis demonstrated the decrease of Gli1, Ptch1, and Smo mRNA levels at different concentration Cyclopamine. We found that 48 h treatment of 5 μM Cyclopamine significantly suppressed HH/Gli1 pathway activity of U87 and SHG44. Effect of 10 μM Cyclopamine suppressing HH/Gli1 pathway activity of U87 and SHG44 was more significant [Figure 4]a. Expression levels of VEGF, MMP2, and MMP9 mRNA and protein were sequentially analyzed in U87 and SHG44 with HH/Gli1 pathway inhibited by different concentrations of Cyclopamine. Quantitative real-time PCR analysis showed that 48 h treatment with 5 μM Cyclopamine markedly reduced mRNA expressions of VEGF, MMP2, and MMP9 to 95.0 ± 4.1%, 41.3 ± 2.8%, and 33.3 ± 3.3% in U87, respectively. Ten micromolar of Cyclopamine made mRNA expressions of these genes reduce to 43.4 ± 1.7%, 35.7 ± 2.9%, and 14.7 ± 2.9% more significantly. The same trends in SHG44 were observed in the experiment [Figure 4]b. These findings indicated that Cylopamine could suppress VEGF, MMP2, and MMP expression by HH Gli1 pathway in dose dependent ways. To verify these findings, we performed Western blot for VEGF, MMP2, and MMP protein in U87 and SHG44 cell lines. The Western blot results also supported this hypothesis that inhibition of HH/Gli1 pathway could down-regulate the expression of VEGF, MMP2, and MMP9.
HH/Gli1 pathway in LV-shGli1 transferred glioma cell lines suppresses mRNA and protein expression of VEGF, MMP2, and MMP9
To further verify whether HH/Gli1 pathway affects the regulation of angiogenesis-related genes mentioned above, we developed LV-shGli1 constructs to inhibit Gli1 function and used Western blot and real-time PCR to analyze HH/Gli1 pathway inhibition. SHG44 and U87 with relatively higher levels were chosen to induce the activation of HH/Gli1 pathway. The transfection efficiency in this study was estimated to be approximately 90%. The Western blot analysis showed that Gli1 expression of U87 and SHG44 cells treated after transfection with LV-shGli1 for 48 h was obviously inhibited. As key components of HH/Gli1 pathway, Gli1, Ptch1, and Smo mRNA of U87 and SHG44 decreased to 6.1 ± 1.3%, 8.6 ± 2.0%, 21.9 ± 1.8% and 8.9 ± 1.6%, 13.6 ± 2.5%, 21.9 ± 2.8 compared with controls, respectively [Figure 5]a. This implicated that the activation of HH/Gli1 pathway was reduced with inhibited Gli1 expression. Meanwhile, VEGF, MMP2, and MMP9 mRNA expressions in U87 was reduced to 7.3 ± 0.5%, 38.6 ± 2.8%, 42.3 ± 2.7% and 17.9 ± 1.3%, 20.6 ± 1.9%, 47.0 ± 3.6% in SHG44 compared with controls, respectively. The same trends were observed for the protein levels of VEGF, MMP2, and MMP9 [Figure 5]b. These findings demonstrate that HH/Gli1 pathway plays an important role in the regulation of VEGF, MMP2, and MMP9 expression.
Malignant gliomas are among the most lethal tumors with a very dismal prognosis, despite advances in standard therapy, including surgery, radiation, and chemotherapy, the median survival of patients with malignant gliomas remained unchanged in the past several years.  In an attempt to develop new therapeutic strategies and identify the molecular mechanism involved in glioma occurrence, development, and recurrence, there has been extraordinary scientific interest. Recently, we reported that there is an abnormally active HH/Gli1 pathway in a subset of brain malignant gliomas, and this pathway activation is closely related to glioma recurrence. , Our studies focused on effects of this pathway on regulating expression of MDR- and Cyclin-related genes, but more work is needed to be done to fully elucidate its intrinsic mechanisms. By focusing on the regulation of angiogenesis, this study further promoted the understanding of mechanisms of HH/Gli1 pathway in gliomas.
Angiogenesis is one of important factors of glioma occurrence, development, and recurrence. The concept of the angiogenesis driving tumor growth and malignant progression was introduced by Folkman in the early 1970s.  This concept presumed that only tumors with angiogenic activity might grow beyond size of 2 mm, once tumor volume beyond this size, tumor cells could no longer be nourished by mere diffusion. It is widely accepted that most tumors and metastases originate from small vascular structures. Glioma studies also confirmed this hypothesis. Indeed, the best characterized pro-angiogenic factor is VEGF in gliomas, except that, ECM degradation induced by MMP2 and MMP9 play a major role in angiogenesis. Many researchers confirmed that overexpression of VEGF, MMP2, and MMP9 was positively related to tumor grade and MVD, and negatively related to tumor prognosis. ,,,,,, However, the link between these genetic events and HH/Gli1 pathway activation is only partly understood. Recent attention has been focused on the role of HH/Gli1 pathway in angiogenesis. ,,,
In our study, we documented a heterogeneous expression pattern of Gli1 and CD31 in gliomas. More importantly, an association between Gli1 and MVD in gliomas was demonstrated for the first time. Despite recent studies were reported to describe Gli1 abnormal activation in angiogenesis, our studies further confirmed this vital phenomenon in gliomas. Gli1 is a key protein of HH/Gli1 pathway. Previous reports indicated that nuclear translocation of Gli1 protein correlated closely with HH/Gli1 pathway and was an essential marker in the activation of HH/Gli1 pathway.  Therefore, in the present study, we used the percentage of cells with nuclear staining of Gli1 to evaluate the degree of HH/Gli1 pathway activation. A positive association between the grade of HH/Gli1 pathway activation and tumor MVD was discovered. This suggests that HH/Gli1 pathway could promote glioma angiogenesis by some of intrinsic mechanisms.
HH/Gli1 pathway is an important signaling pathway, which controls a variety of developmental processes during normal development. In adult, the pathway remains active in stem and/or progenitor cells and participates in necessary regeneration and repair after injury.  In these processes, the activation of HH/Gli1 pathway is transient and tightly regulated. In contrast, aberrant activations of tumor HH signaling cascades is not controlled by regulatory mechanisms, and further contributes to clonogenic survival of tumors and tumor regrowth.  Several studies have demonstrated that this pathway can participate in several tumor pathological process by regulating the expressions of VEGF and MMP genes. ,, Notwithstanding the scarcity of similar studies in brain tumors, Gli1 is originally identified as a gene amplified in malignant human gliomas, which strongly suggests the possibility of the link.
To explore molecular background between HH/Gli1 pathway and angiogenesis in gliomas, we used glioma cell lines with various active degree of HH/Gli1 and classified these cell lines into two groups in light grade of HH/Gli1 pathway activated. Then we regulated HH/Gli1 pathway activity at different levels, Ptch1 by SHH or Smo by Cyclopamine and analyzed changes of VEGF, MMP2, and MMP9 expressions. Our results indicated that HH/Gli1 pathway status could be related to regulation of VEGF, MMP2, and MMP9 expressions. Additionally, down-regulation of VEGF, MMP2, and MMP9 expression was induced by Cyclopamine to present concentration-dependent characteristics. Therefore, our conclusions were obtained by indirect means. In order to prove our hypothesis, we performed RNAi at direct Gli1 level in U87 and SHG44. The data showed that Gli1 inhibition can indeed down-regulate mRNA and protein expressions of VEGF, MMP2, and MMP9. Therefore, we have reasons to speculate the abnormally active HH/Gli1 pathway contributes to angiogenesis in part through the involvement of VEGF, MMP2 and MMP9.
This study documented the heterogeneous expressions of Gli1 and CD31 protein in a fraction of gliomas. Secondly, the study revealed that the degree of the HH/Gli1 pathway activated was positively correlated with MVD of gliomas. Finally, the study demonstrated that the HH/Gli1 pathway status could regulate VEGF, MMP2, and MMP9 expressions. Findings of this study provided a molecular basis of HH/Gli1 pathway in glioma angiogenesis. The data indicated that the aberrantly active HH/Gli1 pathway contributed to angiogenesis in part through the involvement of VEGF, MMP2, and MMP9.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2]