| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Barbara Jane Levy Laboratory of Molecular Neuro-Oncology, Hermelin Brain Tumor Center, Department of Neurosurgery, Henry Ford Health Sciences Center, Detroit, Michigan 48202
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
SPARC is a developmentally regulated glycoprotein that is secreted into the ECM (2) . It functions as a counteradhesive protein, modifying cell shape through the dissociation of focal adhesions (3) , and modulates cell-matrix interactions, presumably by binding to ECM components (2) . It delays cell cycle progression in endothelial cells and fibroblasts by concentration-independent and -dependent mechanisms, respectively (4) . This delay is probably accomplished, in part, by the direct interference of SPARC with platelet-derived growth factor (5) and vascular endothelial growth factor (6) receptor interactions and by indirect effects on basic fibroblast growth factor mechanisms (7) . Therefore, SPARC influences several biological processes including differentiation, migration, and proliferation.
Early in development, SPARC is expressed in tissues undergoing remodeling (8) and during vascular morphogenesis (2) . In the adult, expression becomes more restricted but can be activated in controllable pathological situations such as wound healing (9) . However, increasing evidence in the literature suggests that SPARC expression plays a role in certain malignancies, including melanoma (10) and breast (11) , colon (12) , esophageal (13) , and prostate (14) cancers. Our studies examining SPARC expression in human astrocytomas (1) and meningiomas (15) and an in vivo skin-fold model of invasion (16) indicated that SPARC was highly expressed in invading brain tumor cells. Therefore, our results, combined with reported data, suggest that SPARC might functionally contribute to the invasive phenotype.
To assess the effects of SPARC on tumor invasion, we selected the established cell line U87MG because it has low levels of endogenous SPARC (1) , and it is not invasive when stereotactically injected into the nude rat brain. U87MG cells were transfected with the transactivator, and a selected parental clone designated as U87T2 was transfected with SPARC. Doxycycline-regulatable, SPARC-transfected clones, designated A2b2, A2bi, and C2a4, were established for further analyses. Using these clones, we demonstrated that SPARC promoted invasion in vitro using the spheroid confrontation assay (17) . Additional in vitro studies demonstrated that SPARC modulated growth, attachment, and migration of tumor cells in vitro (18) . Increased SPARC expression correlated with delayed cell growth and altered attachment to brain ECM proteins in ECM-specific and concentration-dependent manners (18) .
Although all clones acted similarly in the growth and attachment assays, the clones differed from one another with respect to migration on the ECM proteins. Assessing our results in the context of a model of adherence proposed by Greenwood and Murphy-Ullrich (3) , we hypothesized that the ability of the SPARC-transfected clones to migrate on the ECMs correlated with the amount of SPARC secreted and, consequently, the level of adherence induced (18) . Furthermore, we proposed that the effects of SPARC were complex, such that both lower and higher levels of secreted SPARC promoted a level of adherence conducive for migration, whereas intermediate levels would promote stronger adherence and less migration (18) . In addition, we hypothesized that the increased invasion would be accompanied by delayed cell growth (18) . Therefore, to test these hypotheses and determine whether SPARC promotes invasion and delays growth in vivo, we have used these same clones and stereotactically injected them into nude rat brain. Tumors were allowed to grow for either 7, 9, or 20 days before brain removal, fixation, paraffin embedding, and sectioning. Sections were then serially H&E stained for morphological assessment of invasion and to measure tumor volume, immunohistochemically stained to visualize SPARC, subjected to in situ hybridization with the human AluII DNA-binding probe to identify human cells, and immunohistochemically stained with MIB-1 to measure proliferation index.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Brain Xenograft.
Before implantation, 85–90% confluent cells were trypsinized, rinsed with DMEM + 10% FBS, and centrifuged at 1000 rpm for 10 min, and the pellet was resuspended in PBS (1x PBS; pH 7.0–7.2; Life Technologies, Inc.). Viable cells were counted using a Hausser Scientific Brightline hemocytometer, and the concentration was adjusted to 5 x 105 cells/5 µl of PBS. The cells were placed on ice. For each cell line, six nude rats were injected with 5 µl each. Injections were performed in sets of two, and the cells were freshly prepared before each set of injections. Using Institutional Animal Care and Use Committee-approved protocols, each nude rat was anesthetized and placed in a stereotactic frame (David Kopf Instruments, Tujunga, CA), and the skull was exposed. A hole was made 3 mm to the right of the bregma, and the cells were injected using a 10-µl Hamilton (#2701) syringe with a #4 point 26 s-gauge needle mounted in a stereotactic holder. The syringe was lowered to a depth of 3.5 mm and then raised to a depth of 2.5 mm. The tumor cells were injected at a rate of 0.5 µl every 10 s. On completion, a 2-min waiting period was observed before withdrawing the syringe slowly to prevent any reflux. The skull was then cleaned, the hole was sealed with bone wax, and the incision was sutured. Tumors were allowed to grow, and animals were sacrificed on day 7 (U87T2 and SPARC-transfected clones) or on day 9 (U87T2) and on day 20 (SPARC-transfected clone A2b2), when the animals developed signs of neurological deficit and had to be sacrificed. Before being sacrificed, the animals were anesthetized, and death followed cardiac puncture, perfusion with 250 ml of sterile 0.9% saline solution, and fixation with 250 ml of 10% formalin. The brains were removed and stored in 10% formalin for at least 24 h.
Sections.
Formalin-fixed rat brains were placed in a coronal rat brain matrix (Activational Systems, Inc., Warren, MI) for 200–400 g rats, and sliced into 2-mm blocks. These blocks were then processed routinely, paraffin embedded, and serially sectioned at 7 µm. Sections for H&E staining were placed onto uncoated slides. Sections intended for immunohistochemistry and in situ hybridization were placed onto positively charged Opti Plus Barrier Slides (BioGenex, San Ramon, CA). Sections from all six tumors were used as follows. Serial sections were: (a) routinely H&E stained for histomorphological assessment of invasion; (b) immunohistochemically processed for SPARC expression and localization; (c) subjected to AluII in situ hybridization to identify human tumor cells; and (d) immunohistochemically processed for MIB-1 expression to assess proliferative index. Three tumors from each set of six were completely sectioned, H&E stained to visualize the tumor, and used for tumor volume measurements.
SPARC and MIB-1 Immunohistochemistry.
The 7-µm sections were dried in a 60°C oven for 1 h and routinely deparaffinized to water. Heat-induced antigen retrieval using citrate buffer (pH 6.0) was performed using a hot plate as described previously (1
, 15)
. The sections were then processed using the BioGenex staining system. Unless indicated, all reagents used were provided by BioGenex as predilutes supplied in kit form. All steps were performed at room temperature using room temperature reagents. The slides were labeled with a BioGenex protocol barcode, placed on the stainer, and processed automatically according to the indicated protocol. Briefly, for SPARC and MIB-1 detection, sections were blocked with hydrogen peroxide for 5 min, rinsed in wash buffer, blocked with Power Blocking reagent for 5 min, and rinsed in wash buffer. Sections were then incubated for 1.5 h with anti-SPARC antibody (Hematologic Technologies, Inc.; 1.5 µl diluted in 20 ml of 0.25% BSA in PBS) or for 30 min with MIB-1 antihuman Ki-67 antibody (Dako Corp.; 1:100 dilution in 0.5% BSA in PBS). Control sections were processed omitting the primary antibody or substituting with the appropriate immunoglobulin isotype. Sections were rinsed in wash buffer, incubated in supersensitive multilink reagent for 20 min, and rewashed. The sections were then incubated in supersensitive streptavidin/horseradish peroxidase reagent for 20 min, and washed. For MIB-1, DAB substrate (16 drops of chromagen, 8 drops of peroxide in 20 ml of substrate buffer) was applied to the sections for 8 min. For SPARC detection, the slides were removed from the system and hand processed because of time constraints. DAB substrate was applied for 15 s. Sections were then washed, counterstained in Gill II hematoxylin for 8 s, blued in ammonia water, rinsed in water, dehydrated, cleared, and coverslipped.
AluII in Situ Hybridization.
The 7-µm sections were dried for 1 h in a 60°C oven. The slides were submerged in BioGenex Nucleic Acid Retrieval solution, and the loosely covered container was placed in a microwave at full power. When the solution started to boil, the container was removed, and the slides were quickly immersed in another staining container of Nucleic Acid Retrieval solution. The sections were again subjected to full power until boiling. The microwave door was opened to allow air to circulate while the oven was reset to 30% power for 15 min. Once restarted, care was taken so that the solution level was never allowed to go below the sections. After the 15 min, the container was removed from the microwave and cooled at room temperature for 30 min. The sections were then washed thoroughly in running deionized water for 5 min, and barcode labels were attached to the slides. The slides were placed on the stainer, and the sections were subjected to the BioGenex in situ hybridization protocol. The following water washes were in deionized water or BioGenex buffer wash as indicated. Briefly, the sections were denatured in solution I for 5 min, water washed, incubated with BioGenex AluII probe for 23 min, water washed, and reincubated in AluII probe for 23 min. Control sections were performed omitting the AluII probe. The sections were then washed in water and stringency washed in 2x SSC twice for 5 min each with a water rinse in between and after. Sections were then stringency washed in 0.5x SSC for 5 min, followed by three rinses in buffer wash. Protein Block was applied for 20 min, and sections were washed in buffer, blocked in Peroxide Block for 10 min, washed in buffer, blocked in avidin/biotin blocking reagents for 12 min each, and washed in buffer. Sections were then exposed to Link I (mouse antifluorescein in PBS) for 20 min, washed in buffer, exposed to Link 2 [biotinylated F(ab)2 fragments of antimouse immunoglobulins in PBS] for 8 min, washed in buffer, incubated with Label (peroxidase-labeled streptavidin in PBS) for 20 min, washed in buffer, exposed to DAB substrate (prepared as above) for 10 min, washed in buffer, rinsed in water, counterstained for 8 s in Gill II hematoxylin, blued in ammonia water, dehydrated, cleared, and coverslipped.
Tumor Volume.
Images of H&E-stained sections containing tumor were captured in Photoshop 5.0 with a SPOT model 1.3 camera (Diagnostic Instruments, Inc.) using either the 1x or 2x objective. Using the NIH Image 1.62 software, the tumor in each section was manually outlined using the freehand selection tool to measure tumor area in mm2. The area was then multiplied by the section thickness to achieve a section volume measurement. The volumes of all of the sections were added to arrive at the total volume for each tumor. Tumor volumes for three animals/tumor cell line were measured, and their volumes averaged. Results are presented as the mean ± SD.
MIB-1 Proliferation Index.
For assessment of proliferative index, the tumor core was used. Images of MIB-1 staining were captured as described above using a x20 objective. Two fields were used to cover the majority of the A2b2 and C2a4 tumors, whereas four fields were required to cover the majority of the U87T2 and A2bi tumors. Using Photoshop 5.0 software, the images were overlaid with a grid, and both total and MIB-1 positively stained cells were counted. The proliferation index was calculated as the number of MIB-1-positive cells divided by the total number of cells multiplied by 100. The proliferation indices for the six tumors/cell line were averaged. Results are presented as the mean ± SD. To assess proliferation at the invading edge, images were specifically captured at the brain-tumor interface and similarly assessed.
MIB-1 Adjusted Proliferation Index.
The in situ hybridization results indicted that the tumors contained various amounts of nonstaining cells that were either rat brain or vessel cells. Therefore, the tumor proliferation indices were corrected for these non-tumor cells. The same fields used for MIB-1 measurement were assessed on the adjacent sections subjected to in situ hybridization. The number of AluII positively stained cells were divided by the total number of cells counted to obtain the fraction of tumor cells within the tumor. The adjusted proliferation index was then calculated as the proliferation index multiplied by the fraction of tumor cells. The adjusted proliferation indices for the six tumors/cell line were averaged. Results are presented as the mean ± SD.
Statistical Analyses.
Three animals/group were used for tumor volume calculations. Six animals/group were used for the MIB-1 proliferation index and adjusted proliferation index. Student’s t tests were used to calculate the Ps. Within each experimental group, an adjustment for multiple testing was applied using the Hochberg’s method.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and adjacent sections were immunohistochemically evaluated for SPARC expression and localization (Fig. 1, I–L)
. A comparison of the immunohistochemical staining patterns (Fig. 1, I–L)
indicates that, as expected, all of the SPARC-transfected clones expressed SPARC at levels greater than the endogenous level of the parental U87T2 clone. However, differences were observed between the clones, with A2bi expressing the least, A2b2 expressing the most, and C2a4 expressing an intermediate level. These results are consistent with our previously reported Western results demonstrating that all clones secrete more SPARC than the parental U87T2; however, when comparing clones, A2bi secretes the least, A2b2 the most, and C2a4 an intermediate level of SPARC (18)
. Therefore, for clarity regarding the level of SPARC expression and secretion, the L, H, and I designations are now included with the clone names as follows A2bi (L), A2b2 (H), and C2a4 (I).
|
, the SPARC-transfected clones all produced invasive tumors (Fig. 1, B and F; C and G; D and H)
. Whereas both the A2bi (L) and the A2b2 (H) clones (Fig. 2A)
produced tumors invading into adjacent brain (Fig. 1
, B, F, J; C, G, K), as well as migrating along the corpus collosum, the clone C2a4 (I) primarily invaded as a bulk tumor along the corpus collosum (Fig. 1, D, H, L)
. Therefore, although all of the SPARC-transfected clones were invasive, distinct patterns of invasion were observed.
|
. The parental tumors neither invaded adjacent parenchyma nor traveled along the corpus collosum, consistently giving rise to well-circumscribed tumors (Fig. 2, A–C)
. Both A2bi (L) (Fig. 2, D–F)
and A2b2 (H) (Fig. 2, G–I)
tumors invaded adjacent brain, with the majority of tumor cells invading either as fingerlike projections into adjacent parenchyma or as distinct satellite tumors separate from the main tumor mass. The clone A2b2 (H) tumor cells invaded more aggressively, having more satellites and fingerlike projections that traveled farther into the brain. C2a4 (I) tumors all invaded primarily as bulk tumors, traveling along the corpus collosum (Fig. 2, J and K)
, where individual SPARC-positive tumor cells could be observed clearly (Fig. 2L)
. To ensure that the results were reproducible and not clonal artifacts, additional clones were selected and implanted, and the same patterns of invasion were observed (data not shown).
SPARC Decreases Glioma Growth in Vivo as Assessed by Tumor Volume Measurements.
The day 7 cross-sections (Fig. 1, A–D)
are through the largest area of the tumors, and a comparison suggests that SPARC decreased tumor growth. Therefore, to quantitate the SPARC-induced changes in overall tumor growth, the tumor volumes were measured (Fig. 3A)
. Although the parental U87T2 clone gave rise to large tumors by day 7 (13.4 ± 0.6 mm3), the SPARC-transfected clones all produced smaller tumors; however, the tumor volumes differed among the clones. Of the SPARC-transfected clones, tumor volume was greatest for the A2bi tumors (4.5 ± 0.6 mm3), intermediate for A2b2 (1.1 ± 0.1 mm3), and lowest for C2a4 (0.4 ± 0.1 mm3).
|
and the proliferation index (Fig. 3B)
within the tumor core was calculated for U87T2 (75.8 ± 1.0), A2bi (54.0 ± 3.0), A2b2 (20.3 ± 4.4), and C2a4 (9.3 ± 7.7). An adjusted proliferation index (Fig. 3C)
was calculated for U87T2 (95.3 ± 1.4), A2bi (73.4 ± 4.0), A2b2 (30.8 ± 6.7), and C2a4 (15.7 ± 13.0) that corrected for rat cells present within the tumor as detected by in situ hybridization (Fig. 4, B, D, F, and H)
. (The high SD for the C2a4 clone is attributable to the fact that two of the tumors had no detectable MIB-1 signal.) The results indicate that both the proliferative index and the adjusted proliferative index were reduced for all of the SPARC-transfected tumors. The level of proliferation correlated with the size of tumor in that the greater the adjusted proliferation index (Fig. 3C)
, the larger the tumor volume (Fig. 3A)
.
|
and adjusted for rat cells (Fig. 1, Q–T)
in each of these regions for individual sections. For example, the adjusted proliferation indices for the core versus the invading edge were: 80.3 versus 68.5 for A2bi (Fig. 4C
versus Fig. 1N
), 24.2 versus 15.0 for A2b2 (Fig. 4E
versus Fig. 1O
), and 23.5 versus 14.0 for C2a4 (Fig. 4G
versus Fig. 1P
), respectively. Therefore, the adjusted proliferative index at the invading edge was reduced 1.2-, to 1.6-, to 1.7-fold for A2bi, A2b2, and C2a4, respectively, in comparison with the tumor core and, on average, 1.5-fold for all of the SPARC-transfected clones.
SPARC Delays but Does Not Inhibit Growth in Vivo.
Because our in vitro data indicated that SPARC delays but does not inhibit growth (18)
, we determined whether SPARC had a similar affect in vivo. The parental U87T2 and SPARC-transfected A2b2 clones were injected, and the tumors were allowed to grow until animals exhibited neurological deficits and had to be sacrificed. Rats were sacrificed on day 9 (parental) and day 20 (A2b2). Brains were imaged before paraffin embedding and demonstrated that the tumors from the SPARC-transfected clone continued to grow larger over the 20 days, surpassing the size of the tumors derived from the parental clone (Fig. 5, A and D)
. H&E-stained cross-sections of the parental and A2b2 tumors demonstrated that the parental clone produced a well-circumscribed tumor (Fig. 5B)
with edema (Fig. 5C)
, whereas the A2b2 tumors gave rise to highly invasive tumors (Fig. 5, E and G)
. Furthermore, in addition to the fingerlike projections and distinct satellite tumors noted previously, these tumors also contained regions of frank necrosis surrounded by pseudopalisading cells, a hallmark of glioblastomas (Fig. 5, F and H)
. As was observed with day 7 tumors, SPARC expression was observed in the small invading tumor satellite cells (Fig. 5, I and J)
and in more distant invading tumor masses (Fig. 5, K and L)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study was undertaken to test our hypothesis that SPARC promotes brain tumor invasion in vivo as a result of in vitro studies that were interpreted in the context of a model of adherence proposed by Greenwood and Murphy-Ullrich (3) . We hypothesized that the ability of the SPARC-transfected clones to migrate on brain ECMs correlated with the amount of SPARC secreted and, consequently, the level of adherence induced (18) . Furthermore, we proposed that the effects of SPARC were complex, such that both lower and higher levels of secreted SPARC promoted a level of adherence conducive for migration, whereas intermediate levels would promote stronger adherence and less migration (18) . In addition, we hypothesized that the increased invasion would be accompanied by delayed cell growth (18) .
To assess the effects of SPARC in vivo, we used the well-characterized nude rat brain xenograft model that permits an adequate brain size for tumor development and growth comparisons. To determine whether SPARC promotes invasion, we correlated SPARC expression with morphological evidence of invasion. Both lower and higher levels of SPARC correlated with invasion by fingerlike projections and satellite tumor masses into the brain parenchyma, whereas the intermediate level of secreted SPARC promoted a tumor that appeared to invade more by bulk movement along the corpus collosum than into adjacent parenchyma. These differences in the ability of the clones to promote invasion in vivo are in agreement with the differences observed between clones in our reported in vitro confrontation assays, where both A2bi and A2b2 clone spheroids were better able to invade fetal brain aggregates (17) . These data are consistent with the contention that SPARC promotes glioma invasion via its ECM interactions that modulate the extent of cell adherence and, thereby, cell migration (18) .
Although we chose to investigate several clones so that we could compare effects relative to the amount of SPARC secreted, an alternative method would have been to choose the parental and one clone and examine invasion in the absence and presence of doxycycline. However, the use of doxycycline in vivo would likely confound our results because of its own ability to affect gene expression and protein activation. Of particular relevance, doxycycline not only inhibits MMP-2 activity but also reduces enzyme expression at the transcriptional level (19) . However, we suspect that MMP-2 activation may, in part, contribute to SPARC’s promotion of invasion in gliomas because we found MMP-2 expression to be up-regulated by SPARC using cDNA array analysis of the U87T2 and A2b2 clones (20) , and SPARC has been shown to induce the activation of MMP-2 in breast cancer cell lines (11) . Therefore, when dealing with the in vivo model, we thought that it was better to examine SPARC-specific effects on invasion.
The results also indicated that although SPARC promoted tumor invasion, it also decreased overall tumor growth, and that this decrease was accompanied by a decrease in cell proliferation as assessed by MIB-1 proliferation index. Because the AluII in situ hybridization results indicated that some of the cells contained within the tumor were not human, the MIB-1 proliferation index was adjusted to remove these cells. We found that increased SPARC corresponded with a statistically significant decrease in the adjusted proliferation index in all clones. However, a concentration effect was noted in that the lowest level of SPARC was associated with a higher proliferation index than those observed for the intermediate and high levels of SPARC. These data are consistent with our flow cytometry data that demonstrated that increased SPARC increased the percentage of cells in G0-G1 for C2a4 and A2b2 versus a greater percentage of cells in G2-M for A2bi (18) . Therefore, as hypothesized, we observed that the lower level of secreted SPARC correlated with more proliferation but sufficient adherence to increase infiltrative invasion, the highest level of secreted SPARC correlated with less proliferation and increased infiltrative invasion, whereas the intermediate level of secreted SPARC correlated with the least proliferation and bulk invasion.
Our in vivo data are also consistent with our in vitro data demonstrating that SPARC delays but does not inhibit tumor cell growth (18) . When the cells were implanted and allowed to grow until neurological deficit was observed, the parental U87T2 clone gave rise to well-circumscribed tumors on day 9 (similar to those observed on day 7), whereas the SPARC-transfected clone A2b2 gave rise to tumors on day 20 that were much larger than the parental clone. These tumors were not only highly invasive but also presented with regions of frank necrosis surrounded by pseudopalisading cells, features consistent with human glioblastomas. This concomitant SPARC-induced invasion and decreased proliferation is interesting relative to ongoing scientific debate as to whether cells can both divide and move at the same time. The inverse relationship we observed supports the hypothesis that cells can either "go" or "grow" but not do both (21) , and the concept that the delay in cell cycle progression provides a mechanism whereby cells might temporarily exit the cell cycle to facilitate migration (22) . These concepts were further supported by the additional decrease in proliferation associated with tumor cells at the invading edge versus in the tumor core. Tumor cells for both A2b2 and C2a4 were less proliferative at the invasive periphery than cells within the tumor core. Thus, these combined data support the concept that a decrease in proliferation is more conducive for the invasive phenotype.
The MIB-1 signal for the A2bi clone, however, was almost as high at the invading edge as in the tumor core, suggesting that these tumor cells might be both invading and dividing. Because the MIB-1 antigen Ki-67 is present throughout the cell cycle and its presence cannot distinguish between cells that have entered, are stalled, or completing the cell cycle, and because our flow cytometry data indicate that a greater percentage of cells for this clone were in the G2-M phase of the cell cycle (18) , we propose that the lower level of SPARC secreted by the A2bi clone permitted the cells to progress past G0-G1and stall later in the cell cycle. This interpretation is consistent with the tumor volume data demonstrating that SPARC decreased tumor growth for all clones, and the tumor volume for clone A2bi was more similar to that of A2b2 and C2a4. These data suggest that tumor growth for A2bi was impacted more than the MIB-1 proliferation index would suggest. MIB-1 is the standard proliferation marker used in pathology to assess tumor proliferation for grading purposes. However, our results suggest that the MIB-1 proliferative index may not be the most appropriate diagnostic marker to use to assess a tumor’s proliferative potential and aggressiveness.
Thus, we conclude that SPARC promotes glioma invasion by mechanisms that both promote tumor cell migration and delay cell proliferation, and that the extent of its effects on both phenotypes depends on the amount of SPARC secreted by the tumor cells. These results support our hypothesis that unlike normal cells in culture that are adherent and express SPARC, the U87T2 parental cells are weakly adherent, and the low and intermediate levels induced by the transfected SPARC induce intermediate and high adherence. However, very high levels of SPARC act to promote its de-adhesive function and, therefore, also induces an intermediate level of adherence (3 , 18) . This interpretation explains the paradoxical situation that both low and high levels of SPARC can promote more aggressive invasion than intermediate levels.
Whether the effects of SPARC on invasion and proliferation are induced by mechanisms that are linked or mutually exclusive is not yet known. For example, attenuation of cell cycle progression may be attributable independently to the direct interference of SPARC with growth factors and their receptors and the resultant attenuation of the growth signaling pathways (5 , 6) . The effects on tumor invasion are likely a result of the effects of SPARC on focal adhesions and cytoskeletal structure, which result in different levels of cell adherence and migration capacity (3) . It is not known whether SPARC induces these changes solely by modulating interactions with integrins and/or ECM proteins at the cell surface or through a signaling mechanism via its putative receptor (3) . However, our cDNA array analyses suggest that the effects of SPARC on tumor invasion and growth are accomplished, in part, through signaling mechanisms that alter the expression of proteins that participate in the regulation of tumor migration and cell proliferation (20) .
In summary, these data indicate that SPARC functionally promotes glioma invasion and delays cell growth in vivo. Because it is up-regulated in all grades of human gliomas, SPARC is a candidate therapeutic target for the design of treatment strategies to inhibit glioma invasion in glioma tumors for all grades of glioma progression.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by NIH Grant R01 CA86997. 
2 To whom requests for reprints should be addressed, at Barbara Jane Levy Laboratory of Molecular Neuro-Oncology, Hermelin Brain Tumor Center, Department of Neurosurgery, Room 3096, Education and Research Building, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI 48202. Phone: (313) 916-8689; Fax: (313) 916-9855; E-mail: nssan{at}neuro.hfh.edu
3 The abbreviations used are: SPARC, secreted protein acidic and rich in cysteine; ECM, extracellular matrix; DAB, 3,3'-diaminobenzidine.
Received 4/26/02. Accepted 9/ 5/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. C. MALONEY, J.-C. MARSHALL, E. ANTECKA, M. E. ORELLANA, B. F. FERNANDES, C. MARTINS, E. CASTIGLIONE, S. DI CESARE, P. LOGAN, and M. N. BURNIER JR. SPARC is Expressed in Human Uveal Melanoma and its Abrogation Reduces Tumor Cell Proliferation Anticancer Res, August 1, 2009; 29(8): 3059 - 3064. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. K. Lee, C. Xiang, S. Cazacu, S. Finniss, G. Kazimirsky, N. Lemke, N. L. Lehman, S. A. Rempel, T. Mikkelsen, and C. Brodie GRP78 is overexpressed in glioblastomas and regulates glioma cell growth and apoptosis Neuro-oncol, January 1, 2008; 10(3): 236 - 243. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Said, I. Najwer, and K. Motamed Secreted Protein Acidic and Rich in Cysteine (SPARC) Inhibits Integrin-Mediated Adhesion and Growth Factor-Dependent Survival Signaling in Ovarian Cancer Am. J. Pathol., March 1, 2007; 170(3): 1054 - 1063. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. O. Sakariassen, L. Prestegarden, J. Wang, K.-O. Skaftnesmo, R. Mahesparan, C. Molthoff, P. Sminia, E. Sundlisaeter, A. Misra, B. B. Tysnes, et al. Angiogenesis-independent tumor growth mediated by stem-like cancer cells PNAS, October 31, 2006; 103(44): 16466 - 16471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Said and K. Motamed Absence of Host-Secreted Protein Acidic and Rich in Cysteine (SPARC) Augments Peritoneal Ovarian Carcinomatosis Am. J. Pathol., December 1, 2005; 167(6): 1739 - 1752. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhou, J. Zhang, Q. Liu, R. Bell, D. A. Muruve, P. Forsyth, M. Arcellana-Panlilio, S. Robbins, and V.W. Yong The chemokine GRO-{alpha} (CXCL1) confers increased tumorigenicity to glioma cells Carcinogenesis, December 1, 2005; 26(12): 2058 - 2068. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Koblinski, B. R. Kaplan-Singer, S. J. VanOsdol, M. Wu, J. A. Engbring, S. Wang, C. M. Goldsmith, J. T. Piper, J. G. Vostal, J. F. Harms, et al. Endogenous Osteonectin/SPARC/BM-40 Expression Inhibits MDA-MB-231 Breast Cancer Cell Metastasis Cancer Res., August 15, 2005; 65(16): 7370 - 7377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Alvarez, F. Prada, E. Salvatierra, A. I. Bravo, V. P. Lutzky, C. Carbone, F. J. Pitossi, H. E. Chuluyan, and O. L. Podhajcer Secreted Protein Acidic and Rich in Cysteine Produced by Human Melanoma Cells Modulates Polymorphonuclear Leukocyte Recruitment and Antitumor Cytotoxic Capacity Cancer Res., June 15, 2005; 65(12): 5123 - 5132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N. Rich, C. Hans, B. Jones, E. S. Iversen, R. E. McLendon, B.K. A. Rasheed, A. Dobra, H. K. Dressman, D. D. Bigner, J. R. Nevins, et al. Gene Expression Profiling and Genetic Markers in Glioblastoma Survival Cancer Res., May 15, 2005; 65(10): 4051 - 4058. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Shi, S. Bao, J. A. Maxwell, E. D. Reese, H. S. Friedman, D. D. Bigner, X.-F. Wang, and J. N. Rich Secreted Protein Acidic, Rich in Cysteine (SPARC), Mediates Cellular Survival of Gliomas through AKT Activation J. Biol. Chem., December 10, 2004; 279(50): 52200 - 52209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Puolakkainen, R. A. Brekken, S. Muneer, and E. H. Sage Enhanced Growth of Pancreatic Tumors in SPARC-Null Mice Is Associated With Decreased Deposition of Extracellular Matrix and Reduced Tumor Cell Apoptosis Mol. Cancer Res., April 1, 2004; 2(4): 215 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sivridis, A. Giatromanolaki, and M. I. Koukourakis "Stromatogenesis" and Tumor Progression International Journal of Surgical Pathology, January 1, 2004; 12(1): 1 - 9. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Sangaletti, A. Stoppacciaro, C. Guiducci, M. R. Torrisi, and M. P. Colombo Leukocyte, Rather than Tumor-produced SPARC, Determines Stroma and Collagen Type IV Deposition in Mammary Carcinoma J. Exp. Med., November 17, 2003; 198(10): 1475 - 1485. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. V. Vadlamuri, J. Media, S. S. Sankey, A. Nakeff, G. Divine, and S. A. Rempel SPARC affects glioma cell growth differently when grown on brain ECM proteins in vitro under standard versus reduced-serum stress conditions Neuro-oncol, October 1, 2003; 5(4): 244 - 254. [Abstract] [PDF]
|
|
|
|
|
|
B. Hu, P. Guo, Q. Fang, H.-Q. Tao, D. Wang, M. Nagane, H.-J. Su Huang, Y. Gunji, R. Nishikawa, K. Alitalo, et al. Angiopoietin-2 induces human glioma invasion through the activation of matrix metalloprotease-2 PNAS, July 22, 2003; 100(15): 8904 - 8909. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
