Answers are grounded in this paper and cite their sources.
Proline oxidase is a p53-induced gene that can mediate apoptosis in lung carcinoma cells. Here, we provide evidence implicating a role for proline oxidase in renal carcinoma. We observed absent or reduced expression of proline oxidase in 8 of 12 primary renal cell carcinomas, with respect to their normal tissue counterparts. Two renal cell carcinomas, which displayed little or no expression of proline oxidase, expressed p53s that were less capable of inducing proline oxidase than p53 isolated from normal renal tissue. One of those tumor-derived p53s contained a double transition mutation at amino acid residues 125 (Ala to Thr) and 193 (Arg to His), and the other exhibited a single transition mutation at amino acid 149 (Ser to Phe). Forced up-regulation of proline oxidase induced the formation of reactive oxygen species and mediated apoptosis in the 786–0 renal cell carcinoma cell line. A proline oxidase antisense vector repressed p53-induced up-regulation of proline oxidase, release of cytochromec from mitochondria, and apoptosis in 786–0 renal carcinoma cells. Taken together, these findings support a role for proline oxidase as a downstream effector in p53-mediated apoptosis. We hypothesize that its altered expression can contribute to the development of renal carcinomas. The presence of proline oxidase in mitochondria, a primary organelle that regulates apoptosis, places this molecule in a subcellular localization that can directly influence the apoptotic pathway and thus tumorigenesis. Proline oxidase is a p53-induced gene that can mediate apoptosis in lung carcinoma cells. Here, we provide evidence implicating a role for proline oxidase in renal carcinoma. We observed absent or reduced expression of proline oxidase in 8 of 12 primary renal cell carcinomas, with respect to their normal tissue counterparts. Two renal cell carcinomas, which displayed little or no expression of proline oxidase, expressed p53s that were less capable of inducing proline oxidase than p53 isolated from normal renal tissue. One of those tumor-derived p53s contained a double transition mutation at amino acid residues 125 (Ala to Thr) and 193 (Arg to His), and the other exhibited a single transition mutation at amino acid 149 (Ser to Phe). Forced up-regulation of proline oxidase induced the formation of reactive oxygen species and mediated apoptosis in the 786–0 renal cell carcinoma cell line. A proline oxidase antisense vector repressed p53-induced up-regulation of proline oxidase, release of cytochromec from mitochondria, and apoptosis in 786–0 renal carcinoma cells. Taken together, these findings support a role for proline oxidase as a downstream effector in p53-mediated apoptosis. We hypothesize that its altered expression can contribute to the development of renal carcinomas. The presence of proline oxidase in mitochondria, a primary organelle that regulates apoptosis, places this molecule in a subcellular localization that can directly influence the apoptotic pathway and thus tumorigenesis. The product of the p53 gene plays pivotal roles in several biological processes that are important in reducing the tumorigenic potential of cells. The protein mediates cell growth arrest by controlling cell cycle checkpoints (1Debbas M. White E. Genes Dev. 1993; 7: 546-554Google Scholar), induces apoptosis (2Lin D. Shields M.T. Ullrich S.J. Apella E. Mercer W.E. Proc. Natl. Acad. Sci. 1992; 89: 9210-9214Google Scholar), and can influence the differentiation state of cells (3Aloni-Grinstein R. Zanbar I. Alboum I. Goldfinger N. Rotter V. Oncogene. 1993; 8: 3297-3305Google Scholar). Many studies have implicated p53 in genomic stability, surveillance of DNA damage, and DNA repair, which are related to its control over cell growth (4Lane D.P. Nature. 1992; 358: 15-16Google Scholar, 5Kastan M.B. Onyekwere O. Sidransky D. Vogelstein B. Craig R. Cancer Res. 1991; 51: 6304-6311Google Scholar, 6Zhan Q. Carrier F. Fournace A.J. Mol. Cell. Biol. 1993; 13: 4242-4250Google Scholar, 7Merritt A.J. Potten C.S. Kemp C.J. Hickman J.A. Balmain A. Lane D.P. Hall P.A. Cancer Res. 1994; 54: 614-617Google Scholar, 8Levine A.J. Cell. 1997; 88: 323-331Google Scholar, 9Ko L.J. Prives C. Genes Dev. 1996; 10: 1054-1072Google Scholar). A series of p53-induced genes has been identified by a serial analysis of gene expression in a colon carcinoma cell line; some of those genes are predicted to encode proteins that could generate or respond to oxidative stress (10Polyak K. Xia Y. Zweier J.L. Kinzler K.W. Vogelstein B. Nature. 1997; 389: 300-305Google Scholar). We performed a comparative gene expression analysis between p53-sensitive and -resistant cells, which also revealed the close association of redox enzymes with p53-mediated apoptosis (11Maxwell S.A. Davis G.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 24: 13009-13014Google Scholar). One of those redox enzymes, proline oxidase, and its product, pyrroline-5-carboxylate (P5C), 1The abbreviations used are: P5C, pyrroline-5-carboxylate; GFP, green fluorescence protein; RT, reverse transcription; DIC, 2,6-dicholoroindophenol; POX, proline oxidase 1The abbreviations used are: P5C, pyrroline-5-carboxylate; GFP, green fluorescence protein; RT, reverse transcription; DIC, 2,6-dicholoroindophenol; POX, proline oxidase exhibited the ability to inhibit cell growth and to induce marked apoptosis in a lung carcinoma cell line, implicating a role for the proline/P5C cycle in p53-mediated growth inhibition and apoptosis (11Maxwell S.A. Davis G.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 24: 13009-13014Google Scholar). Our work was supported by a later study showing that proline oxidase can generate reactive oxygen species (12Donald S.P. Sun X.Y. Hu C.A. Yu J. Mei J.M. Valle D. Phang J.M. Cancer Res. 2001; 61: 1810-1815Google Scholar), which can initiate apoptosis by directly acting on the mitochondrial permeability core complex and effecting the mitochondrial permeability transition (13Li P.F. Dietz R. von Harsdorf R. EMBO J. 1999; 18: 6027-6036Google Scholar, 14Jacotot E. Costantini P. Laboureau E. Zamzami N. Susin S.A. Kroemer G. Ann. N. Y. Acad. Sci. 1999; 887: 18-30Google Scholar, 15Lee H.C. Wei Y.H. J. Biomed. Sci. 2000; 7: 2-15Google Scholar). Here, we report that the expression of proline oxidase was absent or reduced in a significant number of renal cell carcinomas, which was associated with mutant p53 in two cases. A correlation of reduced proline oxidase expression with mutated p53 in two cases of renal carcinoma and the ability of proline oxidase to induce apoptosis in renal carcinoma cells functionally implicate the enzyme as a downstream player in p53-mediated apoptosis. The p53 monoclonal antibody, Bp53–12, was purchased from Santa Cruz Biotechnology, the M30 antibody specific for the caspase-cleaved form of cytokeratin 18 (16) from Roche Molecular Biochemicals, a monoclonal antibody for cytochromec (C-8) from Santa Cruz Biotechnology, and an actin monoclonal antibody (N350) from Amersham Biosciences. A proline oxidase antibody (T338) was generated in rabbits against a synthetic peptide composed of amino acids 338 through 353 of the proline oxidase protein (TGQLEPLLSRFTEEEE). Kidney p53 cDNAs isolated by reverse-transcription PCR from primary kidney tissues were cloned into the pcDNA3.1 ECHO expression system (Invitrogen). The complete proline oxidase cDNA was cloned into the pAdtrack vector (17He T.C. Zhou S. da Costa L.T. Yu J. Kinzler K.W. Vogelstein B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 2509-2514Google Scholar), which coexpresses GFP as a normalization control for transfection efficiency. The pAdtrack-proline oxidase vector was used to generate recombinant adenovirus using a simplified procedure as previously described (17He T.C. Zhou S. da Costa L.T. Yu J. Kinzler K.W. Vogelstein B. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 2509-2514Google Scholar). A proline oxidase antisense vector was constructed by subcloning the proline oxidase cDNA encoding amino acids 186–516 in the pEGFPC2 expression vector (Clontech) in the antisense orientation. The proline oxidase and a wild type p53 recombinant adenovirus were amplified in 293 cells and virus-containing medium used to infect kidney cells. The 786–0 renal carcinoma and the H1299 nonsmall cell lung carcinoma cell lines were obtained from the American Tissue Type Collection and propagated in Dulbecco's modified Eagle's medium containing 10% fetal calf serum. To conduct transfections of renal and lung carcinoma cells, a T75 flask of confluent cells was incubated for 2 h at 37 °C in 3 ml of Optimem medium (Invitrogen) containing 4 μg of plasmid DNA and 3 μl of LipofectAMINE-2000 (Invitrogen). The transfection medium was then replaced with Dulbecco's modified Eagle's medium and the cells incubated for 24–36 h to allow expression of the transfected gene. To normalize for transfection efficiency, pAdtrack, which expresses green fluorescence protein, was included in the transfection assay where appropriate. Western blotting was performed by using the SuperSignal immunodetection system (Pierce Chemical), as previously described (18Maxwell S.A. Davis G.E. Apoptosis. 2000; 5: 277-290Google Scholar). Proline oxidase activity was detected by the reduction of the colorimetric indicator 2,6-dicholoroindophenol (an electron acceptor) as a result of oxidation of proline in extracts of isolated mitochondria, as previously described (19Sakuraba H. Takamatsu Y. Satomura T. Kawakami R. Oshima T. Appl. Environ. Microbiol. 2001; 67: 1470-1475Google Scholar). Mitochondria were isolated by homogenization of cells in a 0.25 m sucrose buffer (20Phang J.M. Curr. Top. Cell Regul. 1985; 25: 91-132Google Scholar). After microcentrifugation of whole-cell lysates, the mitochondria-enriched pellet was washed twice in sucrose buffer. Flow cytometry to quantitate apoptosis was performed on ethanol-fixed and permeabilized cells that were stained with propidium iodide according to a previously published protocol (11Maxwell S.A. Davis G.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 24: 13009-13014Google Scholar). Apoptotic cells were quantitated in the subG1 DNA content portion of the flow cytometric scans. Reactive oxygen species were detected by the oxidation of dihydroethidium in the cytoplasm of viable cells. The oxidized compound then migrated to the nucleus as ethidium bromide and emitted a red-orange fluorescence under ultraviolet light when bound to DNA (21Huang C. Zhang Z. Ding M. Li J. Leonard S.S. Shen H.M. Butterworth L. Lu Y. Costa M. Rojanasakul Y. Castranova V. Vallyathan V. Shi X. J. Biol. Chem. 2000; 275: 32516-32522Google Scholar). Early apoptosis was detected by immunohistochemistry using a monoclonal antibody (M30) specific for the caspase-cleaved form of cytokeratin 18 (16, 22), according to the manufacturer's specifications (Roche Molecular Biochemicals). Apoptotic cells were also identified by morphological criteria that included cell blebbing, fragmented and shrunken nuclei, and apoptotic bodies. Paired normal/tumor cDNA blot arrays were purchased from Clontech. The cDNA blot was hybridized with a proline oxidase cDNA probe according to the manufacturer's specifications and then rehybridized with an ubiquitin probe for the purpose of normalizing the expression data. Paired normal and carcinoma cDNAs generated from kidney tissues were purchased from Clontech. RT-PCR on nonsmall cell lung and renal carcinoma cell lines was performed by isolating total RNA using the TRI reagent extraction solution (Molecular Research Center, Inc.) and generating cDNA from the RNA by using the RT-for-PCR system (Clontech). RT-PCR to analyze p53, proline oxidase, Bax, p21waf1/cip1, and glyceraldehyde-3-phosphate dehydrogenase expression was performed as previously described (11Maxwell S.A. Davis G.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 24: 13009-13014Google Scholar). Ribosomal protein S9 primers for normalization of PCR were purchased from Clontech. A commercial array containing cDNA amplified from paired normal and tumor mRNA from 12 patients presenting with renal carcinoma was hybridized with a 32P-labeled proline oxidase cDNA probe (Fig. 1, upper panel). The cDNA was normalized by rehybridizing the blot with a ubiquitin housekeeping cDNA probe (Fig. 1, lower panel). The cDNA spotted on the membrane has been shown to correlate closely with the expression of a number of different known mRNAs in the tissue (Clontech). As shown in Fig. 1, five kidney tumors (POX panel, T row,lanes 3, 4, 5, 7,10) appeared to exhibit little or no expression of proline oxidase mRNA as compared with the matched normal tissue (N row), even though similar amounts of normal and tumor cDNA were present on the blot as shown by rehybridization of the blot with a ubiquitin housekeeping cDNA probe (Fig. 1, Ubiq.). When taking the ubiquitin normalization blot into consideration, it appeared that several other renal cell carcinomas exhibited reduced expression of proline oxidase mRNA (POX panel, T row,lanes 1, 8, 9). Reduced or absent proline oxidase expression thus occurred in two-thirds of the renal cell carcinomas that we examined. This tumor array data suggested that altered expression of proline oxidase is frequently associated with renal cell carcinomas. However, to validate the differential expression of proline oxidase between normal and tumor tissues that we observed in the tumor array cDNA blot, the cDNA derived from kidney 4 and kidney 7 normal and tumor tissues was subjected to PCR. As shown in Fig. 2, the expected 270-bp-size proline oxidase DNA product was generated from normal kidney tissue (N lanes) cDNA, whereas little or no product was derived from tumor tissue cDNA (T lanes), confirming the array data showing that kidney tumors 4 and 7 lacked apparent expression of proline oxidase. In both renal cell carcinomas 4 and 7, the reduced expression of proline oxidase was very specific, because transcripts for p53, p21waf1/cip1, and Bax were expressed at similar levels in both normal and tumor tissues, relative to the ribosomal S9 protein that was used as a normalization control. From an analysis of the classification and staging of these renal carcinomas, it appeared that reduced expression of proline oxidase was more prevalent in clear cell-type renal cell carcinomas at stages II to IV (Table I; tumors 1,7, 8, 9, 10). However, two clear cell-type carcinomas did display normal levels of proline oxidase (Table I; tumors 6,11). No differences in expression of proline oxidase were detected by the tumor array blotting method in three prostate carcinomas relative to their normal tissue counterparts (data not shown).Figure 2RT-PCR confirmed altered expression of proline oxidase in renal carcinomas. RT-PCR was performed on two renal cell carcinomas (4 and 7) that displayed reduced or absent proline oxidase expression in the tumor array analysis shown in Fig. 1. Ribosomal protein S9 was employed as a normalization control. The expression of p53 and two of its target genes (Bax and p21) was also investigated in the same experiment.View Large Image Figure ViewerDownload (PPT)Table IRenal cell carcinoma and proline oxidase expressionRenal cell carcinomaClassification/StageProline oxidase expression1Clear cell/grade II-III++2Poorly differentiated/grade III-IV++++3Unclassified4Grade III-IV5Unclassified6Clear cell+++7Clear cell/grade cell/grade cell/grade cell/grade cell/grade proline oxidase expression was the different kidney tumors by normalizing with the ubiquitin control probe on the tumor array blot shown in Fig. 1. of proline oxidase detected by the array was confirmed in tumors 4 and 7 by in a proline oxidase expression was the different kidney tumors by normalizing with the ubiquitin control probe on the tumor array blot shown in Fig. 1. of proline oxidase detected by the array was confirmed in tumors 4 and 7 by To the biological of proline oxidase in renal cell the 786–0 renal carcinoma cell was transfected with the proline oxidase expression vector to proline oxidase. Proline oxidase expression was by Western and a colorimetric assay to oxidation of proline in cell As shown in Fig. 3 3, little or no expression of proline oxidase mRNA was detected in 786–0 cells transfected with the expression whereas levels of proline oxidase transcripts were detected in cells transfected with the vector of proline oxidase protein was observed in 786–0 cells h transfection with the vector (Fig. 3 or with two different of recombinant proline oxidase adenovirus (Fig. 3 and No proline oxidase protein was detected in 786–0 cells with a recombinant GFP adenovirus To proline oxidase we on a method the reduction of the colorimetric indicator 2,6-dicholoroindophenol an electron as a result of oxidation of proline in extracts of isolated (19Sakuraba H. Takamatsu Y. Satomura T. Kawakami R. Oshima T. Appl. Environ. Microbiol. 2001; 67: 1470-1475Google Scholar). of in a in at over As shown in Fig. 3 the reduction of occurred to a in mitochondrial extracts from cells than in mitochondrial extracts obtained from cells. of transfected cells were used as a for the Forced of p53 in 786–0 cells by with the type p53 expression vector or with the recombinant p53 adenovirus induced apoptosis (Fig. and the proline oxidase protein of p53 was observed by Western blotting in 786–0 cells with p53 in to the apparent of p53 protein in cells whole-cell lower panel). of the induced proline oxidase protein was observed in the (Fig. 4 with little or no enzyme in the cytoplasm panel, Forced up-regulation of proline oxidase by transfection with in significant in apoptosis in 786–0 cells, as by the in the subG1 in flow cytometric relative to those in cells of proline oxidase induces apoptosis and reactive oxygen species in 786–0 renal carcinoma cells. p53-mediated of proline oxidase in 786–0 renal carcinoma cells. 786–0 cells were or with GFP or p53 recombinant adenovirus Mitochondria were isolated h and proteins in the were in mitochondrial and were in buffer and for proline oxidase expression in Western by using the proline oxidase extracts were also by 786–0 cells with recombinant GFP and p53 in buffer. The extracts were for proline oxidase p53, or actin a protein 786–0 cells were transfected with or pAdtrack-proline oxidase. h apoptotic cells were quantitated by flow cytometry of ethanol-fixed and propidium cells. of reactive oxygen species in and proline 786–0 cells. were with GFP or proline oxidase recombinant adenovirus for 12 h and then incubated with which when oxidized by reactive oxygen to DNA as an fluorescence under ultraviolet of reactive oxygen was then quantitated by flow cytometry of GFP and ethidium Proline cells more than cells, the of reactive oxygen a proline cell h showing and that is of the later of Large Image Figure ViewerDownload Proline oxidase has been shown to induce the formation of reactive oxygen species (12Donald S.P. Sun X.Y. Hu C.A. Yu J. Mei J.M. Valle D. Phang J.M. Cancer Res. 2001; 61: 1810-1815Google Scholar). in the cytoplasm when oxidized by reactive DNA and an fluorescence under As shown in Fig. 4 proline 786–0 cells with dihydroethidium more fluorescence than transfected cells, that proline oxidase induced the of reactive oxygen Fig. 4 the and fragmented of the stages of apoptosis in 786–0 cells. of cytokeratin 18 is an associated with the stages of apoptosis C. Oshima G. J. Cell Biol. 1997; Scholar). antibody (M30) specific for an that is in caspase-cleaved of cytokeratin 18 can used as a of apoptosis V. T. G. B. P. B. M. H. B. J. 1991; Scholar). that 786–0 cells both GFP and proline oxidase were reactive with the primary M30 monoclonal antibody, as detected by the fluorescence emitted by the (Fig. or no fluorescence was detected in transfected cells (Fig. panel). A role for proline oxidase in p53-mediated apoptosis was suggested from proline oxidase antisense expression Forced up-regulation of wild type p53 in 786–0 cells by transfection with the expression vector induced apoptosis h transfection of these cells with wild type p53 and a proline oxidase antisense vector in less apoptosis (Fig. and expression of proline oxidase protein by to (Fig. 2 with the proline oxidase antisense vector p53-mediated release of from (Fig. 3 Our published work (11Maxwell S.A. Davis G.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 24: 13009-13014Google and data that proline oxidase is a p53-induced gene expression is frequently reduced or absent in renal cell carcinomas. We the altered expression of proline oxidase in kidney tumors associated with mutant p53 To the of kidney tumor-derived p53, we cloned the p53 cDNAs from both normal and tumor tissue of 4 and p53 cDNAs were cloned into the ECHO expression system to were capable of inducing proline oxidase in H1299 lung carcinoma cells. As shown in Fig. 7 Western blotting that p53 cDNAs isolated from normal and tumor kidney tissues were expressed in H1299 cells PCR revealed that p53 isolated from normal kidney 4 induced the expression of proline oxidase in H1299 cells (Fig. 7 lower with of the p53s from tumors 4 and 7 also in the of proline oxidase to lower levels than that induced by the normal kidney p53 (Fig. 7 8 with Western blotting that the levels of expression of proline oxidase protein closely with the expression of mRNA detected by RT-PCR (Fig. 7 both p53 protein and mRNA levels were between the normal and tumor cells and proline oxidase mRNA and protein were by normal and tumor-derived we that the tumor-derived p53s were less at the proline oxidase gene. of the p53 cDNAs was thus performed in an to a for the differential of proline oxidase between these normal and tumor-derived The p53 isolated from renal cell carcinoma 4 contained a double transition mutation at amino acid residues 125 (Ala to Thr) and 193 (Arg to His), and that cloned from tumor 7 exhibited a single transition mutation at amino acid 149 (Ser to Phe). We that proline oxidase is a downstream in p53-mediated apoptosis in renal carcinomas, on a number of findings in this up-regulation of p53 induced the expression of the proline oxidase gene. a proline oxidase antisense expression vector p53-mediated apoptosis in a renal carcinoma cell line. two renal tumor-derived p53 were less in inducing the proline oxidase gene than wild type p53 isolated from normal renal tissue. expression of proline oxidase induced apoptosis in a renal carcinoma cell line. proline oxidase can generate reactive oxygen which mediate apoptosis by mitochondrial membrane potential by acting on the mitochondrial membrane permeability core complex (13Li P.F. Dietz R. von Harsdorf R. EMBO J. 1999; 18: 6027-6036Google Scholar, 14Jacotot E. Costantini P. Laboureau E. Zamzami N. Susin S.A. Kroemer G. Ann. N. Y. Acad. Sci. 1999; 887: 18-30Google Scholar, 15Lee H.C. Wei Y.H. J. Biomed. Sci. 2000; 7: 2-15Google Scholar). the by which proline oxidase mediates apoptosis is not the enzyme apoptosis through generated from the proline redox Proline oxidase the of proline to and into the mitochondrial electron with an (20Phang J.M. Curr. Top. Cell Regul. 1985; 25: 91-132Google Scholar). The of by proline can generate two is then the where it is to proline by the oxidation The oxidation state has been shown to important in the of cell growth N. D. K. J. Res. 1999; and apoptosis M. M. L. V. L. P. Res. 2001; Scholar). proline oxidase can generate reactive oxygen species (12Donald S.P. Sun X.Y. Hu C.A. Yu J. Mei J.M. Valle D. Phang J.M. Cancer Res. 2001; 61: 1810-1815Google Scholar), which can the mitochondrial membrane permeability transition (13Li P.F. Dietz R. von Harsdorf R. EMBO J. 1999; 18: 6027-6036Google Scholar, 14Jacotot E. Costantini P. Laboureau E. Zamzami N. Susin S.A. Kroemer G. Ann. N. Y. Acad. Sci. 1999; 887: 18-30Google Scholar, 15Lee H.C. Wei Y.H. J. Biomed. Sci. 2000; 7: 2-15Google Scholar). In data implicate a role for altered expression of proline oxidase in renal cell tumorigenesis. The data also a role for proline oxidase in p53-mediated apoptosis of renal carcinoma cells. We have also observed in proline oxidase expression in primary lung carcinomas, lung carcinoma cell and a primary data. a role for this redox enzyme in other of as
No takes yet. Share an insight, caveat, or question.
Maxwell et al. (2003) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: