Peroxisome proliferator-activated receptor gamma and spermidine/spermine N1-acetyltransferase gene expressions are significantly correlated in human colorectal cancer

Background The peroxisome proliferator-activated receptor γ (PPARγ) is a transcription factor that regulates adipogenic differentiation and glucose homeostasis. Spermidine/spermine N1-acetyltransferase (SSAT) and ornithine decarboxylase (ODC) are key enzymes involved in the metabolism of polyamines, compounds that play an important role in cell proliferation. While the PPARγ role in tumour growth has not been clearly defined, the involvement of the altered polyamine metabolism in colorectal carcinogenesis has been established. In this direction, we have evaluated the PPARγ expression and its relationship with polyamine metabolism in tissue samples from 40 patients operated because of colorectal carcinoma. Since it is known that the functional role of K-ras mutation in colorectal tumorigenesis is associated with cell growth and differentiation, polyamine metabolism and the PPARγ expression were also investigated in terms of K-ras mutation. Methods PPARγ, ODC and SSAT mRNA levels were evaluated by reverse transcriptase and real-time PCR. Polyamines were quantified by high performance liquid chromatography (HPLC). ODC and SSAT activity were measured by a radiometric technique. Results PPARγ expression, as well as SSAT and ODC mRNA levels were significantly higher in cancer as compared to normal mucosa. Tumour samples also showed significantly higher polyamine levels and ODC and SSAT activities in comparison to normal samples. A significant and positive correlation between PPARγ and the SSAT gene expression was observed in both normal and neoplastic tissue (r = 0.73, p < 0.0001; r = 0.65, p < 0.0001, respectively). Moreover, gene expression, polyamine levels and enzymatic activities were increased in colorectal carcinoma samples expressing K-ras mutation as compared to non mutated K-ras samples. Conclusion In conclusion, our data demonstrated a close relationship between PPARγ and SSAT in human colorectal cancer and this could represent an attempt to decrease polyamine levels and to reduce cell growth and tumour development. Therefore, pharmacological activation of PPARγ and/or induction of SSAT may represent a therapeutic or preventive strategy for treating colorectal cancer.


Background
The peroxisome proliferator-activated receptor γ (PPARγ) is a member of the nuclear receptor superfamily [1]. It functions as a transcription factor, heterodimerizes with the retinoid X receptor (RXR) and regulates the expression of target genes by binding to the PPAR response elements (PPERs). The PPARγ, initially identified for its role in the metabolism and adipocyte differentiation, is also expressed in tissue outside the adipocyte lineage, including colonic epithelia [2]. In cancer biology PPARγ is the most intensively studied PPAR isoform and several studies have shown the protective role of this nuclear receptor in colorectal tumour [3]. On the contrary, other authors have questioned the antineoplastic activity of PPARγ ligands [4]. In addition, although most of the known target genes activated by PPARγ belong to the pathway of metabolism and lipid transport, the target genes mediating the pro-or anti-cancer activity by affect of the activated PPARγ have not yet been identified.
Recently, Babbar et al. [5] demonstrated that in colorectal tumour cells, the promoter of spermidine/spermine N 1acetyltransferase (SSAT) gene contains a functional PPAR responsive element and activated PPARγ is able to activate transcription of this gene.
SSAT and ornithine decarboxylase (ODC) are key enzymes involved in the catabolism and biosynthesis, respectively, of the polyamines (putrescine, spermidine and spermine), compounds that play an important role in both normal and neoplastic cell proliferation [6]. Increased ODC activity and the associated elevation in intracellular polyamines have been implicated in carcinogenesis of many human tissues, including those of the gastrointestinal tract [7][8][9].
Induction of SSAT plays an important role in lowering the polyamine pool and typically gives rise to growth inhibition as well as exerting tumour-suppressive effects [10,11]. Particularly, Wallace et al. have hypothesized that the increase in SSAT activity may be an attempt to reduce the polyamine pool in tumour cells. [12] Little is know about the relationship between polyamine metabolism and PPARγ expression in colorectal tumours and available information is derived from in vitro studies [5,13].
We have provided evidence about a close relationship between polyamine biosynthesis and K-ras mutation in human colorectal carcinoma [14]. Moreover, in the human colon adenocarcinoma cell line, it has been shown that mutated K-ras suppressed SSAT via a transcriptional mechanism that involves PPARγ signalling pathway [15].
In order to investigate the relationship between PPARγ and polyamine metabolism in human colorectal carcinoma, the aim of this study was to evaluate the PPARγ, SSAT and ODC gene expression as well as polyamine levels and ODC and SSAT activities, in neoplastic samples and in surrounding uninvolved mucosa. The profile of polyamine metabolism and PPARγ mRNA expression were also investigated in relation to the presence of K-ras mutation in our tumour samples. We have exclusively focalised our attention on mutated codon 12 as it occurs more frequently in human colorectal cancer [16].

Patients
Tissue samples were obtained at the time of surgical resection for colorectal adenocarcinoma from 40 patients: 24 men (mean age 64.8 years, range 36-88) and 16 women (mean age 64.4 years, range 28-87). Pieces of tissue were removed from the peripheral region of the tumour and surrounding mucosa in a region at least 10 cm from the neoplasm. The normal mucosa was found to be macroscopically and microscopically free of neoplastic growth. Immediately after removal, tissue was washed in cold saline solution and then rapidly (within 15-30 min) transported on ice to the laboratory and stored at -80°C until assay. The quality of the tumour and surrounding normal tissue specimens and the relative cellular composition were determined by histopathological assessment. The clinical and histopathological features of colorectal cancer patients are reported in Table 1. The design of this study was approved by the local ethics committee.
Informed consent was obtained from all patients before their surgery and examination of specimens used in this study.

Analysis of the PPARγ, ODC and SSAT gene expression
PPARγ, ODC and SSAT mRNA levels were evaluated by reverse transcriptase and real-time PCR. The RNA isolation, reverse transcriptase and real-time PCR (RT-PCR) methods have been described previously [17]. In brief, after RNA extraction, aliquots of 5 µl of the reverse transcriptase reaction (20 µl), corresponding to 0.5 µg of total RNA, were subjected to the RT-PCR in 25 µl of final volume contained master mix (iQ SYBR Green Supermix, BioRad, Milan, Italy). The amplification program on a iCycler iQ Real-Time detection system (BioRad) was: 30 s at 95°C and then 10 s at 94°C, 10 s at 55°C and 30 s at 72°C for 45 cycles. A further melt-curve step (melt-curve started at 55°C and the temperature was increased 0.5°C per cycle for 80 cycles) was added in the thermocycler program at the end of the last extension phase of 45 cycles to control the presence of the secondary non specific PCR products or primer-dimer products. For quantitative analysis the external standard curves were used and the mRNA levels of each target gene were then normalized to those of the β-actin, as control gene.

Polyamine analysis
Each tissue sample (approximately 10-15 mg) was homogenized in 700 µl of 0.9 percent sodium chloride mixed with 5 µl (174 µM) of the internal standard 1, 10-Diaminodecane (1, 10-DAD). In order to precipitate the proteins, 50 µl of perchloric acid (PCA) 3 M were added to the homogenate. After 30 minutes incubation in ice, the homogenate was centrifuged for 15 minutes at 7000 g. The supernatant was filtered (Millex-HV 13 pore size 0.45 µm, Millipore, Bedford, MA) and lyophilized. The residue was dissolved in 250 µl of HCl (0.1 M). Aliquots (100 µl) were reacted with dansyl chloride and the dansylpolyamine derivatives were determined by high performance liquid chromatography (HPLC) as previously described [7]. Polyamine levels were expressed as concentration values in nmol/g wet weight tissue.

ODC activity
ODC activity was measured with a radiometric technique as previously described [14]. In brief, the reaction mixture consists of DL-[1-14 C]-ornithine and tissue extract in Tris-HCL buffer; the 14 CO 2 product generated by the enzyme reaction was trapped on filter paper pre-treated with NaOH, which was suspended in a center well above the reaction mixture. Radioactivity on the filter papers was determined by a liquid scintillation counter (model 1219 Rackbeta, LKB-Pharmacia, Uppsala, Sweden). ODC activity was expressed as pmol CO 2 /h/mg of protein.

SSAT activity
SSAT activity was measured according to the method described by Wallace et al. [18] with minor modifications. The reaction was started by adding 70 µL of cytosol to a microfuge tube containing 30 µL of incubation buffer (30 mM spermidine, 0.1 µCi [acetyl-1-14 C] acetyl coenzyme A, 1 M Tris-HCL, pH 7.8). The incubation was carried out for 10 min at 37°C, then 20 µL hydroxylamine hydrochloride were added to stop the reaction. The tubes were boiled for 3 min with pierced tops and then spun in a microfuge for 5 min. A quantity of 30 µL of supernatant was spotted on to Whatman P81 discs which were then washed five times with distilled water and three times with 100% ethanol. The radioactivity on the filter papers was determined by a liquid scintillation counter. SSAT activity was expressed as pmol of N 1 -[ 14 C] acetylspermidine generated/min/mg of protein.

K-ras mutation analysis
Genomic DNA was isolated from 50 mg of normal mucosa and tumour tissue by TRI-Reagent (Mol. Res. Centre Inc. Cincinnati, OH, USA) following manufacturer's instruction. The extracted DNA was dissolved in 50 µl of sterile water and used for PCR. Mutations in K-ras exon 1, codon 12 were screened by PCR amplification followed by RFLP analysis as previously described [14].

Statistical analysis
Differences among polyamine levels, ODC activity, SSAT activity, and mRNA expression of relative genes were analyzed by Wilcoxon matched pairs test and Mann Whitney test when appropriate. The significance of differences, for the above parameters, in location of tumour, Dukes'stages, and differentiation grade was determined by Kruskal-Wallis analysis of variance. Correlations were evaluated by Spearman correlation test. All data are expressed as mean ± SEM. A p-value < 0.05 was considered as significant.

Results
The PPARγ, ODC and SSAT mRNA levels were significantly higher in colorectal carcinoma than in normal surrounding mucosa. In addition, polyamine levels, ODC and SSAT activities were significantly increased in neoplastic samples than in normal ones (p < 0.05; Wilcoxon matched pairs test) ( Table 2). Of note, a significant and positive correlation between PPARγ and SSAT gene expression was observed in both normal and neoplastic tissue (r = 0.73, p < 0.0001; r = 0.65, p < 0.0001, respectively; Spearman correlation coefficient) (Figure 1).
K-ras codon 12 mutation was found in 12 out of 40 patients (30%). PPARγ expression, as well as ODC and SSAT mRNA levels were significantly higher (p < 0.05; Mann Whitney test) in colorectal carcinoma samples expressing K-ras mutation as compared to non mutated Kras samples. Mutated K-ras tissues also showed significantly higher (p < 0.05) polyamine levels, as well as ODC and SSAT activities, as compared to non mutated K-ras tissues ( Table 3).
The gene expression, the polyamine levels and the enzymatic activities were not associated with age, sex, tumour site, histological differentiation, and Dukes' stage (data not shown).

Discussion
Several studies have described increased or decreased levels of PPARγ expression in different human tumour tissue as compared to normal mucosa [19][20][21]. Hence, taking into consideration the diversity of human cancer, the expression of PPARγ may likely be dependent on tissue specificity and/or mutational events that are prerequisite for cancer development. In our study, real time RT-PCR analysis demonstrates a significant enhancement in the expression of PPARγ in colorectal tumours as compared to normal surrounding mucosa. Few studies have investi- Representative correlations between PPARγ and SSAT mRNA expression in colorectal tissue On the other hand, it is well known that polyamine metabolism is an integral component of the mechanism of colorectal carcinogenesis [9,28]. In this study, a close and positive correlation was observed between the PPARγ and SSAT expression both in normal and neoplastic human colorectal tissues. Studies performed with colorectal tumour cell lines have shown that cells transfected with the PPARγ restored the SSAT promoter activity, and an activated PPARγ could increase SSAT expression in these cells [15]. Thus, our findings in human tissue are compatible with the in vitro studies about the transcriptional induction of PPARγ on the SSAT gene.
In the light of these data, we suggest that, in colorectal normal mucosa, PPARγ induces SSAT expression and influences polyamine availability, regulating cell proliferation and differentiation. In fact, it is known that the requirement of polyamines in cell growth is typically met by a biosynthetic pathway regulated by ODC and balanced by a polyamine catabolic pathway regulated by SSAT [29].
In colorectal tumour tissue, we observed an increase in PPARγ and SSAT expression as well as in SSAT activity.
The relationship between SSAT and cancer is only now being defined. In this respect, in vitro and in vivo studies are consistent with the notion that SSAT suppresses cell growth and tumour development [30,31]. On the contrary, a Min mouse model in which SSAT expression promotes tumorigenesis has been proposed. [32]. Furthermore, the role of SSAT in the polyamine metabolism in the cancer tissue seems to be tissue dependent. In fact, it has been observed that the increase in SSAT activity leads to polyamine pool depletion and growth inhibition in breast tumour [12,33].
On the other hand, in prostatic cancer the SSAT overexpression induces growth inhibition without determining polyamine pool depletion since SSAT activity seems to modulate ODC activity [11,34].
In this study, we report both higher polyamine levels and ODC activity in neoplastic tissue samples than in normal ones, thus confirming our previous data and those of other groups obtained in human colorectal cancer or in other types of carcinoma [35][36][37][38][39].
We have also found that while colon cancer tissue exhibits about a 13-fold increase of the ODC activity in comparison to normal tissue, SSAT activity rises by 1.5-fold only. The tumour polyamine pool seems to be maintained at higher levels by a robust increase in ODC activity not properly balanced by a severe increase in SSAT activity. Therefore, the SSAT response could be secondary to a procancerous increase in ODC activity and polyamine levels.
In this direction, we can suppose that the increase in polyamine catabolic activity, as a result of PPARγ induction without a concomitant downregulation of the polyamine biosynthetic pathway, could be insufficient for counteracting tumour development.
Our findings about PPARγ and polyamine metabolism in mutated K-ras colorectal cancer seem to follow this direc- tion. In fact, we have detected higher levels of PPARγ and SSAT expression, as well as higher SSAT activity, in mutated K-ras samples. These data suggest that in cancer tissue with high rate of proliferation and likely with enhanced polyamine levels and ODC activity [14], there is an attempt to restore a growth control via SSAT with a mechanism involving PPARγ. However, in a vitro study, it has been shown that mutated K-ras suppressed the SSAT expression via a transcriptional mechanism involving the PPARγ signalling pathway [15]. The in vitro findings may not be directly relevant to the in vivo effects and different tumour models can result in diversified functional outcome.

Conclusion
In conclusion, our data show a close relationship between PPARγ and SSAT in human colorectal cancer and this could represent an attempt to decrease polyamine levels and to reduce cell growth and tumour development. These findings support the hypothesis that pharmacological activation of PPARγ and/or induction of SSAT may represent, in future, an effective therapeutic or preventive strategy for colorectal cancer. In this direction a therapeutic or chemopreventive approach using a combination of polyamine biosynthesis inhibitor like difluoromethylornithine (DFMO), and an agent, like a non-steroidal antiinflammatory drug (NSAID) which induces catabolism, should be more efficacious than a single approach. This rationale has provided the basis of two ongoing chemoprevention studies [40].