99 induction of PRAP1 mRNA expression by serum deprivation and ethanol. The common mechanism of action underlying the agents that induced PRAP1 mRNA expression was DNA damage. We investigated whether PRAP1 is a genotoxic responsive gene by using gamma irradiation, a well-characterized model to study DNA-damage. Using real-time quantitative RT-PCR, we showed that PRAP1 mRNA transcript levels were induced by gamma irradiation in a time-dependent manner (Figure 3.31-B). The expression of PRAP1 mRNA was induced by fourfold at four hours after exposure to gamma irradiation. The response to DNAdamage peaked at six-fold, six hours after gamma irradiation. These results indicated that PRAP1 mRNA levels are regulated by DNA damage. 3.5.2 Transcriptional regulation of PRAP1 by genotoxic agents To further study the regulation of PRAP1 by the various stressors, we selected 5-FU and CPT for our downstream studies as they induced the highest levels of PRAP1 mRNA and are used in the treatment of colorectal cancer. Firstly, we investigated whether the regulation of PRAP1 occurs at the transcriptional level. To achieve that, we first characterized the PRAP1 promoter in HCT 116 by using various PRAP1 promoter constructs as described previously under section 3.2.2.1. The majority of the promoter constructs showed an average promoter activity of 30-fold over baseline levels (without promoter) (Figure 3.32). The longest PRAP1 promoter construct was 3900bp long and demonstrated the highest promoter activity (50-fold). In HCT 116 cells, the core promoter was also identified to be within 203 base pairs upstream of PRAP1 transcription start site (Figure 3.33), similar to our previous observation in another cell line, L8 (Figure 3.20). 100 Figure 3.31 PRAP1 is induced by genotoxic stressors A: Representative RT-PCR gel picture showing the expression of prap1 and gapdh in HCT 116 cells after treatment with the indicated stressors (5-FU (25 µM), camptothecin (20 nM), etoposide (20 µM), serum free medium or 5% ethanol for 48 hours; hydrogen peroxide (100 µM) for hour and recovered for hours; and UV (20J)). B: Representative graph showing fold induction of prap1 expression in HCT 116 cells after treatment with gamma-irradiation and recovered at 4, and 24 hours. Real-time RT-PCR was performed and relative expression of prap1 was normalized against gapdh and calculated as fold induction. Column, mean of duplicates; Bar, SEM. 101 Figure 3.32 Luciferase assay to identify the regions required for the prap1 gene promoter activity in HCT 116 cells Left panel: Illustration representing various portions of the prap1 5’ flanking region of the prap1 gene subcloned upstream of the firefly luciferase gene (pGL3Basic). The number in the box indicates the size of each prap1 fragment with respect to the transcription start site (+1). Right panel: Representative figure showing the fold induction of prap1 promoter activity of each fragment in HCT 116. For each transfection, the firefly luciferase activity was normalized with the Renilla reniformis luciferase activity by the cotransfected pRL-TK. The relative activity of each construct is expressed as a ratio to the activity of the pGL-Basic. Bar, mean of three replicates. Bar, SE. Figure 3.33 Identification of core promoter of prap1 gene Left panel: Illustration showing the deletion construct pGL (-461/-203) generated by deleting the 203 base pairs upstream of the transcription start site from the pGL(-461/0) construct. Right panel: Representative figure showing the fold induction of prap1 promoter activity of each construct in HCT 116. The promoter activity was measured as described in Figure 3.32. Column, mean of three replicates; Bar, SE. 102 To study the transcriptional regulation of PRAP1 by genotoxic agents, the longest promoter construct, pGL(-3900/0) and the core promoter construct, pGL(203/0) of PRAP1 were used. There was a two-fold increase in the core promoter activity after exposure to either 5-FU or CPT, as compared to the untreated control (Figure 3.34). This demonstrated that induction of PRAP1 was regulated at the transcriptional level. However, this two-fold increase in the transcriptional activity does not seem to fully account for the marked induction of PRAP1 mRNA by the genotoxic agents as shown in Figure 3.35, suggesting the possible involvement of other transcriptional regulatory mechanisms such as cis- or trans-regulatory elements and/or stability of mRNA. 3.5.3 Regulation of PRAP1 protein by genotoxic agents In order to study the intracellular level of PRAP1, which is a highly secreted protein (Zhang, Wong et al. 2003), we used Brefeldin A (BFA), a lactone antibiotic that inhibits the protein transport from the endoplasmic reticulum (ER) to the Golgi apparatus, to trap PRAP1 in the ER (Pelham 1991; Klausner, Donaldson et al. 1992). After exposing HCT 116 cells to 5-FU and CPT for 48 hours, PRAP1 was trapped in the ER with BFA for hours before harvesting. As shown in Figure 3.36, PRAP1 protein was highly induced by both 5-FU and CPT at various dosages which corresponded to the increase observed at the mRNA level. 5-FU induced a higher level of PRAP1 protein expression as compared to that of the CPT at all doses. The induction of PRAP1 by CPT was dose-dependent whereas the induction of PRAP1 by 5-FU was saturated at the lowest dosage used. Our results indicated genotoxic stress can induce the expression of PRAP1 at both mRNA and protein levels. 103 Figure 3.34 Induction of prap1 promoter activity by 5-FU and CPT The longest promoter construct (pGL-3900) and core promoter construct (pGL203) were cotransfected with pRL-TK into HCT 116 cells for 24 hours. 5fluorouracil, 25µM of 5-FU (A) and 20nM of camptothecin, CPT (B) was added to the cells for another 24 hours. The firefly luciferase activity was normalized with the Renilla reniformis luciferase activity. Promoter activity of each construct was expressed as fold induction relative to that of the pGL3-Basic. Column, mean of three replicates; Bar, SE. 104 Figure 3.35 Induction of PRAP1 by 5-FU and CPT at mRNA level Representative RT-PCR gel picture showing the mRNA expression of prap1 and gapdh in the cells from Figure 3.34. Figure 3.36 PRAP1 was induced at protein level by 5-FU and CPT Representative western blot of PRAP1 and GAPDH in HCT116 cells after treatment with the indicated doses of either 5-fluorouracil (5-FU) or camptothecin, (CPT) or untreated (control) for 48 hours. Brefeldin A was added to trap the PRAP1 protein in endoplasmic reticulum for hours before harvesting. Cytoplasmic protein (100 µg) was subjected to western blot analysis with a specific anti-PRAP1 antibody. The membrane was reprobed with anti-GAPDH antibody to confirm equal loading. 105 3.5.4 Dose- and time-dependent regulation of PRAP1 To study the dose and time response of PRAP1 mRNA level by 5-FU and CPT, HCT 116 cells were treated with different doses of 5-FU (2.5 and 25 µM) and CPT (20 and 250nM) for days. Cells were harvested for RT-PCR and western blot analysis at 24, 48 and 72 hours. The expression of PRAP1 mRNA was induced as early as 24 hours and sustained at 72 hours when treated with both low and high doses of 5-FU (Figure 3.37). In the presence of low dose of CPT, PRAP1 mRNA was induced at 24 hours and peaked at 48 hours, whereas with high dose of CPT, the expression of PRAP1 mRNA peaked at 24 hours. The time-course regulation of PRAP1 expression at protein level was studied over 72 hours using high doses of 5-FU or CPT. The induction of PRAP1 protein by 5-FU occurred as early as 24 hours and peaked at 48 hours (Figure 3.38), whereas the induction of PRAP1 by CPT peaked at 24 hours. This temporal regulation of PRAP1 by genotoxic stress suggests that it has a specific temporal function in the cellular response to genotoxic stress. 3.6 Wild-type-p53-dependent induction of PRAP1 3.6.1 Genotoxic agents failed to induce PRAP1 in p53-/- cells As p53 plays an important role in the response to DNA damage (Sionov and Haupt 1999), we investigated whether p53 is a regulator of PRAP1 expression upon exposure to DNA damage agents. We used two cell lines derived from the parental HCT 116 cells, one with p53 knockout (p53-/-) and one with p21 knockout (p21-/-). PRAP1 mRNA was induced in both HCT 116 wild-type and p21-/- cell lines, but not in p53-/- 106 Figure 3.37 Early upregulation of PRAP1 in a dose- and time-dependent manner Representative RT-PCR gel picture of prap1 and gapdh in HCT 116 cells after treatment with 5-FU and CPT at the indicated doses for 24, 48 and 72 hours. Figure 3.38 Early upregulation of PRAP1 protein by 5-FU and CPT Representative western blot of PRAP1 and GAPDH in HCT116 cells after treatment with 5-FU (25µM) and CPT (20nM) for 24, 48 and 72 hours. Brefeldin A was added to trap the PRAP1 protein in endoplasmic reticulum for hours before each harvesting. Cytoplasmic protein (100 µg) was subjected to immunoblot analysis with a specific anti-PRAP1 antibody. The membrane was reprobed with anti-GAPDH antibody to confirm equal loading. 107 cells at both low and high doses of 5-FU and CPT (Figure 3.39). These results suggested that induction of PRAP1 by DNA damaging agents was dependent on p53. 3.6.2 Restoration of PRAP1 induction by reintroduction of wild-type p53 in p53-/- cells To confirm that the induction of PRAP1 was dependent on p53, we reintroduced the wild type p53, pCMV-p53 (WT) or the DNA-binding deficient mutant p53, pCMV-p53mt153 (Mut) into p53-/- cells. Our results showed that, the PRAP1 mRNA was detectable in WT-transfected p53-/- cells, but not in Mut or empty vector-transfected p53-/- cells, suggesting that PRAP1 may be a direct target gene of p53. In the presence of 5-FU or CPT, the expression of PRAP1 mRNA was further induced in the WT-transfected p53-/- cells, but not in the Mut or vector-transfected cells (Figure 3.40). This indicated that the induction of PRAP1 requires the expression of wild-type p53. While p53 is best characterized as a transcription factor, the data presented here does not distinguish between the following two possible roles of p53 in the induction of PRAP1 expression following exposure to DNA damaging agents. Firstly, PRAP1 may be a direct or indirect transcriptional target of p53. Secondly, p53 which is a DNA binding protein, may displace an unknown repressor from the PRAP1 promoter. 3.6.3 Genotoxic agents failed to induce PRAP1 in Hep 3B and HT 29 cells To further validate that wild-type p53 is required for the induction of PRAP1 expression by genotoxic agents, we used two cell lines with inherent p53 defects: HT 29, a cell line with transcriptional mutant p53 (HT 29) and Hep 3B, a 108 Figure 3.39 Induction of prap1 by 5-FU and CPT is dependent on p53 Representative RT-PCR gel picture of prap1 and gapdh in HCT116 cells and its two derivatives, p53-/- and p21-/- after treatment with the indicated doses of 5-FU and CPT for 24 hours. Figure 3.40 Reintroduction of wild-type p53 rescues the induction of prap1 by 5-FU and CPT in p53-/- cells Representative RT-PCR gel picture of prap1 and gapdh in p53-/- cells after transfection with either empty vector control (V) or wild-type p53 (WT) or mutant p53 (Mut), followed by treatment with either 5-FU (25µM) or CPT (20nM) for 24 hours. 122 Figure 3.53 Repression of PRAP1 expression results in morphological changes Representative phase-contrast images showing the morphology of HCT116 cells after transfection with control siRNA or PRAP1 siRNA for 48 hours, and followed by addition of 5-FU for 24 hours (320X Magnification). 123 Figure 3.54 Repression of PRAP1 expression abrogates the S-phase arrest induced by 5-FU Representative histograms showing the cell cycle distribution of HCT116 cells after transfection with control siRNA or PRAP1 siRNA and for 48 hours, and followed by the addition of 5-FU for 24 hours. Total cells were harvested and stained with propidium iodide for DNA content analysis by flow cytometry. Box indicates population of cells at S-phase. 124 similar results to that of the untransfected control, indicating the effects observed are specific to PRAP1siRNA treatment. Our result showed that treatment of cells with 5-FU induced an increase in cyclin E1 and cyclin A1 protein levels, reflecting an arrest at S-phase (Figure 3.55). Repression of PRAP1 induction by 5FU caused a significant reduction in cyclin A1 expression, whereas cyclin E1 expression level remained unchanged. There was also an accompanied loss of CDK but not CDK (Figure 3.56). These results suggested that the abrogation of S-phase arrest by repression of PRAP1 may be mediated by disruption of the cyclin A1/CDK complex. 3.9.3 Effect of PRAP1 knockdown on cell cycle checkpoint proteins It has been shown that 5-FU acts through p21 to arrest the cells for DNA repair (el-Deiry, Kern et al. 1992). The function of cell cycle arrest proteins is dependent on their localization in the cellular compartments. To investigate whether PRAP1 interferes with cell cycle checkpoint proteins and their functions, we harvested the cells after treatment and subjected them to cell fractionation. As shown in Figure 3.57, there were no obvious changes in the level of p21 or p27 in the total cell lysate after 5-FU treatment. The effects on cell cycle checkpoint proteins were more evident after fractionation. There was an increase in p21 in both cytoplasmic and nuclear extracts of cells treated with 5-FU as compared to untreated control. For p27, the increase was more evident in the nuclear extract when compared to that of the untreated cells. Repression of PRAP1 did not interfere with either the level or the localization of p27. However, in cells with PRAP1 inhibition, there was a further increase in the cytoplasmic p21 level with a corresponding decrease in nuclear p21. This relocation of p21 from nucleus to cytoplasm may inhibit the function of p21 in causing cell cycle arrest. These 125 Figure 3.55 Repression of PRAP1 expression reduces cyclin A1 Representative western blot of cyclin E1, cyclin A1, p53 and B-ACTIN in HCT 116 cells after transfection with transfection reagent alone or siRNA (control siRNA and PRAP1 siRNA) for 48 hours, and followed by the addition of 5-FU for 24 hours. Lane 1: control; 2: transfection reagent control; 3: PRAP1 siRNA; 4: control siRNA. Figure 3.56 PRAP1 knockdown reduces CDK2 Representative western blot of CDK 4, CDK 2, p53 and B-ACTIN in HCT 116 cells after transfection with control siRNA or PRAP1 siRNA for 48 hours, and followed by the addition of 5-FU for 24 hours. 126 Figure 3.57 Repression of PRAP1 expression affects p21 localization Representative western blot of p21, p27, GAPDH and B-ACTIN in HCT 116 cells after transfection with control siRNA (C) or PRAP1 siRNA (P) for 48 hours, and followed by the addition of 5-FU for 24 hours. The cells were harvested for total protein extraction (Total) or subjected to fractionation for cytoplasmic (Cytosol) and nuclear extract (Nucleus). 127 results indicated that PRAP1 may mediate its effect on cell cycle profile by retaining p21 in the cytoplasm. 3.9.4 Effect of PRAP1 knockdown on p53 level The level of p53 was reported to regulate cell fate. Low level of p53 mainly induces cell cycle arrest while apoptosis is activated with high level of p53 (Vousden and Lu 2002). As shown in the immunoblot analysis in Figure 3.55 and Figure 3.56, no significant change was observed in the level of p53 in the sample with PRAP1 repression as compared to that of the control siRNA treated sample. This demonstrated that the mechanism by which PRAP1 knockdown modulates cell fate was not by interfering with the protein level of p53, suggesting PRAP1 protein may mediate its function downstream of p53. However, our data presented here does not rule out the possibility that PRAP1 may exert its effect at the level of post-translational modification of p53 such as phosphorylation. 3.9.5 PRAP1 expression was up-regulated in cells arrested at S-phase To further validate the relationship of PRAP1 and cell cycle, we examined the regulation of PRAP1 in cells synchronized at S-phase by thymidine or at G2/M phase by taxol or nocodazol. PRAP1 mRNA expression was highly induced in cells arrested at S-phase by single or double-thymidine block assay, but not in cells arrested at G2/M phase by taxol or nocodazole (Figure 3.58). These results strongly demonstrated that PRAP1 may be involved in cell division especially at S-phase, consistent with our observation with 5-FU. 128 Figure 3.58 PRAP1 was upregulated in S-phase arrested cells Representative RT-PCR agarose gel picture of prap1 and gapdh in HCT 116 cells after treatment with either 400ng/ml of nocodazole (N) or 1µM taxol (TAX) for 16 hours to arrest cells at G2/M phase, or 5mM thymidine for 16 hours to arrest at G1/S phase. Alternatively, cells were arrested at G1/S using double thymine (2mM) block. 3.9.6 PRAP1 inhibition in double-thymidine block assay In order to further confirm the abrogation of S-phase arrest by the repression of PRAP1 expression, we employed the classical S-phase arrest model, double-thymidine block assay. Cells were transfected with PRAP1 siRNA for 24 hours before subjecting them to double-thymidine block treatment. There was a marked loss of the proportion of cells in S-phase following PRAP1 knockdown as compared to that of the control siRNA (Figure 3.59-A). Consistent with our hypothesis, this significant abrogation of S-phase arrest (Figure 3.59-B) was accompanied by a significant increase in cell death (Figure 3.59-C). These data supported the hypothesis that repression of PRAP1 induction enhanced the apoptosis caused by 5-FU by inhibiting S-phase arrest. 129 3.9.7 PRAP1 overexpression and cell cycle To access whether PRAP1 alone is sufficient to interfere with cell cycle profile, we cloned PRAP1 into a mammalian expression vector, pcDNA 3.1 (+). HCT 116 cells were transfected with pcDNA-PRAP1 plasmid or vector control for 72 hours. Cells were harvested, fixed and stained with propidium iodide. Flow cytometry was performed for cell cycle analysis. No obvious changes in cell cycle profile were observed in cells with PRAP1 overexpression compared to that of the vector control (Figure 3.60). Together, these results indicated that PRAP1 alone is not sufficient to induce cell cycle arrest at S-phase, but it is necessary in inducing S-phase arrest in the context of 5-FU treatment. 3.9.8 PRAP1 knockdown in p53-/- cells As the cytotoxic effect of 5-FU is dependent on p53 (Johnson, Wang et al. 1997) and p53 is inactivated in most colorectal tumors (Hollstein, Rice et al. 1994), we investigated whether repression of PRAP1 induction was able to exert its effect in the absence of p53. As shown in the histograms of Figure 3.61, Sphase arrest was detected with low dose of 5-FU (1 and µM) in p53-/- cells. This indicated that 5-FU was able to induce a p53-independent S-phase arrest at low doses. Repression of PRAP1 in p53-/- cells treated with 5-FU did not result in any significant reduction in the S-phase population. This indicated that the abrogation of S-phase arrest by a suppression of PRAP1 occurs in a p53-dependent manner. Consistently, without PRAP1 exerting its effect in abrogating the S-phase, there was no enhancement in the cytotoxicity of 5-FU observed (Figure 3.62). 130 A Non-transfected Control siRNA1 PRAP1 siRNA1 50 B % of S-phase 40 30 * 20 10 -1 A N R si P1 ro A PR U co nt nt ns ls fe iR ct N A ed 25 * C % of sub-G1 20 15 10 -1 A N R si PR A P1 ro nt co U nt ns ls fe iR ct N A ed Figure 3.59 Abrogation of the S-phase arrest in double-thymidine block A: Representative histograms of cell cycle profile of HCT116 cells after transfection with control siRNA or PRAP1 siRNA1 for 48 hours, and followed by double-thymidine block. Box indicates the population of cells at S-phase. B: The statistical analysis of S-phase reduction by PRAP1 knockdown in the double-thymidine block model was performed using GraphPad Prism. Column, mean of three replicates; Bars, SEM. * p[...]... cells with PRAP1 inhibition, there was a further increase in the cytoplasmic p 21 level with a corresponding decrease in nuclear p 21 This relocation of p 21 from nucleus to cytoplasm may inhibit the function of p 21 in causing cell cycle arrest These 12 5 Figure 3.55 Repression of PRAP1 expression reduces cyclin A1 Representative western blot of cyclin E1, cyclin A1, p53 and B-ACTIN in HCT 11 6 cells after... transfection with PRAP1 specific siRNA (PRAP1 siRNA1 and 2) or PRAP1 siRNA specific control siRNA (control siRNA1 and 2) for 48 hours, and followed by the addition of 5-FU for another 24 hours Figure 3 .47 Suppression of PRAP1 induction by 5-FU at protein level Representative western blot of PRAP1 and GAPDH in HCT 11 6 after transfection with PRAP1 specific siRNA (PRAP1 siRNA1 and 2) or PRAP1 siRNA specific... responsive elements in PRAP1 matched the consensus p53 binding site used by the algorithm by 84% and 70% (Table 3 .11 ) In comparison, the known p53-response element in p 21 promoter showed a match of 90% The positions of the two p53-responsive elements ( +13 16 and + 14 60) in PRAP1 are illustrated in Figure 3 .42 In order to determine whether the two identified p53-response elements in PRAP1 (PRAP1-p53BS) are functional,... 3B and HT 29 cells failed to induce prap1 gene expression Representative RT-PCR gel picture of prap1 and gapdh in HCT 11 6 (p53 WT) versus HT 29 (p53 mutant) and Hep G2 (p53 WT) versus Hep 3B (p53 deficient) cells after treatment with 5-FU (25µM) and CPT (20nM) for 48 hours 11 1 Figure 3 .42 Schematic diagram of PRAP1 gene Two p53 binding sites were identified in intron 1 of PRAP1 Their distribution and. .. two corresponding control siRNAs by mutating 4 bases of PRAP1 siRNA1 and 2 (control siRNA1 and control 11 4 Figure 3 .45 Predicted p53 binding elements of PRAP1 response to wild-type p53 Representative figure showing the fold induction of SV40 promoter activity of pGL3-P (basic vector), pGL3-PRAP and pGL3-p 21 cotransfected with pcDNA vector, pcDNA-p53 and pcDNA-p53Mut in p53-/- cells for 48 hours For... of protein complexes The main cell cycle protein complexes involved in G1/S-phase are cyclin D/CDK4 and cyclin E/CDK 2 complexes Both mock transfection and control siRNA showed 12 2 Figure 3.53 Repression of PRAP1 expression results in morphological changes Representative phase-contrast images showing the morphology of HCT 116 cells after transfection with control siRNA or PRAP1 siRNA for 48 hours, and. .. 11 6 cells with twenty times higher dosage of 5-FU (200μM) to induce the same level of cell death as observed in the cells with PRAP1 repression Our results showed that pre-treatment of cells with Z-VAD and specific inhibitors of caspase 3, 8 and 9 were able to block cell death induced by 5-FU in HCT 11 6 cells, indicating that 5-FU induces cell 13 4 5-FU A - + Control siRNA PRAP1 siRNA1 B % of Sub-G1... the intronic fragment from positions +11 97 to +15 34 which contains both p53-response elements of PRAP1 This fragment was cloned into a pGL3-promoter luciferase reporter vector upstream of a minimal SV -40 promoter as illustrated in Figure 3 .43 The constructed plasmid (pGL3-PRAP) was checked by restriction enzyme digestion (insert size, 337bp; Figure 3 .44 ) and verified by sequencing 11 0 Figure 3. 41 Hep... position was showed in the diagram with transcription start site of PRAP1 gene denoted as +1 Table 3 .11 Sequences of the two p53 binding sites located in PRAP1 gene The consensus sequence of p53 was showed in the second row of the table The p53 binding site located in p 21 was included as positive control for comparison Row three and four indicate the sequences of p53 binding site in PRAP1 gene as predicted... SEM *p . (p53-/-) and one with p 21 knockout (p 21- /-). PRAP1 mRNA was induced in both HCT 11 6 wild-type and p 21- /- cell lines, but not in p53-/- 10 6 Figure 3.37 Early upregulation of PRAP1 in. ( +13 16 and + 14 60) in PRAP1 are illustrated in Figure 3 .42 . In order to determine whether the two identified p53-response elements in PRAP1 (PRAP1-p53BS) are functional, we amplified the intronic. two corresponding control siRNAs by mutating 4 bases of PRAP1 siRNA1 and 2 (control siRNA1 and control 11 4 Figure 3 .45 Predicted p53 binding elements of PRAP1 response to