The p8 protein is involved in the cellular stress response of many tissues. Because p8 is overexpressed in many cancers, we investigated whether its expression was required for tumour development. Mouse embryo fibroblasts (MEFs) from p8+/+ and p8−/− animals were transformed with the pBabe‐rasV12/E1A retroviral vector, which expresses both the rasV12 mutated protein and the E1A oncogene. As expected, transformed p8+/+ MEFs could form colonies in soft agar. However, transformed p8−/− MEFs could not. In addition, subcutaneous or intraperitoneal injections of transformed p8+/+ MEFs always led to tumour formation in nude mice, but, again, no tumour was observed with transformed p8−/− MEFs. However, restoring p8 expression in transformed p8−/− MEFs before injection led to tumour formation. In the tumours, p8 expression was induced during tumour development. It was concluded that p8 expression in transformed MEFs is necessary for tumour formation, suggesting that the stress‐response mechanisms governed by p8 are required for tumour establishment.
Cancer progression occurs in several steps. During transformation, some cancer cells are positively selected within the tumour on the basis of their growth capacity, low response to apoptotic signals and ability to escape the immunological survey of the host. After leaving the primary tumour, transformed cells migrate through the body. However, metastases will not develop in all tissues (Yeatman and Nicolson, 1993). Capacity for invading the target organ is a first limitation. But once within the organ, metastasis will develop only if transformed cells can cope with their new microenvironment. In fact, the phenotype of the cells is adapted to their original environment but not necessarily to the environment generated by the invaded tissue. Therefore, invading cells are exposed to a stress, and their capacity to react by activating stress‐associated genes should be determinant in metastasis formation. Several stress‐associated genes have been found overexpressed in tumours (Jamora et al., 1996; Storm et al., 1996) and their expression level correlated with aggressiveness. Therefore, stress‐associated genes might facilitate tumour progression and metastasis formation by helping cell adaptation to the microenvironment of the host tissue.
Previously, using a systematic approach, we identified a new gene, called p8, whose expression is strongly induced in pancreatic acinar cells during the acute phase of pancreatitis (Mallo et al., 1997). Further experiments have shown that p8 activation is not restricted to pancreatic cells. In vivo, p8 mRNA expression is activated in several tissues in response to systemic liposaccharides (Jiang et al., 1999), and in vitro studies showed that a variety of cell lines exhibited transient p8 mRNA expression in response to several stress agents (Mallo et al., 1997; Vasseur et al., 1999; Garcia‐Montero et al., 2001a) through a p38‐dependent pathway (Garcia‐Montero et al., 2001a). The deduced protein sequence revealed that p8 was an 80‐amino‐acid polypeptide whose primary structure could not be aligned with any of the protein sequences present in public databases. In contrast, p8 contains a canonical bipartite signal for nuclear targeting, suggesting that p8 should be located in the nucleus. We detected p8 within the nucleus of Cos‐7 cells transfected with a p8 expression plasmid, although it was also partly located to the cytoplasm (Vasseur et al., 1999). Furthermore, analysis of p8 primary structure suggested that it should be a DNA‐binding protein (Mallo et al., 1997). More recently, we performed biochemical and biophysical studies showing that human recombinant p8 was in many structural aspects very similar to the high mobility group (HMG) proteins, although sharing with them only 35% amino‐acid sequence homology (Encinar et al., 2001). In addition, p8 binding to DNA, rather weak with the native protein, was strongly enhanced after phosphorylation by PKA of serine/threonine residues (Encinar et al., 2001). HMG protein binding to DNA is also regulated by post‐translational modifications (Bianchi and Beltrame, 2000). Thus, in spite of a low sequence homology, p8 can be considered as an HMG‐like protein. We found recently that p8 is a co‐transcription factor, since it can enhance the transcription activity of the Smad proteins in response to TGF‐β (Garcia‐Montero et al., 2001b). Therefore, p8 is a ubiquitous protein induced by cellular stress that, acting as a co‐transcription factor, may participate in a general defence programme against cellular injury or cellular stress.
Concomitant studies by another laboratory (Ree et al., 1999) suggested that expression of the candidate of metastasis 1 (Com 1) protein, which is identical to the human p8 (Vasseur et al., 1999), could mediate the growth of tumour cells after metastatic establishment in a secondary organ, suggesting that activated expression of p8 in metastatic cells is required for tumour progression. Taken together, these results strongly suggest that p8 expression might regulate cellular functions required for tumour progression and metastasis. To explore the function of p8 in cancer progression, we used embryonic fibroblasts obtained from p8−/− mice and their p8+/+ counterparts. We transformed these fibroblasts and compared their growth properties in vitro and in vivo.
Results and Discussion
A conventional targeting vector was used to ablate p8 exon 2 in mouse embryonic stem cells. Details of this construction can be obtained from the authors upon request. Heterozygous mice for the mutated p8 allele were bred to generate mice homozygous for the disruption of p8. Primary embryo fibroblasts were isolated from 14.5‐day‐old mouse embryos following standard protocols (Harvey et al., 1993). p8 genotypes of cultured cells were confirmed by PCR on genomic DNA. Cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS). Expression of p8 mRNA in p8+/+ and p8−/− mouse embryo fibroblasts (MEFs) was evaluated by RT–PCR using specific primers (Figure 1A). Oncogenic ras transforms most immortal rodent cells to a tumorigenic state, whereas transformation of primary cells requires either a cooperating oncogene or the inactivation of a tumour suppressor. The adenovirus E1A oncogene cooperates with ras to transform primary rodent fibroblasts (Ruley, 1990) and abrogates ras‐induced senescence (Serrano et al., 1997). Therefore, we transduced p8+/+ and p8−/− MEFs with the pBabe‐rasV12/E1A retroviral vector, which expresses both the rasV12 mutated protein and the E1A oncogene, to obtain transformed MEFs with a defined p8 status. As control, we evaluated the ability of the retroviral vector to transduce p8+/+ and p8−/− MEFs by using a retroviral vector expressing the EGFP under the control of the retroviral promoter located in the long terminal repeat. More than 80% of p8‐expressing and p8‐deficient MEFs expressed high levels of EGFP fluorescence 48 h after transduction (data not shown), indicating that the retroviral vectors are well adapted tools to our experimental set‐up. Transformation of p8+/+ and p8−/− MEFs by the pBabe‐rasV12/E1A retroviral vector was evaluated by examining changes in their morphological aspect, by quantifying expression of the RAS protein by western blotting and by monitoring cell proliferation. The original pBabe retroviral vector that only contains and expresses the puromycin resistance gene was used as a control. As shown in Figure 1B, transduction of MEFs with the pBabe‐rasV12/E1A retroviral vector induced important morphological changes typical of cell transformation (i.e. cells lost their flat and enlarged cytoplasm and became refractile with thin projections), whereas no change in morphology was observed in cells transduced with the pBabe retroviral vector; both transformed p8+/+ and p8−/− MEFs expressed high levels of RAS compared with control MEFs transduced with empty vector (Figure 1C); finally, transformed MEFs grew 2–3 times faster than control MEFs (Figure 1D). It is noteworthy that lack of p8 expression did not affect cell growth. On the contrary, p8−/− and transformed p8−/− MEFs grew more rapidly than p8+/+ and transformed p8+/+ MEFs, respectively. Based on these results, we assumed that p8+/+ and p8−/− MEFs were transformed by the pBabe‐rasV12/E1A retroviral vector. The similar features acquired by p8+/+ and p8−/− MEFs upon transformation by rasV12/E1A co‐expression suggested that p8 deficiency did not interfere with MEF transformation.
The tumorigenic properties of transformed p8+/+ and p8−/− MEFs were assessed by checking their ability to grow in vitro and to form tumours in immunocompromised mice. In soft‐agar assays, transformed p8+/+ MEFs formed colonies at high frequency, as expected (Figure 2), but transformed p8−/− MEFs, like control p8+/+ or p8−/− MEFs transduced with the pBabe vector, were unable to form colonies (Figure 2). Similarly, transformed p8+/+ MEFs produced tumours in all athymic nude mice (6/6) when injected subcutaneously, whereas transformed p8‐deficient MEFs, like control MEFs, did not (0/6) (Table 1 and Figure 3). As a control, we restored p8 expression in transformed p8−/− MEFs and found that they had recovered their ability to form tumours when injected into nude mice, as expected (Table 1 and Figure 4). Finally, we injected intraperitoneally the transformed p8+/+ and p8−/− MEFs to test their ability for dispersed growth. Remarkably, only transformed p8+/+ MEFs (Table 1 and Figure 5) and transformed p8−/− MEFs in which p8 expression was restored (Table 1) showed intraperitoneal spreading. Again, transformed p8−/− MEFs, like non‐transformed control MEFs, did not produce dispersed tumours (Table 1). Therefore, although p8+/+ and p8−/− MEFs seem to be similarly transformed by the pBabe‐rasV12/E1A retroviral vector as judged by their behaviour in standard culture conditions (Figure 1D), the inability of the transformed p8‐deficient cells to form colonies in soft agar, form subcutaneous tumours or generate intraperitoneal spreading strongly suggest that expression of p8 is required for the organization and development of tumours.
The molecular mechanisms by which p8 allows tumour progression are unknown. It might involve Ki‐ras and PPARδ pathways, because their disruption and that of the p8 gene result in similar phenotypes (continued growth in vitro but poor growth in vivo) (Shirasawa et al., 1993; Park et al., 2001), but this has not been demonstrated. Nevertheless, in that context, it is conceivable that suppression of p8 expression in cancer cells would impede tumour progression. Consequently, p8 might be a new drug‐targetable gene whose blockade would prevent cancer progression and metastasis development.
So far, p8 induction has been associated with stress (Mallo et al., 1997; Jiang et al., 1999; Vasseur et al., 1999; Garcia‐Montero et al., 2001a), suggesting that activation of p8 in transformed p8+/+ MEFs during subcutaneous or intraperitoneal tumour formation (Figure 6) is a consequence of the stress generated by the new microenvironment of MEFs. We propose that transformed p8‐deficient cells fail to form tumours because they are unable to respond efficiently to such stress. This is supported by clinical data indicating that p8 expression in breast and pancreatic cancers correlates with aggressiveness (Ree et al., 2000; Su et al., 2001) and that metastatic cells express high levels of p8 (Ree et al., 1999; S. Garcia and J. Iovanna, unpublished data). Further studies are required to demonstrate that cells losing their capacity to cope with environmental stress are unable to develop tumours; if this is achieved, it will open up an original and promising area of therapeutic cancer research.
Preparation of p8+/+ and p8−/− MEFs.
Primary embryo fibroblasts were isolated from 14.5‐day‐old p8+/+ and p8−/− mouse embryos following standard protocols (Harvey et al., 1993). Cells were grown in DMEM supplemented with 10% FCS.
Retroviral vector and retroviral‐mediated gene transfer.
pBabe‐rasV12/E1A and pBabe plasmids were obtained from M. Serrano (Spanish National Center of Biotechnology, Madrid, Spain) and S. Lowe (Cold Spring Harbor Laboratory, NY). Bosc 23 ecotropic packaging cells (106) were plated in a six‐well plate, incubated for 24 h and then transfected with polyethylene imine with 5 μg of retroviral plasmid. After 48 h, the medium containing the virus was filtered (0.45 μm filter, Millipore) to obtain the first supernatant. Target MEFs were plated at 2 × 105 cells per 35‐mm dish and incubated overnight. For infections, the culture medium was replaced by an appropriate mix of the first supernatant and culture medium (v/v), supplemented with 4 μg/ml polybrene (Sigma), and cells were incubated at 37°C. MEFs infected with the empty vector were used as control. To restore p8 expression in transformed p8−/− MEFs, we infected these cells with a retrovirus expressing human p8. The retroviral vector was constructed as follows: human p8 cDNA was subcloned in HindIII and XhoI restriction sites of the pLPC plasmid (obtained from S. Lowe). Ecotropic human p8‐expressing retrovirus was then generated by transient transfection using Bosc 23 ecotropic packaging cells. Virus supernatant was used to infect transformed p8−/− MEFs, and the population of p8‐expressing transformed p8−/− MEFs was isolated by selection in the presence of puromycin (0.7 μg/ml). As control, transformed p8−/− MEFs were infected with pLPC empty vector
One‐hundred thousand cells per well were plated in a series of 35‐mm culture dishes. The cell number was estimated daily in triplicate (from 1 to 7 days) in a hemocytometer. Within an experiment, each point was determined in triplicate and each curve was established at least two times.
RNA was extracted using Trizol (Life Technologies) procedures. Total RNA (1 μg) was analysed by RT–PCR with the SuperScript One‐step RT–PCR System and the Platinum Taq kit (Life Technologies). RT–PCR was performed using different numbers of cycles to verify that the conditions chosen were within the linear range. The mRNA coding for p8 was specifically amplified with sense (5′‐GGAGAGAGCAGACTAGGCATA‐3′) and antisense (5′‐GTTGCTGCCACCCAAGGGCAT‐3′) primers, in positions 16 and 581 of the cDNA, respectively. As control, the transcript coding for the ribosomal protein RL3 was specifically amplified for 19 cycles with sense (5′‐Gaaagaagtcgtggaggctg‐3′) and antisense (5′‐Atctcatcctgcccaaacac‐3′) primers, in positions 216 and 637 of the cDNA, respectively.
One‐hundred micrograms of total protein extracted from cells were separated with standard procedures on 12.5% SDS–PAGE using the Mini Protean System (Bio‐Rad) and transferred to a nitrocellulose membrane (Sigma). The intracellular level of RAS was estimated by western blotting using the H‐ras (C‐20) polyclonal antibody (1:200) (Santa Cruz Biotechnology), and p8 was detected with a polyclonal antibody raised against human p8 (Vasseur et al., 1999).
Colony growth assays.
To determine the ability of different cell lines to grow in soft agar, 5 × 104 cells were resuspended in 0.3% agar in DMEM plus 10% FCS and overlaid on 0.6% agar in the same medium in 12‐well plates. Experiments were performed in triplicate and repeated two (pBabe‐transduced cells) or three times (pBabe‐rasV12/E1A‐transduced cells).
In vivo tumour growth, spreading assays and histological analysis.
Suspensions of the p8+/+ or p8−/− pBabe‐rasV12/E1A‐transformed MEFs or the p8+/+ or p8−/− pBabe‐transduced MEFs [106/200 μl phosphate‐buffered saline (PBS)] were injected subcutaneously into the flank of male athymic 7–8‐week‐old nu nu mice, and tumours were allowed to develop for 20 days. Tumours were measured externally with a caliper, in two dimensions, on the indicated days. Tumour volumes were determined from the equation V = (L × W2) × 0.5, where L is length and W is width. For tumour spreading assays, 106 cells in 200 μl PBS were injected intraperitoneally into male nude mice, and animals were killed after 20 days. For histological examination, necropsies were performed on representative animals. Tumours were fixed in 10% buffered neutral formalin, processed in paraffin, sectioned at 5 μm and stained with hematoxylin and eosin.
We thank Drs M. Serrano and S. Lowe for pBabe‐rasV12/E1A and pBabe plasmids and R. Feil and S. Kuhbandner for help in constructing the p8‐deficient mice. We also thank Drs C. Mawas, I. Dikic, A. Ree, S. Moreno and M. Serrano for their help and critical comments. Production of retroviral vector was generously sponsored by the Institut Paoli‐Calmettes (Marseille, France) and the GVPN (Gene Vector Production Network). This work was supported by grants from the Société de Secours des Amis des Sciences and an EMBO short‐term fellowship (to S.V.), the Fondation pour la Recherche Médicale (to A.H.), the Ligue Nationale Contre le Cancer (LNCC) and the Association pour la Recherche sur le Cancer (ARC).
- Copyright © 2002 European Molecular Biology Organization