CG&D
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cell Growth & Differentiation
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kraemer, M.
Right arrow Articles by Binétruy, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kraemer, M.
Right arrow Articles by Binétruy, B.
Cell Growth & Differentiation Vol. 10, 193-200, March 1999
© 1999 American Association for Cancer Research

Rat Embryo Fibroblasts Transformed by c-Jun Display Highly Metastatic and Angiogenic Activities in Vivo and Deregulate Gene Expression of Both Angiogenic and Antiangiogenic Factors1

Michel Kraemer2, Roselyne Tournaire2, Valerie Dejong, Nicole Montreau, Dominique Briane, Claude Derbin and Bernard Binétruy3

Université Paris XIII, UFR Léonard de Vinci, Laboratoire d’Oncologie cellulaire et d’imagerie des tumeurs, 93017 Bobigny cedex [M.K., R.T., D.B., C.D.], and Institut de Recherche sur le Cancer, 94801, Villejuif, cedex [V.D., N.M., B.B.], France


   Abstract
TOP
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 References
 
The comparative tumorigenicity in rats and nude mice of cell lines derived from FR3T3 and transformed by either c-jun, ras, SV40 lt, or bovine papilloma virus type 1 (BPV1) oncogenes was investigated. c-Jun-transformed cells were as tumorigenic and metastatic as Ras-transformed cells. Latencies were short, and numerous pulmonary metastases were observed in all injected animals. In contrast, tumors induced by s.c. injection of SV40-transformed cells developed slower, and none of the animals who received injections i.v. presented with metastases. BPV1-transformed cells had an intermediate tumorigenic and metastatic activity. Microvessels present in the different tumors were revealed by immunostaining with Griffonia (Bandeiraea) Simplicifolia lectin 1. Tumors obtained with c-Jun-transformed cells exhibited more neovascularization than those induced by the other oncogenes. By comparison to FR3T3 cells or SV40- or BPV1-transformed cells, c-Jun-transformed fibroblasts repress the antiangiogenic thrombospondin-1 and SPARC genes, whereas we found that they express higher levels of gene expression of the angiogenic vascular endothelial growth factor. Finally, as compared with cells before passage in animals, thrombospondin-1, SPARC, and VEGF gene expression was also deregulated in cell lines isolated from primary tumors induced by BPV1-transformants. Our results indicate that the high transforming potential of c-Jun, evidenced as soon as transformation is established in vitro, correlates with deregulation of gene expression of both angiogenic and antiangiogenic factors leading to rapid neovascularization of tumors.


   Introduction
TOP
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 References
 
The transfer of cloned oncogenes into fibroblast cell cultures is a powerful model that has been extensively used to study the molecular events leading to cell transformation and tumorigenesis. Although the precise biochemical mechanisms taking place in transformed cells have been analyzed and dissected at the molecular level, the in vivo mechanisms of tumorigenesis remain poorly understood. An important feature of tumorigenesis is tumor neovascularization, on which depends the growth of solid tumors beyond 2–3 mm in diameter (1) . Tumor angiogenesis, quantified after intratumoral microvessel density in primary tumors, has been reported to be associated with metastasis in various neoplasms (2 , 3) . More particularly, intratumoral microvessel density is highest in relatively aggressive tumors (4) and may, thus, serve as a prognostic factor (5 , 6) .

In different cellular models, including rat or mouse fibroblasts, cellular transformation by the activated Ras oncogene leads to up-regulation of genes encoding angiogenic factors, such as the VEGF4 or the FGF-2 (7, 8, 9) . Via direct induction of angiogenesis, these results show a link between a specific molecular function of an oncogene product and tumorigenicity. Other works have led to the concept that oncogenes and inactivated tumor suppressor genes contribute directly to induction of the tumor angiogenic phenotype (10) . Activated Ras proteins elicit pleiotropic effects in cells. Ras can turn on signal transduction cascades activating specific transcription factors, one of them being the c-Jun oncogene (11, 12, 13) . The respective contributions of the different pathways in the angiogenic effect of Ras are unclear. Particularly, we do not know whether the c-Jun-induced transformation process involves activation of angiogenesis.

The nontumoral-established rat fibroblast cell line FR3T3 (14) is sensitive to transformation by a variety of viral and cellular oncogenes. The aim of this work was to determine the tumorigenicity and the angiogenic potential of FR3T3-derivative transformants obtained by gene transfer of the c-Jun oncogene. Biological properties of these cell lines can be compared with FR3T3 transformants obtained with either the Ras oncogene or SV40 or BPV1 viral oncogenes. Presumably, the oncoproteins encoded by the latter use different molecular mechanisms to induce cellular transformation. SV40 lT oncoprotein interferes with the activities of p53 and retinoblastoma cellular proteins encoded by these tumor suppressor genes (15 , 16) . The precise effect of BPV1 oncoproteins in cellular transformation is still unclear, but oncoproteins from human papillomaviruses are also known to counteract p53 and retinoblastoma (17 , 18) . In contrast, the oncogenic potential of c-Jun in rat fibroblasts requires its activity as a transcription factor (19) . Furthermore, c-Jun-transformed fibroblasts specifically regulate the expression of genes potentially involved in angiogenesis, such as those encoding the antiangiogenic matricellular proteins TSP-1 and SPARC (20) . Interestingly, the expression of these target genes is not affected by SV40 or BPV1 oncogenes, confirming that these oncogenes use different molecular mechanisms to transform cells.

We demonstrate that c-Jun-transformed cells are highly tumorigenic, metastatic, and neoangiogenic in vivo and that these biological properties correlate with specific deregulation of expression of both angiogenic and antiangiogenic genes in c-Jun-transformed cells.


   Results
TOP
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 References
 
c-Jun-transformed Cells Are Highly Tumorigenic and Metastatic in Vivo.
FR3T3 cells transformed by either SV40, BPV1, Ras, or c-Jun oncogenes have been shown to be tumorigenic (20 , 21) . All c-Jun-transformed cell lines tested induce tumors with short latency periods, similarly to high tumorigenic Ras-transformed cells (20) . By contrast, SV40-transformed FR3T3 cell lines are poorly tumorigenic, inducing tumors only after long latency periods (22) . BPV1-transformed cells induce tumors with intermediate latency periods (21) . It is noteworthy that, despite some variations in their in vitro-transformed characteristics, FR3T3 derivatives transformed by one specific oncogene display in a given experiment a relatively homogenous tumorigenicity, as measured by the latency period. Because these studies were not performed in parallel experiments, a direct comparison of tumorigenicity is not possible. We, therefore, analyzed the tumorigenicity of several cell lines in concomitant injections. As characteristics of the tumors, we monitored the lag before the development of tumors and measured the surface of the tumors after various times. Cell lines representative of the different oncogenes were analyzed: two transformed by c-Jun (FRcJ-3 and FRcJ-4), one transformed by BPV1 (RV145-4), and one transformed by SV40 lT (SVWT-N2). Cells (1 x 106) of each cell line were injected s.c. into rats. FRcJ-3, FRcJ-4, and RV145-4 cell lines induced sarcomas in all injected animals after a latency period of 1 week (Table 1)Citation . However, the growth of c-Jun-induced tumors was faster than that of BPV1-induced tumors: sixteen days after injection, c-Jun-induced tumors were between 13 and 21 times larger than BPV1-induced tumors (Fig. 1A)Citation . The SVWT-N2 cell line presented a very low tumorigenicity, with a long latency period (41 days for the appearance of the first tumor and 50 days for 100% incidence; Table 1Citation ).


View this table:
[in this window]
[in a new window]
  		Table 1 Tumorigenesis after s.c. injection of 1 x 106 cells into Fisher rats and nude mice

 


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 1. Tumor growth after injection of 1 x 106 transformed cells of the different cell lines into Fisher rats (A) and nude mice (B). The tumor surface area, as a measure of growth, is plotted against time. # indicates the end of the experiment (see text for details).

 
Tumorigenicity was also tested in nude mice (Fig. 1B)Citation . Although latency periods were shorter than in rats, the tumorigenicity of SVWT-N2 cells compared with the other cell lines was very low. Surprisingly, unlike the findings in rats RV145-4 cells were as tumorigenic as FRcJ-3 cells in nude mice. FRcJ-4 cells were the most tumorigenic, but tumors were only up to 6 times larger than those induced by BPV1. Nevertheless, in mice, the tumors induced by both c-Jun-transformed cell lines were all necrotic and hemorrhagic, and the animals either died or had to be sacrificed by day 24 after injection. c-Jun-transformed cells similarly induced necrotic and hemorrhagic tumors in rats. In contrast, no necrosis was observed in BPV1-induced tumors.

To know whether these in vivo results were a simple reflection of the in vitro growth properties of the different cell lines, we measured their doubling time in culture. c-Jun-transformed cells grew faster (generation time, 12 hours) than normal FR3T3 cells, BPV1-transformed, or SV40-transformed FR3T3 cells (generation time, 18 hours). Discrepancies became apparent between in vitro parameters and in vivo parameters. For example, in culture, RV145-4 cells grow slower than FRcJ-3 cells and present the same doubling time as SVWT-N2 cells, but RV145-4 tumors grow faster than SVWT-N2 tumors. Therefore, the different tumorigenic potentials of these cell lines cannot be simply explained by differences in their in vitro growth properties.

Metastatic potential of the different cell lines was investigated after i.v. injection into the tail of nude mice. As a positive control, an FR3T3-derived cell line transformed by the activated Ras oncogene was also injected. Fig. 2Citation shows survival of the animals plotted against time. Autopsy revealed lung metastasis in most animals. No mouse survived the injection of 2 x 104 c-Jun- or Ras-transformed cells for more than 5 weeks. In contrast, all animals injected with BPV1- or SV40-transformed cells were alive after 11 weeks (when we stopped the experiment; Fig. 2Citation ). When 2 x 105 cells of each line were injected, the results were similar, except that the BPV1-transformed cells were partially metastatic (one survival among four injected animals) and with longer latency period than c-Jun- or Ras-transformed cells (data not shown).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 2. Metastatic potential of the different transformed cell lines after i.v. injection into nude mice of 2 x 104 cells. Four to seven animals received injections for each cell line. Values are the percentage of animals surviving at the indicated times.

 
As observed by the growth of primary tumors and by the analysis of the metastatic potential, c-Jun transformation is associated with a high tumorigenic potential in rats and mice. This potential is much higher than the potential of viral oncogenes and reaches values similar to Ras transformation.

c-Jun-transformed Cells Are Highly Neoangiogenic in Nude Mice.
We tested whether the strong tumorigenicity associated with c-Jun correlates with a high neovascularization of the tumors. Microvessel density on histological tumor sections is heterogeneous. The areas of highest microvessel count (i.e., "hot spots") reveal the presence of the most angiogenic tumor cells. As these cells are presumably those influencing the clinical behavior of the tumor, hot spots are presently widely used to assess angiogenesis (23) . Paraffin sections were prepared from mouse tumors of similar sizes induced by the different cell lines. Endothelial cells were stained indirectly to identify tumor vessels. Intratumoral microvessels can be visualized using immunohistochemical methods with a variety of endothelial-specific antibodies. We used the GSL I-isolectin B4, which is useful as a marker for endothelial cells in nonprimate mammals such as mouse, rat, rabbit, and goat (24, 25, 26) . This lectin agglutinates blood group B cells and is specific for {alpha}-galactosyl residues. Two measures were used: the microvessel area and the microvessel count (see "Materials and Methods"). The quantitative results are summarized in Table 2Citation .


View this table:
[in this window]
[in a new window]
  		Table 2 Microvessel areas and counts in tumors induced in nude mice

 
Tumors obtained with FRcJ-4 or FRcJ-3 cells had significantly higher microvessel areas than those obtained with RV145-4 and SVWT-N2 cells (P < 0.001). Thus, as assessed from microvessel area, c-Jun-transformed cells induced about twice as much neoangiogenesis as SV40- or BPV1-transformed cells. Similarly, microvessel counting indicated that FRcJ-3 and FRcJ-4 cells induced, respectively, 2 and 1.7 times more microvessels than SV-40- and BPV1-transformed cells (in both comparisons, P < 0.001).

We then investigated whether these differences in neoangiogenesis could be detected by analyzing the tumors at earlier stages. Injections of FRcJ-4 and RV145-4 cells were administered, and tumors were dissected at various times thereafter. Tumors were classified into three groups according to their weight (Table 3)Citation . In the tumors obtained with BPV1-transformed cells, there was a small but significant increase in the microvessel count during the growth of the tumors (i.e., between the groups 1 and 3: 41.7 ± 20 versus 47.35 ± 8; P < 0.046). The increase in neoangiogenesis evaluated as microvessel area was greater: the area was 2.5 times larger in group 3 than group 1 tumors. Surprisingly, tumors obtained with c-Jun-transformed cells did not exhibit significant differences according to weight (i.e., hot spot scores were similar in all FRcJ-4-induced neoplasma, and the microvessel area was only 1.4 times greater in group 3 than group 1 tumors). Fig. 3Citation illustrates this characteristic of c-Jun-induced tumors: well-formed vessels with lumen were observed even in small group 1 FRcJ-4 tumors. These results indicate that the neovascularization starts very early in these tumors and that there was no, or only a marginal, increase with growth. For each of the three groups of tumors, the microvessel count was significantly higher (between 1.2 and 1.3 times more; P < 0.001) in FRcJ-4-induced tumors than in RV145-4-induced tumors. The differences in microvessel areas were even more pronounced (2.5-fold for group 1 of tumors; Table 3Citation ).


View this table:
[in this window]
[in a new window]
  		Table 3 Microvessel areas and counts in FRcJ-4- and RV145-4-induced tumors in nude mice. For each cell line, the tumors are classified into three groups according to their weight.

 


View larger version (119K):
[in this window]
[in a new window]
  		Fig. 3. Neoangiogenesis in tumors induced by injection of either RV145-4 (left) or FRcJ-4 (right) cells. Photographs of paraffin sections of small (group 1, top) and large (group 3, bottom) tumors are shown. Sections were treated as described in "Materials and Methods." Host endothelial cells appear stained brown. Nuclei are counter-stained to visualize other tumor cells. Vessels with lumen are visible in all photographs except that of RV145-4 group 1 tumors.

 
Thus, the analysis of two measures of angiogenesis indicated that c-Jun-transformed cell lines induced more tumor vascularization than cell lines transformed by the other tested oncogenes. Furthermore, in the case of FRcJ-4 cells, the induction of vascularization was detectable in the smallest (group 1) tumors.

Deregulation of Gene Expression of Both Angiogenic and Antiangiogenic Factors in c-Jun-transformed Cells.
We then investigated whether these in vivo properties could be correlated with specific changes in gene expression. c-Jun-induced transformation in rat cells is associated with repression of the antiangiogenic (TSP-1 and SPARC genes (20) . However, the regulation of genes encoding angiogenic factors by oncogenic transcription factors is not known. We analyzed their expression by Northern blots in our cellular model. No specific signal was detected using either FGF-1 or FGF-2 cDNA radioactive probes, even with 5 µg of polyA+ mRNA of the different cell lines (data not shown). Again, no signal was detected with RNAs from normal FR3T3 hybridized to a VEGF cDNA probe (data not shown). By contrast, strong bands were observed with 5 µg of total RNA from FRcJ-4 cells (Fig. 4Citation , Lane 1), FRcJ-3 or Ras-transformed cells (data not shown). Using a radio-imager, quantitation measures of these bands do not show significant differences in VEGF expression between the three cell lines. VEGF expression was also detected with RNAs from RV145-4 cells (Fig. 4Citation , Lane 4) and SVWT-N2 (data not shown), but to levels 5- and 15-fold lower, respectively. Therefore, as compared with normal FR3T3 cells, VEGF expression was found up-regulated in all transformed cells, and its level of activation correlates with the tumorigenic potential. The lowest VEGF expression was found with the least tumoral cell line: SVWT-N2. Our results could suggest that the VEGF gene promoter might be a new c-Jun target. Using a VEGF promoter/luciferase construct, we performed transient transfection experiments. Although basal expression of the promoter was detected in rat fibroblasts, no activation was found when we cotransfected a c-Jun expression vector, even with a construct that includes the AP1 site located at position -1131 (Ref. 27 ; data not shown).



View larger version (78K):
[in this window]
[in a new window]
  		Fig. 4. VEGF, TSP-1, and SPARC expression levels in FR3T3 cells transformed by either c-Jun or BPV1 oncogenes, before and after passage in animals. Northern blot analysis of 5 µg of total RNA/lane. Bottom, an ethidium-bromide photograph is shown to demonstrate the equal amount of RNA. The following cell lines were analyzed: FRcJ-4 (Lane 1), FRcJ4-TD1 (Lane 2), FRcJ4-TD2 (Lane 3), RV145-4 (Lane 4), RV145-4-TD3 (Lane 5), and RV145-4-TD4 (Lane 6). The filter was successively probed with VEGF, TSP-1, and SPARC cDNAs. Results of quantification analysis are given for each cell line and normalized to 1 (in the case of VEGF probe) or 100% (in the cases of TSP1 and SPARC) for RV145-4 cells.

 
BPV1-transformed FR3T3 cells acquire a more advanced transformed phenotype during the passage in animals (21) . We investigated a potential change in gene expression during this step. Two cell lines derived from two distinct FRcJ-4-induced tumors (FRcJ4-TD1 and -TD2) and two cell lines derived from two RV145-4-induced tumors (RV145-4-TD3 and -TD4) were established in culture. Total RNA (5 µg) of each cell line was analyzed by Northern blot using either VEGF, TSP-1, or SPARC cDNA probes (Fig. 4)Citation . Whereas FRcJ4 tumor-derived cell lines do not present significant changes in gene expression (compare Lanes 2 and 3 with Lane 1), RV145-4 tumor-derived cell lines present a 2–3-fold activation of VEGF expression, a 3–6-fold repression of TSP1 expression, and a 1.4–2-fold repression of SPARC expression as compared with their respective levels in RV145-4 cells (compare Lanes 5 and 6 with Lane 4). These results indicate that the phenotypic changes observed in the case of BPV1 transformants after passage in animals is accompanied by a deregulation of expression of genes encoding factors involved in angiogenesis. In contrast, c-Jun transformants do not modify the expression of these genes during the in vivo passage.

We investigated whether the up-regulation of VEGF and repression of TSP-1 and SPARC in RV145-4 tumor-derived cell lines correlate with an activation of c-Jun during the tumoral progression. We first analyzed the expression of the transin/stromelysin1 gene, as a classical target gene activated by c-Jun, and the c-Jun gene itself. As VEGF expression, transin expression was found up-regulated in RV145-4 tumor-derived cell lines as compared with RV145-4 cells (Fig. 5ACitation , Lanes 5 and 6 compared with Lane 4). The activation of transin reaches the levels found in c-Jun-transformed cells or their tumor derivatives (Fig. 5ACitation , Lanes 1–3). This result was, thus, compatible with a c-Jun activation during tumoral progression. However, Northern blot analysis of RNAs from FRcJ-4, RV145-4, and their corresponding tumor derivatives did not evidence any significant change in endogenous c-Jun expression (Fig. 5A)Citation . Nevertheless, in BPV1 tumor-derived cell lines, c-Jun could still be activated at the protein level without change in gene expression, for example, by change in its phosphorylation status. We, therefore, measured in the different cell lines the relative specific AP1 transcriptional activity as a marker of c-Jun activation. We transfected the synthetic plasmid 5 x TRE-tk-CAT, which contains five binding sites for c-Jun protein cloned in front of the tk-CAT reporter construct. As control, we transfected the {delta}-TRE-tk-CAT plasmid, which is deleted from the TRE sites (28) . After dosage of the CAT activities in transfected cells, we calculated the 5 x TRE-tk-CAT:{delta}-TRE-tk-CAT ratio as an indicator of the specific AP1 activity. As expected, we found high relative AP1 activities in FRcJ-4 and cJ-4-TD1, 8 and 5.4, respectively (Fig. 5B)Citation . By contrast, modest AP1 activities, all comprised between 1 and 2, were found in either FR3T3, RV1454, RV145-4-TD3, or RV145-4-TD4 cell lines (Fig. 5B)Citation . Our data demonstrate that there is no c-Jun activation during the tumoral progression of BPV1-transformed cells, but evidence of c-Jun-independent deregulations of VEGF, Transin, TSP-1, and SPARC gene expression appears in this model.



View larger version (51K):
[in this window]
[in a new window]
  		Fig. 5. Transin expression and endogenous mRNA level and specific transcriptional activity of c-Jun in transformed cells before and after passage in animals. A, Lanes 1–6: Northern blot analysis of the same mRNA and from the same cell lines as in Fig. 4. The filter was probed with c-Jun and transin cDNAs, as indicated. Results of quantification analysis are given for each cell line and normalized to 1 for RV145-4 cells. B, relative AP-1 transcriptional activity. Cells (3 x 105) of the different cell lines were transfected with 1 µg of either 5 x TRE-tk-CAT or {delta}TRE-tk-CAT constructs. For each cell line, the 5 x TRE-tk-CAT:{delta}TRE-tk-CAT ratio between CAT activities is shown.

 

   Discussion
TOP
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 References
 
The molecular events leading to cell transformation and tumorigenicity have been extensively investigated in a large variety of biological systems. There have been studies of natural spontaneous tumors and experimental tumors produced by a variety of means, such as physical or chemical inducers or transfer of cloned viral or cellular oncogenes. The comparison of these complementary approaches is limited by the specificity of the different mechanisms involved in each particular biological system. Using a single cellular background, nontumoral FR3T3 fibroblasts, we undertook an analysis of the in vivo tumorigenic properties of cell lines transformed by oncogenes known to act by different molecular mechanisms. We compared Ras- or c-Jun-transformed cells (oncogenes acting on the same pathway) with cells transformed by the SV40 and BPV1 viral oncogenes.

Similarly to Ras-transformed cells, c-Jun-transformed FR3T3 fibroblasts were highly aggressive on subcutaneous or i.v. injection into animals. In contrast, FR3T3 transformed by the SV40 lT viral oncogene displayed a low tumorigenic potential, and those transformed by BPV1 oncogenes had an intermediate potential. We evidenced two main in vivo properties of c-Jun-transformed cells: (a) the primary tumors grew faster than those induced by BPV1 or SV40 (Table 1Citation and Fig. 1Citation ); and (b) they are more metastatic than BPV1- or SV40-transformed cells (Fig. 2)Citation . Several studies have demonstrated that tumor angiogenesis correlates with metastasis in invasive carcinomas (1 , 2 , 23 , 29) . Thus, our results may be due to c-Jun-transformed cells having a higher angiogenic potential than SV40- or BPV1-transformed cells. Using two measures of angiogenesis, microvessel count, and microvessel area, we demonstrate that the cellular transformation induced by the c-Jun oncogene correlates with a higher neovascularization of the tumors than of tumors induced by viral oncogenes (Table 2)Citation . Furthermore, neovascularization of RV145-4 tumors developed slowly, whereas the neovascularization of FRcJ-4 tumors appeared very early and did not increase significantly during growth (Table 3Citation and Fig. 3Citation ). Our results demonstrate that the c-Jun-induced transformation process is associated with an early and strong stimulation of angiogenesis. This is strengthened by the observation that two specific c-Jun targets, TSP-1 and SPARC, are genes involved in angiogenesis processes (20) . Because the stimulation of angiogenesis is also found in the case of Ras transformation, it is possible that in this case the stimulatory effect is mediated by the activation of endogeneous c-Jun proteins. This hypothesis is in good agreement with the phenotypic reversion observed after overexpression of trans-dominant negative c-Jun mutant proteins into Ras-transformed cells (30) .

It has been found in various cellular systems that one of the consequences of Ras transformation is the activation of gene expression of angiogenic factors such as VEGF or FGF-2. Conversely, previous data from our laboratory showed that, by comparison with normal FR3T3 cells, expression of the antiangiogenic factors TSP-1 and SPARC in either Ras- or c-Jun-transformed FR3T3 cells is strongly repressed (20) . Stimulation of angiogenesis by both up-regulation of angiogenic factors and down-regulation of antiangiogenic factors has been described in transformed human fibroblasts (31) . It is likely that fibroblasts transformed by c-Jun or Ras present also this dual regulation. To test this hypothesis, we performed Northern blot experiments (Fig. 4)Citation . No VEGF expression was detected in normal FR3T3 and, as expected, a strong expression was found in Ras-transformed cells. We found that c-Jun-transformed cell lines express VEGF to the same levels as Ras-transformed cells. These results suggest that VEGF gene represents a new c-Jun target gene. However, the VEGF promoter is not activated by transient transfections with a c-Jun expression vector, suggesting that VEGF activation is not an immediate early event of the c-Jun transformation process. Additional experiments are necessary to understand the molecular mechanisms involved in this regulation. It is possible, for example, that c-Jun acts synergistically with another transcription factor not available during the early step of transformation. Such a factor could be the hypoxia-inducible factor-1, which has been shown to synergize with c-Jun to transactivate the VEGF promoter in transformed cells (27) .

Regarding tumorigenicity, the importance of the concomitant dual regulation of angiogenic and antiangiogenic factors is strengthened by the analysis of their expression in cell lines isolated from tumors. Before passage in animals, BPV1-transformed cells present the same level of TSP-1 and SPARC gene expression as in FR3T3 cells (20) . Surprisingly, a repression of these genes is observed in BPV1 tumor-derived cell lines. This repression reaches similar levels as those in c-Jun- or Ras-transformed cells. Conversely, VEGF and transin/stromelysin1 expressions are up-regulated in BPV1 tumor-derived cell lines as compared with RV145-4 cells. Such changes in TSP-1, SPARC, VEGF, and transin gene regulation before and after passage in animals are not observed in cell lines derived from tumors induced by c-Jun-transformed cells (Figs. 4Citation and 5)Citation . These results suggest that regulation of angiogenic and antiangiogenic factors in c-Jun-transformed cells has reached threshold levels as soon as transformation is established. In contrast, BPV1-induced transformation does not achieve these levels which, nevertheless, can be reached after passage in animals. This is consistent with the stepwise tumoral progression induced by BPV1 oncogenes described in our cellular model (21) . One of these steps could have been c-Jun activation, resulting in up-regulation of VEGF and transin and repression of TSP-1 and SPARC. However, analysis of c-Jun expression and c-Jun activity did not show any change during BPV1-induced tumoral progression (Fig. 5)Citation . The molecular mechanisms accounting for this gene regulation specific to the passage in animals need to be further characterized.

In conclusion, our findings are consistent with: (a) a direct relationship between the tumorigenic and the angiogenic potentials of a transformed cell; and (b) a correlation between these biological properties and the deregulation of gene expression of both angiogenic and antiangiogenic factors.


   Materials and Methods
TOP
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
References
 
Fibroblast Cell Lines, Tumorigenicity, and Experimental Metastasis.
FR3T3, a clonal diploid fibroblastic line, is derived from Fisher rat embryo (14) . The transformed FR3T3 derivatives used have been described in Mettouchi et al. (20) . For each cell line tested, groups of 4–12 (depending on the experiment) female nude mice (nu/nu) or Fisher rats, 5 weeks of age, received injections s.c. in the loose skin of the midback with 1.0 x 106 to 5.0 x 106 cells/animal (depending on the experiment). Tumor development was monitored daily. For histochemistry, the animals were sacrificed at various times after injection, and the tumors were dissected out and weighed. The metastatic potential of the different cell lines was investigated after i.v. injection in the tail of nude mice. Two doses of cells, 2 x 104 and 2 x 105, for each cell line, were injected.

Tissues, Immunohistochemistry, and Lectin Histochemistry for Endothelial Cells.
The proposed standard methods described by Vermeulen et al. (32) were used for tissue processing, immunostaining of tumor vessels, selection of quantification fields, and microvessel counting. Mice were sacrificed, and the tumors were taken out, weighed, divided into small blocks, fixed in 4% paraformaldehyde, and embedded in paraffin. Sections (5 µm thick) were obtained from each paraffin block. GSL-1 lectin immunohistochemical staining for endothelial cells was performed by the streptavidin-biotin-peroxidase complex method. The sections were deparaffinized with three changes of xylene and rehydrated with four changes of ethanol before immunostaining. After blocking endogeneous peroxidase activity in 3% H2O2 for 5 min, the sections were pretreated with 10% normal goat serum to block nonspecific binding and then incubated with GSL-1 isolectin B4 (diluted in 1:50; Vector Laboratories; Burlingame, CA) for 45 min at room temperature. The sections were then incubated with goat antibody to GSL-1 isolectin B4 (diluted in 1:400; Vector Laboratories) for 30 min, washed, and incubated with biotinylated rabbit antigoat immunoglobulins (diluted in 1:400; Dako, Glostrup, Denmark) for 20 min in a moist chamber at room temperature. The samples were incubated with streptavidin-biotin peroxidase (Dako LSAB kit; Dako) for 10 min using diaminobenzidine tetrahydrochloride as the chromogen. Between each step, the sections were washed three times for 5 min with 10 mM HEPES buffer (pH 8.5) containing 0.1 mM Ca++. The nuclei were lightly counterstained by incubation with Meyer’s hematoxylin for 1 min, and the sections were mounted in Moviol. Negative control slides were treated in the same manner, except that the lectin GSL-1 or the antilectin GSL-1 or the biotinylated rabbit antigoat immunoglobulins were omitted.

Microvessel Quantitation.
Intratumoral microvessel areas were determined using a point-counting grid over the endothelial cells plus the vessel lumen and expressed as a fraction of the total point count of the point-counting grid (33) . For each kind of tumor, 10 randomly selected nonserial sections from each of two blocks/mouse were studied. Each section was scanned at low magnification (x10 objective) to identify within the tumor section the area with the most microvessels. For each section, two or three pictures were taken at a x250 magnification. The microvessel area as a percentage of the tumor tissue area was determined according to the formula S = Pv/(Pt.n).100, where: Pv = number of grid points over endothelial cells or lumen of microvessels; Pt = number of points in a grid (96 points in this study, corresponding to an area of 1.02 mm2 on the picture); and n = number of fields counted. The most representative zone of tumor was studied in each case, and necrotic and hemorrhagic areas and section borders were omitted.

For microvessel scoring, the procedure and criteria followed were those defined by Weidner et al. (23 , 34) regarding hot spots. Using a Reichert-Jung (Polivar) microscope, each tumor was scanned at low magnification (x100) to detect and select the areas with the most intense vascularization. Each count was expressed as the highest number of microvessels identified within any x250 field (1.02 mm2). Any GSL-1-stained endothelial cell or group of cells, with or without a lumen, clearly separate from other cells was considered as an individual vessel. The coefficient of variation (defined as the SD of microvessel counts divided by their mean) was used to assess the variability of counts among different fields of the same tumor. Mean intratumoral microvessel area values in the various tumors were compared using Student’s t test to identify significant differences. For all statistical analyses, the level of significance was set at 0.05.

Northern Blots and CAT Assays.
Total RNA was prepared as described by Chomczynski and Sacchi (35) . Total RNA (5 µg)/lane was blotted onto nylon membranes (Biotrans; ICN), cross-linked with UV, and hybridized to radiolabeled probes synthesized from purified cDNA inserts. Cell lines were transfected with the different CAT reporters, and CAT assays were performed as previously described (36) . Northern hybridization signals and CAT assays were quantified by the STORM (Molecular Dynamics) radio-imager.


   Acknowledgments
 
We thank Gilles Pagès and Jacques Pouyssegur for VEGF cDNA and promoter construct, Anne-Catherine and Hervé Prats for FGF-2 cDNA, Gilda Raguenez for FGF-1 cDNA, and Roger Jouffre and Italina Cerutti for technical expertise with the animals. We also thank David Lawrence for critical reading of the manuscript.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by grants from Association pour la Recherche sur le Cancer (ARC), Foundation de France, and Ligue Nationale Contre le Cancer (to B. B.). Back

2 These authors contributed equally to this work. Back

3 To whom requests for reprints should be addressed, at Institute de Recherche sur le Cancer, CNRS UPR 9079, 7 rue Guy Moquet, 94801, Villejuif, cedex, France. Phone: 33-1-4958-3396; Fax: 33-1-4958-3674; E-mail: binetruy{at}infobiogen.fr Back

4 The abbreviations used are: VEGF, vascular endothelial growth factor; FGF, fibroblast growth factor; BPV1, bovine papilloma virus type 1; TSP-1, thrombospondin-1; GSL I-isolectin B4, Griffonia (Bandeiraea) simplicifolia lectin I; CAT, chloramphenicol acetyltransferase. Back

Received for publication 11/23/98. Accepted for publication 1/13/99.


   References
TOP
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 References
 

  1. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat. Med., 1: 27-30, 1995.[Medline]
  2. Weidner N., Carroll P., Flax J., Blumenfeld W., Folkman J. Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am. J. Pathol., 143: 401-409, 1993.[Abstract]
  3. Gasparini G. Angiogenesis research up to 1996. A commentary on the state of art and suggestions for future studies. Eur. J. Cancer, 32A: 2379-2385, 1996.
  4. Vartanian R. K., Weidner N. Correlation of intratumoral endothelial cell proliferation with microvessel density (tumor angiogenesis) and tumor cell proliferation in breast carcinoma. Am. J. Pathol., 144: 1188-1194, 1994.[Abstract]
  5. Weidner N. Quantification of angiogenesis in solid human tumors: an international consensus on the methodology and criteria of evaluation. Am. J. Pathol., 147: 9-19, 1995.[Medline]
  6. Morphopoulos G., Pearson P., Ryder W. D., Howell A., Harris M. Tumor angiogenesis as a prognostic marker in infiltrating lobular carcinoma of the breast. J. Pathol., 180: 44-49, 1996.[Medline]
  7. Grugel S., Finkenzeller G., Weindel K., Barleon B., Marmé D. Both v-Ha-Ras and v-Raf stimulate expression of the vascular endothelial growth factor in NIH 3T3 cells. J. Biol. Chem., 270: 25915-25919, 1995.[Abstract/Free Full Text]
  8. Rak J., Mitsuhashi Y., Bayko L., Filmus J., Shirasawa S., Sasazuki T., Kerbel R. S. Mutant ras oncogenes upregulate VEGF/VPF expression: implications for induction and inhibition of tumor angiogenesis. Cancer Res., 55: 4575-4580, 1995.[Abstract/Free Full Text]
  9. Iberg N., Rogerlj S., Fanning P., Klagsbrun M. Purification of 18- and 22-kDa forms of basic fibroblast growth factor from rat cells transformed by the ras oncogene. J. Biol. Chem., 264: 19951-19955, 1989.[Abstract/Free Full Text]
  10. Bouck N., Stellmach V., Hsu S. C. How tumors become angiogenic. Adv. Cancer Res., 69: 135-174, 1996.[Medline]
  11. Binétruy B., Smeal T., Karin M. Ha-Ras augments c-Jun activity and stimulates phosphorylation of its activation domain. Nature (Lond.), 351: 122-127, 1991.[Medline]
  12. Smeal T., Binétruy B., Mercola D., Grover-Bardwick A., Heidecker G., Rapp U., Karin M. Oncoprotein-mediated signaling cascade stimulates c-Jun activity by phosphorylation of serines 63 and 73. Mol. Cell. Biol., 12: 3507-3513, 1992.[Abstract/Free Full Text]
  13. Karin M. The regulation of AP-1 activity by mitogen-activated protein kinases. J. Biol. Chem., 270: 16483-16486, 1995.[Free Full Text]
  14. Seif R., Cuzin F. Temperature-sensitive growth regulation in one type of transformed rat cells induced by the tsa mutant of polyoma virus. J. Virol., 24: 721-728, 1977.[Abstract/Free Full Text]
  15. Weinberg R. A. Oncogenes, antioncogenes, and the molecular bases of multistep carcinogenesis. Cancer Res., 49: 3713-3721, 1989.[Free Full Text]
  16. Levine A. J. The p53 protein and its interactions with the oncogene products of the small DNA tumor viruses. Virology, 177: 419-426, 1990.[Medline]
  17. Dyson N., Howley P. M., Münger K., Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science (Washington DC), 243: 934-937, 1989.[Abstract/Free Full Text]
  18. Crook T., Tidy J. A., Vousden K. H. Degradation of p53 can be targeted by HPV E6 sequences distinct from those required for p53 binding and trans-activation. Cell, 67: 547-556, 1991.[Medline]
  19. Alani R., Brown P., Binétruy B., Dosaka H., Rosenberg R., Angel P., Karin M., Birrer M. The transactivating domain of the c-jun proto-oncogene is required for cotransformation of rat embryo cells. Mol. Cell. Biol., 11: 6286-6295, 1991.[Abstract/Free Full Text]
  20. Mettouchi A., Cabon F., Montreau N., Vernier P., Mercier G., Blangy D., Tricoire H., Vigier P., Binétruy B. SPARC and thrombospondin genes are repressed by the c-jun oncogene in rat embryo fibroblasts. EMBO J., 13: 5668-5678, 1994.[Medline]
  21. Grisoni M., Meneguzzi G., De Lapeyrière O., Binétruy B., Rassoulzadegan M., Cuzin F. The transformed phenotype in culture and tumorigenicity of Fischer rat fibroblast cells (FR3T3) transformed with bovine papilloma virus type 1. Virology, 135: 406-416, 1984.[Medline]
  22. De Lapeyrière O., Hayot B., Imbert J., Courcol M., Arnaud D., Birg F. Cell lines derived from tumors induced in syngeneic rats by FR 3T3 SV40 transformants no longer synthesize the early viral proteins. Virology, 135: 74-86, 1984.[Medline]
  23. Weidner N., Semple J., Welch W., Folkman J. Tumor angiogenesis and metastasis-correlation in invasive breast carcinoma. N. Engl. J. Med., 324: 1-8, 1991.[Abstract]
  24. Alroy J., Goyal V., Skutelskye E. Lectin histochemistry of mammalian endothelium. Histochemistry, 86: 603-607, 1987.[Medline]
  25. Laitinen L., Virtanen I., Saxen L. Changes in the glycosylation pattern during embryonic development of mouse kidney as revealed with lectin conjugates. J. Histochem. Cytochem., 35: 55-65, 1987.[Abstract]
  26. Christie K. N., Thomson C. Bandeiraea simplicifolia lectin demonstrates significantly more capillaries in rat skeletal muscle than enzyme methods. J. Histochem. Cytochem., 8: 1303-1304, 1989.
  27. Damert A., Ikeda E., Risau W. Activator-protein-1 binding potentiates the hypoxia-induciblefactor-1-mediated hypoxia-induced transcriptional activation of vascular-endothelial growth factor expression in C6 glioma cell. Biochem. J., 327: 419-423, 1997.
  28. Angel L., Imagawa M., Chiu R., Stein B., Imbra R. J., Rahmsdorf H. J., Jonat C., Herrlich P., Karin M. Phorbolester-inducible genes contain a common cis element recognized by a TPA-modulated trans-acting factor. Cell, 49: 729-749, 1987.[Medline]
  29. Ellis L. M., Fidler I. J. Angiogenesis and metastasis. Eur. J. Cancer, 32a: 2451-2460, 1996.
  30. Lloyd A., Yancheva N, Wasylyk B. Transformation suppressor activity of a Jun transcription factor lacking its activation domain. Nature (Lond.), 352: 635-638, 1991.[Medline]
  31. Volpert O., Dameron K., Bouck N. Sequential development of an angiogenic phenotype by human fibroblasts progressing to tumorigenicity. Oncogene, 14: 1495-1502, 1997.[Medline]
  32. Vermeulen P. B., Gasparini G., Fox S. B., Toi M., Martin L., McCulloch P., Pezzella F., Viale G., Weidner N., Harris A. L., Dirix L. Y. Quantification of angiogenesis in solid human tumors: an international consensus on the methodology and criteria of evaluation. Eur. J. Cancer, 32a: 2474-2484, 1996.
  33. Weibel E. R. Practical methods for biological morphometry Weibel E. R. eds. . Stereological Methods, 1: 1-415, Academic Press London 1979.
  34. Weidner N., Folkman J., Pozza F., Bevilacqua P., Allred Emoore Dmeli S., Gasparini G. Tumor angiogenesis: a new significant and independent prognostic indicator in early-stage breast carcinoma. J. Natl. Cancer Inst., 84: 1875-1887, 1992.[Abstract/Free Full Text]
  35. Chomczynski P., Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156-159, 1987.[Medline]
  36. Mettouchi A., Cabon F., Montreau N., Dejong V., Vernier P., Gherzi R., Mercier G., Binétruy B. The c-jun induced transformation process involves a complex regulation of tenascin-c expression. Mol. Cell. Biol., 17: 3202-3209, 1997.[Abstract]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
U. E. Knies-Bamforth, S. B. Fox, R. Poulsom, G. I. Evan, and A. L. Harris
c-Myc Interacts with Hypoxia to Induce Angiogenesis In vivo by a Vascular Endothelial Growth Factor-Dependent Mechanism
Cancer Res., September 15, 2004; 64(18): 6563 - 6570.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
A. Alfranca, M. D. Gutierrez, A. Vara, J. Aragones, F. Vidal, and M. O. Landazuri
c-Jun and Hypoxia-Inducible Factor 1 Functionally Cooperate in Hypoxia-Induced Gene Transcription
Mol. Cell. Biol., January 1, 2002; 22(1): 12 - 22.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
R. Thomas, L. D. True, J. A. Bassuk, P. H. Lange, and R. L. Vessella
Differential Expression of Osteonectin/SPARC during Human Prostate Cancer Progression
Clin. Cancer Res., March 1, 2000; 6(3): 1140 - 1149.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kraemer, M.
Right arrow Articles by Binétruy, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kraemer, M.
Right arrow Articles by Binétruy, B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cell Growth & Differentiation