| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 7, 2002 as doi:10.1096/fj.02-0050fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratori d’Investigació Gastrointestinal, Institut de Recerca and
Servei de Cirurgia, Hospital de Sant Pau, Barcelona; and
* Laboratori de Recerca Traslacional, Institut Català d’Oncologia, Barcelona, Spain
2Correspondence: Laboratori d’Investigació Gastrointestinal, Institut de Recerca, Hospital de Sant Pau, Avda. Sant Antoni M. Claret, 167, Barcelona, Spain. E-mail: rmangues{at}santpau.es
SPECIFIC AIM
We tested the hypothesis that the molecular nature of K-ras mutation confers distinct transforming capacity in vivo. We transfected immortalized mouse fibroblasts with K-ras genes containing a codon 12 (K12) or a codon 13 (K13) mutation and subcutaneously (s.c.) injected stable K12 and K13 transfectants in nude mice, using 3T3-neo and 3T3-wt transfectants as negative controls. We compared morphologically, functionally and molecularly the generated sarcomas.
PRINCIPAL FINDINGS
1. K12-derived tumors presented different histological appearance and shorter growth rate than K13-derived tumors
Tumors derived from K12 and K13 transfectants differed histologically. All K13-derived tumors but one contained almost exclusively histiocytoid cells (Fig. 1
a). In contrast, two cell populations with different morphological appearance coexisted in K12-derived tumors, one fusocellular and another histiocytoid (Fig. 1d
). The generation of tumors from an identical number of inoculated cells and until a macroscopically detectable size was faster in K12 than in K13 transformants. At 3.6 wk, 8 of 10 K12 inocula had already generated visible tumors. K13 tumors did not become evident until 4.3 wk; their appearance was always lagging behind K12 tumors until 6.3 wk, when all inoculated animals had developed tumors at each of the 10 injected sites.
|
2. K12- and K13-derived tumors presented different apoptotic and mitotic rates
Next, we wanted to test whether their distinct histology and growth rate were associated with distinct mitotic or apoptotic rates. Most K13 tumors presented high levels of apoptosis as assessed by the TUNEL assay and by nuclear staining with Hoechst (Fig. 1b, c
). In contrast, apoptosis was negligible in all K12-derived tumors (Fig. 1e, f
). Mitotic figures were found at high frequency in both transfectant-derived tumors. Nevertheless, K13 tumors registered a significantly higher number of mitotic figures than K12 tumors.
3. K12- and K13-derived tumors presented different molecular profiles
Full-blown K13 tumors overexpressed the K-Ras oncoprotein compared with the K12-derived tumors (Fig. 2
). Specific proteins regulating the apoptotic process downstream of Ras were activated and/or expressed differently between K12 and K13 tumors. All K12-derived tumors showed high levels of AKT activation and no apoptosis. K13-derived tumors showed high levels of apoptosis and low AKT activation, which still varied within a certain range, showing an inverse correlation between these two parameters. An overexpression of the anti-apoptotic proteins bcl-2, NF-
B, and Stat-3 was observed in K12 tumors. In contrast, overexpression of the preapoptotic protein Bax was detected in K13 tumors. No differences were observed regarding JNK activation between both tumor types. Then we analyzed the levels of key proteins in the regulation of cell–cell (E-cadherin) and cell–substrate interactions (FAK), which also play an important role in the regulation of apoptosis. There was a significant increase in the expression of E-cadherin and FAK in K12 tumors.
|
We also analyzed the status of known proliferative pathways downstream of Ras. K13 and K12 tumors both showed similarly high levels of activation of Erk1 and Erk2 and expression of the metabolic protein GAPDH (Fig. 2)
. There were no differences regarding cyclin D1 or PCNA expression between groups. Nevertheless, K13 tumors showed significantly higher cyclin B1 expression than K12 tumors (Fig. 2)
. With the exception of cyclin B1, all other cell cycle regulatory proteins tested were expressed similarly in K12 and K13 tumors. All tested molecular markers of soft tissue sarcoma aggressiveness, including Ki67, p53, and c-myc proteins, were expressed significantly more in K12 than in K13 tumors (Fig. 2)
.
CONCLUSIONS AND SIGNIFICANCE
We found that K-ras codon 12 (K12) and K-ras codon 13 (K13) mutations induce strikingly different soft tissue sarcoma types after s.c. implantation of transformed 3T3 cells in nude mice. This is based on their distinct histopathology, latency of appearance, apoptotic and mitotic rates, expression of markers for aggressiveness, and in the activation of specific signal transduction pathways, which could explain these functional differences. K-ras 13-derived tumors resembled malignant fibrous histiocytomas (MFH), whereas K-ras 12-derived tumors resembled fibrosarcomas. The histiotypic specificity of these two different K-ras mutations in fibroblast transformation appears to agree with the clinical findings in soft tissue sarcomas.
Though both transfectants were tumorigenic, K13 tumors appear to be less aggressive than K12 tumors. Thus, K13 tumors showed lower growth rates than K12 tumors, based on their longer latency period of appearance, and significantly higher expression of K-Ras than K12 tumors. An overexpression of the K-Ras protein, when mutated at codon 13, appears to build up over time in vivo and to be required to maintain the tumor phenotype, whereas an expression similar only to the endogenous level is enough for the K12 oncogene to induce tumorigenesis.
The most remarkable difference related to tumorigenesis involved apoptotic regulation. The diminished tumorigenic capacity of the K-ras codon 13 mutation in this in vivo model was due at least partially to a significantly higher degree of apoptosis. The degree of apoptosis inversely correlated with the level of anti-apoptotic signaling through the AKT pathway. Therefore, the correlation between the degree of AKT activation and apoptosis in all tumor samples appears to indicate that the AKT pathway plays a key role in regulating apoptosis in this model (Fig. 3
). Concurrent changes in other molecules that regulate apoptosis through the AKT pathway (FAK, E-cadherin, bcl-2, NF-B) add support to this conclusion. Other proteins, of which connection to the Ras pathway has not yet been unveiled (Stat-3 and Bax), are regulated in the expected direction. In contrast to AKT, the Erk pathway, another regulator of apoptosis and cell proliferation in several systems, showed no differences in activation between K12 and K13 tumors. These findings are consistent with the reported structural differences between the oncogenic proteins in the Ras effector domain that depend on the affected codon and alter the affinity of the interaction of Ras with its downstream targets.
|
The increased mitotic rate observed in K13 tumors was unexpected, since mitotic index is usually associated with tumor aggressiveness in human sarcomas. Nevertheless, the concurrent increased of the cyclin B1 levels in these tumors suggests that their enhanced mitotic activity may be a consequence of an increased induction of passage through the G2/M transition (Fig. 3)
. This would be opposed to the significantly lower mitotic rate and cyclin B1 expression observed in K12 tumors. In contrast, the G1 transition (as indicated by cyclin D1 expression and Erk activation) was not altered, nor was DNA synthesis as assessed by PCNA expression. Thus, in K13 tumors, growth is the result of a very high mitotic activity that outweighs a high level of apoptotic induction, implying a very high degree of tumor cell turnover. In K12-derived tumors, growth is accomplished despite the low proliferation rate because of a diminished apoptotic induction in a context of low tumor cell turnover. Therefore, tumor cell turnover varied significantly among K-ras transfectants in vivo because of the differences in apoptotic and cell cycle regulation. Other parameters indicated the higher tumorigenic capacity of the K-ras 12 mutations. Thus, in contrast to K13, K12 tumors overexpress Ki67, p53, and c-myc, markers of high-grade enhanced metastatic potential, poor clinical outcome, and/or reduced patient survival in human soft tissue sarcomas (STS).
Our results demonstrate that K12 and K13 mutations generate two distinct soft tissue sarcoma types and suggest that malignant fibrous histiocytomas and fibrosarcomas, despite of sharing a common ontogeny (they both derived from fibroblasts), are completely different entities. Fibrosarcomas appear to be more undifferentiated and aggressive and regulate apoptosis, the cell cycle, and cell–cell and cell adhesion through completely different molecular pathways. MFH appear to be more differentiated and less aggressive; this is accomplished in part by down-regulating the AKT anti-apoptotic pathway. Thus, the position of the K-ras mutation modulates its transforming capacity of fibroblasts by differentially regulating K-Ras downstream pathways, leading to changes in apoptosis, cell turnover, tumor morphology, and aggressiveness. Therefore, nude mice tumors derived from transformation of 3T3 cells by the K-ras codon 13 mutant gene may be a model for MFH whereas K12-derived tumors could be a model for fibrosarcoma. Nevertheless, the confirmation of the histiotypic specificity of K12 and K13 mutations requires a comparative evaluation of apoptosis and its associated signal transduction pathways in human STS bearing these mutations. Our results also suggest that tumor cell turnover, which includes the evaluation of apoptotic and mitotic rates, may complement the present use of mitotic rate to grade STS. Finally, they suggest the need to separately manage these two biologically distinct entities to improve biological understanding and therapeutic outcomes in STS.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0050fje; to cite this article, use FASEB J. (August 7, 2002) 10.1096/fj.02-0050fje 
This article has been cited by other articles:
|
|
 |
|
  Q.-S. Zhu, W. Ren, B. Korchin, G. Lahat, A. Dicker, Y. Lu, G. Mills, R. E. Pollock, and D. Lev Soft Tissue Sarcoma Cells Are Highly Sensitive to AKT Blockade: A Role for p53-Independent Up-regulation of GADD45{alpha} Cancer Res., April 15, 2008; 68(8): 2895 - 2903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. V. Cespedes, F. J. Sancho, S. Guerrero, M. Parreno, I. Casanova, M. A. Pavon, E. Marcuello, M. Trias, M. Cascante, G. Capella, et al. K-ras Asp12 mutant neither interacts with Raf, nor signals through Erk and is less tumorigenic than K-ras Val12 Carcinogenesis, November 1, 2006; 27(11): 2190 - 2200. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Richter, H. Murua Escobar, K. Gunther, J. T. Soller, S. Winkler, I. Nolte, and J. Bullerdiek RAS Gene Hot-Spot Mutations in Canine Neoplasias J. Hered., November 1, 2005; 96(7): 764 - 765. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Vizan, L. G. Boros, A. Figueras, G. Capella, R. Mangues, S. Bassilian, S. Lim, W.-N. P. Lee, and M. Cascante K-ras Codon-Specific Mutations Produce Distinctive Metabolic Phenotypes in Human Fibroblasts Cancer Res., July 1, 2005; 65(13): 5512 - 5515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Lleonart, S. Ramon y Cajal, J. D. Groopman, and M. D. Friesen Sensitive and specific detection of K-ras mutations in colon tumors by short oligonucleotide mass analysis Nucleic Acids Res., March 22, 2004; 32(5): e53 - e53. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
