Circulating tumour cells as a biomarker for diagnosis and staging in pancreatic cancer.

BACKGROUND: Current diagnosis and staging of pancreatic ductal adenocarcinoma (PDAC) has important limitations and better biomarkers are needed to guide initial therapy. We investigated the performance of circulating tumour cells (CTCs) as an adjunctive biomarker at the time of disease presentation. METHODS: Venous blood (VB) was collected prospectively from 100 consecutive, pre-treatment patients with PDAC. Utilising the microfluidic NanoVelcro CTC chip, samples were evaluated for the presence and number of CTCs. KRAS mutation analysis was used to compare the CTCs with primary tumour tissue. CTC enumeration data was then evaluated as a diagnostic and staging biomarker in the setting of PDAC. RESULTS: We found 100% concordance for KRAS mutation subtype between primary tumour and CTCs in all five patients tested. Evaluation of CTCs as a diagnostic revealed the presence of CTCs in 54/72 patients with confirmed PDAC (sensitivity=75.0%, specificity=96.4%, area under the curve (AUROC)=0.867, 95% CI=0.798-0.935, and P<0.001). Furthermore, a cut-off of ⩾3 CTCs in 4 ml VB was able to discriminate between local/regional and metastatic disease (AUROC=0.885; 95% CI=0.800-0.969; and P<0.001). CONCLUSION: CTCs appear to function well as a biomarker for diagnosis and staging in PDAC.

The proximal signaling network of the BCR-ABL1 oncogene shows a modular organization.

BCR-ABL1 is a fusion tyrosine kinase, which causes multiple types of leukemia. We used an integrated proteomic approach that includes label-free quantitative protein complex and phosphorylation profiling by mass spectrometry to systematically characterize the proximal signaling network of this oncogenic kinase. The proximal BCR-ABL1 signaling network shows a modular and layered organization with an inner core of three leukemia transformation-relevant adaptor protein complexes (Grb2/Gab2/Shc1 complex, CrkI complex and Dok1/Dok2 complex). We introduced an 'interaction directionality' analysis, which annotates static protein networks with information on the directionality of phosphorylation-dependent interactions. In this analysis, the observed network structure was consistent with a step-wise phosphorylation-dependent assembly of the Grb2/Gab2/Shc1 and the Dok1/Dok2 complexes on the BCR-ABL1 core. The CrkI complex demonstrated a different directionality, which supports a candidate assembly on the Nedd9 (Hef1, CasL) scaffold. As adaptor protein family members can compensate for each other in leukemic transformation, we compared members of the Dok and Crk protein families and found both overlapping and differential binding patterns. We identified an additional level of regulation for the CrkII protein via binding to 14-3-3 proteins, which was independent from its inhibitory phosphorylation. We also identified novel components of the inner core complexes, including the kinases Pragmin (Sgk223) and Lrrk1 (Lrrk2 paralog). Pragmin was found as a component of the CrkI complex and is a potential link between BCR-ABL1/CrkI and RhoA signaling. Lrrk1 is an unusual kinase with a GTPase domain. We detected Lrrk1 as a component of the Grb2/Gab2/Shc1 complex and found that it functionally interacts with the regulator of small GTPases Arap1 (Centd2) and possibly participates in the mitogen-activated protein kinase response to cellular stresses. This modular and phosphorylation-driven interaction network provides a framework for the integration of pleiotropic signaling effects of BCR-ABL1 toward leukemic transformation.

Maximal entropy inference of oncogenicity from phosphorylation signaling.

Point mutations in the phosphorylation domain of the Bcr-Abl fusion oncogene give rise to drug resistance in chronic myelogenous leukemia patients. These mutations alter kinase-mediated signaling function and phenotypic outcome. An information theoretic analysis of the correlation of phosphoproteomic profiling and transformation potency of the oncogene in different mutants is presented. The theory seeks to predict the leukemic transformation potency from the observed signaling by constructing a distribution of maximal entropy of site-specific phosphorylation events. The theory is developed with special reference to systems biology where high throughput measurements are typical. We seek sets of phosphorylation events most contributory to predicting the phenotype by determining the constraints on the signaling system. The relevance of a constraint is measured by how much it reduces the value of the entropy from its global maximum, where all events are equally likely. Application to experimental phospho-proteomics data for kinase inhibitor-resistant mutants shows that there is one dominant constraint and that other constraints are not relevant to a similar extent. This single constraint accounts for much of the correlation of phosphorylation events with the oncogenic potency and thereby usefully predicts the trends in the phenotypic output. An additional constraint possibly accounts for biological fine structure.

Bioinformatic identification of potential autocrine signaling loops in cancers from gene expression profiles.

Many biological signaling pathways involve autocrine ligand-receptor loops; misregulation of these signaling loops can contribute to cancer phenotypes. Here we present an algorithm for detecting such loops from gene expression profiles. Our method is based on the hypothesis that for some autocrine pathways, the ligand and receptor are regulated by coupled mechanisms at the level of transcription, and thus ligand-receptor pairs comprising such a loop should have correlated mRNA expression. Using our database of experimentally known ligand-receptor signaling partners, we found examples of ligand-receptor pairs with significantly correlated expression in five cancer-based gene expression datasets. The correlated ligand-receptor pairs we identified are consistent with known autocrine signaling events in cancer cells. In addition, our algorithm predicts new autocrine signaling loops that can be verified experimentally. Chemokines were commonly members of these potential autocrine pathways. Our analysis also revealed ligand-receptor pairs with expression patterns that may indicate cellular mechanisms for preventing autocrine signaling.

Rapid gene repression triggered by interleukin-6 at the onset of monocyte differentiation.

To date, the majority of characterized extracellular ligand-induced rapid changes in gene expression involve upregulation. Hence, rapid gene repression is either less common or less well studied. To study rapid gene repression during cytokine-initiated differentiation programs, we used the mRNA subtractive hybridization technique of representational difference analysis to isolate repressed genes. Cultures of the myeloid leukemia cell line M1 were induced to terminally differentiate by treatment with interleukin-6 (IL-6). The repressed genes identified in our subtraction products include the genes encoding the growth factor receptor Flt3/Flk2/STK-1 (CD135) and the costimulatory protein CD24 [heat-stable antigen] and the c-myb oncogene. Following 4 h of IL-6 treatment, mRNA levels of these genes are decreased by 45-65% relative to controls and after 8 h by 65-80%. Lipopolysaccharide also triggers the repression of these genes. Protein synthesis inhibitors do not block the IL-6-stimulated repression of c-myb, or c-myc, mRNA, yet they do block the repression of flt3 and CD24 mRNA, demonstrating the existence of both protein synthesis-independent and -dependent mechanisms of cytokine-triggered rapid gene repression during differentiation.

The role of neutrophil-derived oxidants as second messengers in interleukin 1beta-stimulated cells.

The role of the inflammatory cytokine interleukin 1beta (IL-1beta) as potent agonist of the PMN respiratory burst signal transduction cascade has been described. We hypothesized that this phenomenon is self-limiting and that polymorphonuclear leukocyte (PMN)-derived reactive oxygen intermediates (ROI) might provide feedback regulation on the IL-1beta surface receptor (IL-1betaR)-G-protein-effector enzyme transducing tripartite complex that ultimately leads to NADPH oxidase activation. Therefore, we separately assessed either baseline or IL-1beta-induced activation of each member of the IL-1betaR-G-protein-phospholipase D (PLD) or IL-1betaR-G-protein-phospholipase C (PLC) signaling systems in the presence or absence of one of several specific ROI scavengers/antioxidants. Purified human PMN were lipopolysaccharide primed, adhered for 2 h, and stimulated with 100 ng/mL IL-1beta with or without 1% v/v dimethyl sulfoxide, 10 mM NaN3, 30 mM L-alanine, 200 U catalase, or 300 U superoxide dismutase (SOD). To validate the use of these antioxidants, the production of O2-, H2O2, hypochlorous acid, or myeloperoxidase (MPO) in the employed experimental model was confirmed in a separate set of experiments. The expression of IL-1betaR type I or II was assessed by binding with corresponding 125I-labeled monoclonal antibodies and corrected for nonspecific binding. PLD activation was assessed by measuring phosphatidyl ethanol formation in the presence of ethanol. PLC activation was determined by quantitative measurement of diacylglycerol. The level of Galpha stimulatory and inhibitory subunits was assessed by Western blotting. IL-1betaR type I expression was significantly up-regulated in the presence of catalase and SOD. PLD activation was increased by dimethyl sulfoxide and NaN3, and PLC activation was up-regulated by NaN3, L-alanine, SOD, and catalase. After 5 min of stimulation with IL-1beta, Gialpha expression was significantly down-regulated by NaN3 and SOD, whereas SOD had an up-regulating effect on the expression of Gs alpha. Increasing concentrations of externally added authentic MPO progressively down-regulated both PLD and PLC activity. Thus, PMN-derived ROI, in addition to their role as antibacterial/fungal agents, serve as second messengers in IL-1beta signal transduction, with MPO having the most ubiquitous role as a modulator of PMN second messenger pathways.

Superiority of combined CK-MB and troponin I measurements for the early risk stratification of unselected patients presenting with acute chest pain.

BACKGROUND: Recent studies have suggested that positive troponin I tests are associated with an increased risk of cardiac death during short-term follow-up. However, it is unknown if troponin I tests alone or in addition to CK-MB measurements are superior to predict unfavorable outcome during long-term follow-up. PATIENTS AND METHODS: In a prospective, double-blind study we assessed the prevalence and prognostic value of combined troponin I and CK-MB tests in an unselected cohort of patients (n = 292) admitted to the emergency department for acute chest discomfort. Patients were grouped according to the diagnosis on discharge in those with acute myocardial infarction (1), unstable angina (2), and noncardiac chest pain (3). Six months after enrollment, death rates were obtained and follow-up interviews were performed with respect to survival, recurrence of chest pain, and myocardial infarction. RESULTS: In patients with evidence of coronary heart disease, the mortality rate for abnormal troponin I and normal CK-MB levels was 5.0%. Baseline troponin I and elevated CK-MB levels were associated with a mortality rate of 4.0%. However, the mortality rate was significantly higher (11.1%) in patients presenting with elevated troponin I and CK-MB values. In patients without myocardial infarction on admission, 10.5% with positive troponin I tests died compared to 1.6% with negative tests. The mortality rate in patients without myocardial infarction was 2.7% for patients with elevated CK-MB but normal troponin I values. In patients with both markers elevated a significantly higher mortality rate (16.7%) was found, representing a 6-fold increase in the death event rate. With the additional knowledge of troponin I values, it could be demonstrated that certain cases were misclassified as having noncardiac chest pain. At least some of the latter patients with above-normal values of troponin I were retrospectively to be reclassified as unstable angina. Acute non-Q-wave myocardial infarctions were occasionally misdiagnosed as either angina pectoris or nonischemic chest pain. CONCLUSIONS: Our data suggest the superiority of combined CK-MB and troponin I measurements in clinical practice for the early risk stratification of patients presenting with acute chest pain. In nonmyocardial infarctions, both CK-MB and troponin I convey independent prognostic information with regard to fatal outcome. Troponin I tests in addition to CK-MB measurements contribute to a lower rate of misdiagnoses.

Selection of human cervical epithelial cells that possess reduced apoptotic potential to low-oxygen conditions.

Since human papillomavirus (HPV) infection is strongly associated with cervical neoplasia and tumor hypoxia has prognostic significance in human cervical carcinomas, we examined the relationship between hypoxia and apoptosis in human cervical epithelial cells expressing high-risk HPV type 16 oncoproteins. In vitro, hypoxia stimulated both p53 induction and apoptosis in primary cervical epithelial cells infected with the HPV E6 and E7 genes but not in cervical fibroblasts infected with E6 and E7. Furthermore, cell lines derived from HPV-associated human cervical squamous cell carcinomas were substantially less sensitive to apoptosis induced by hypoxia, indicating that these cell lines have acquired additional genetic alterations that reduced their apoptotic sensitivity. Although the process of long-term cell culturing resulted in selection for subpopulations of HPV oncoprotein-expressing cervical epithelial cells with diminished apoptotic potential, the exposure of cells to hypoxia greatly accelerated the selection process. These results provide evidence for the role of hypoxia-mediated selection of cells with diminished apoptotic potential in the progression of human tumors and can in part explain why cervical tumors that possess low pO2 values are more aggressive.

Endogenous PMN-derived reactive oxygen intermediates provide feedback regulation on respiratory burst signal transduction.

The role of polymorphonuclear leukocytes (PMN) in stemming systemic infection is executed mainly by the utilization of molecular O2 leading to the production of reactive oxygen intermediates (ROI). PMN-derived ROI also serve as intra- and extracellular second messengers providing both positive and negative feedback on cellular autoregulation. We investigated the effect of endogenous ROI on two signal transducing pathways: the receptor (R)-G-protein-phospholipase D (PLD) and receptor (R)-G-protein-phospholipase C pathways responsible for the subsequent interleukin-8 (IL-8)-induced PMN respiratory burst. Purified human PMN were primed with LPS adhered to plastic surfaces and stimulated with IL-8 with or without the presence of each of five different selective ROI scavengers/antioxidants: DMSO, N(a)N3, L-alanine, catalase, or superoxide dismutase. Total IL-8 surface receptor expression was assessed by 125I-IL-8 and 125I-labeled mAbs against IL-8R type A and B binding assays; PLD activation was assessed by measuring formation of phosphatidyl ethanol (PEt) in the presence of ethanol; PLC activation was measured by quantitative conversion of [32P]ATP-labeled phosphatidic acid (PA) into diacylglycerol (DAG); expression of G alpha-inhibitory subunit was assessed by SDS-PAGE and immunoblotting with polyclonal Abs against this subunit. Production of O2-, H2O2, HClO, and myeloperoxidase (MPO) in the experimental model was confirmed in a separate set of experiments. The overall impact of antioxidants on each component of the transducing tripartite complex was stimulatory; however, N(a)N3 and SOD exhibited the most ubiquitous effect with consistent up-regulation by N(a)N3 of IL-8R expression, whereas even trace amounts of externally added authentic MPO significantly down-regulated the functional activity of both effector enzymes. These results demonstrate a multiple site-specific targeting of the signal-transducing complex by endogenous PMN-derived ROI and an overall protective effect of ROI scavengers/antioxidants.

Cytokine control of PMN phagocytosis: regulatory effects of hypoxemia and hypoxemia-reoxygenation.

We investigated the effects of hypoxemia and hypoxemia-reoxygenation (H/R) on interleukin-8 (IL-8), tumor necrosis factor-alpha (TNF-alpha), or IL-1beta stimulation of whole blood polymorphonuclear leukocyte (PMN) phagocytosis and bactericidal activity. Whole blood PMN were rendered hypoxemic (venous PO2 < 15 mmHg), normoxic (venous PO2 60-80 mmHg), or reoxygenated after hypoxemia (H/ R = venous PO2 150-200 mmHg) and were incubated with IL-8, TNF-alpha, or IL-1beta before sequential addition of serum-opsonized fluorescent microspheres and fluorescein isothiocyanate-conjugated mouse anti-human CD64, CD32w, CD16, CD35, or CD11b/CD18. Concomitant two-color flow cytometric analyses were then performed measuring mean channel fluorescence and the percentage of PMN positive for phagocytosis, with simultaneous subset receptor analysis on populations of PMN that exceeded control levels of phagocytosis. During hypoxemia, whole blood PMN phagocytosis in the presence of IL-8, TNF-alpha, or IL-1beta was increased compared with normoxia. Northern blot analyses revealed an increase in steady-state mRNA levels for CD32w during hypoxemia + IL-8 and CD64 during hypoxemia + IL-1beta. During reoxygenation, both whole blood PMN phagocytosis and bactericidal activity were reduced in the presence of IL-8, TNF-alpha, or IL-1beta, and in subsets of PMN with reduced phagocytosis H/R reduced CD64, CD32w, CD16, CD35, and CD11b/CD18 expression in the presence of each cytokine. Northern blot analyses revealed that H/R reduced mRNA levels for opsonic receptors primarily for IL-1beta-stimulated PMN. These results demonstrate a direct regulatory effect of hypoxemia and H/R on whole blood PMN phagocytosis, receptor expression, and steady-state mRNA levels of both Fc(gamma) and complement receptors.

Modulation of c-Myc activity and apoptosis in vivo.

We have developed an animal tumor model system to study the effects of c-Myc activation on apoptosis induction in vivo. Tumors were generated in SCID mice from Rat-1 fibroblasts that constitutively express an inactive c-Myc-estrogen receptor fusion protein (T.D. Littlewood et al, Nucleic Acids Res., 23: 1686 -1690, 1995), which is activated in vivo by the administration of 4-hydroxytamoxifen in time release pellets. We demonstrate that activation of c-Myc results in a substantial increase in the number of apoptotic tumor cells and that this apoptosis is predominant in regions of tumor hypoxia. c-Myc-induced apoptosis of hypoxic cells is inhibited in tumors that overexpress the human Bcl-2 protein. Bcl-2, however, does not prevent p53 protein accumulation or the down-regulation of the cyclin-cdk inhibitor p27 protein following c-Myc activation by 4-hydroxytamoxifen. This result suggests that Bcl-2 does not affect c-Myc function directly but acts downstream of c-Myc to inhibit apoptosis. We propose that the ability of activated c-Myc to enhance cellular proliferation might contribute to the genesis of early neoplasms that are held in check by the alternate ability of c-Myc to induce apoptosis of cells that have outgrown their supply of oxygen or other factors associated with hypoxic regions of solid tumors. Secondary genetic lesions downstream of c-Myc that suppress the apoptotic potential of tumor cells, such as Bcl-2 overexpression, might play an important role in the malignant progression of these tumors because they would disrupt the balance between apoptosis and proliferation initiated by c-Myc deregulation.

Hypoxemia/reoxygenation down-regulates interleukin-8-stimulated bactericidal activity of polymorphonuclear neutrophil by differential regulation of CD16 and CD35 mRNA expression.

BACKGROUND: The purpose of this study was to determine the effects of hypoxemia/reoxygenation (H/R) on the regulation of interleukin-8 (IL-8)-stimulated human polymorphonuclear neutrophil (PMN) bactericidal activity. METHODS: Venous human whole blood was rendered normoxic (Pvo2 saturation 60% to 80%), hypoxemic (Pvo2 saturation, less than 15%), or H/R (Pvo2 saturation more than 97%) by dialyzing the blood against a gas mixture of N2/H2/CO2 +/- 30% O2. Two hundred microliter aliquots from each study group were incubated with IL-8 (50 ng/ml) for 45 minutes before fluorescein isothiocyanate-conjugated mouse antihuman CD16 or CD35 antibodies were added. Bactericidal activity was measured with the release of 51Cr from labeled bacteria at 1:1, 5:1, and 10:1 PMN-target ratios. Steady-state mRNA levels for CD16 and CD35 were quantified by Northern blot analyses. RESULTS: H/R reduced PMN bactericidal activity compared with hypoxemic levels for staphylococcus aureus (48 +/- 5.6 versus 27 +/- 3.3) and Escherichia coli (58 +/- 7.1 versus 33 +/- 4.2). H/R reduced the surface expression of CD16 but not CD35 (mean channel fluorescence CD16, 610 +/- 70 versus 310 +/- 30 for hypoxemia versus H/R; p < 0.01). After H/R was performed, IL-8 decreased mRNA levels for CD16 but not for CD35 compared with levels seen during hypoxemia + IL-8. CONCLUSIONS: H/R down-regulates IL-8-stimulated PMN bactericidal activity by decreasing steady-state mRNA levels and surface expression of CD16. PMN bactericidal capability after H/R + IL-8 is primarily complementary and not Fc gamma receptor dependent.

Altered oxygen tension modulates cytokine-induced signal transduction in polymorphonuclear leukocytes: regulation of the G PLC pathway.

The purpose of these studies was to examine the sensitivity of the PIP 2-PLC-transducing pathway (GPLC) and its relationship to the respiratory burst in human polymorphonuclear leukocytes (PMN) stimulated by IL-8, TNF-alpha, or IL-1 beta during sequential changes in buffer oxygen tension from normoxia (pO2 = 180-200 mm Hg), to hypoxia (pO2 < 30 mm Hg) and then reoxygenation (pO2 > 140 mm Hg). Our specific hypothesis was that altered oxygen tensions would regulate the G PLC pathway in human PMN. G PLC activity was assayed by investigating phospholipase C activity by measuring inositol phosphates and diacylglycerol (DAG) formation. Respiratory burst activity was assayed as O 2 production and NADPH oxidase activation in intact PMN and in a cell-free system, respectively, and correlated separately to both early and late DAG production. At 1 min, DAG formation during normoxia was decreased by IL-8 plus fibronectin while hypoxia had no regulatory effect on control of DAG formation by any of the cytokines. In contrast to early DAG formation, hypoxia significantly downregulated late DAG formation induced by buffer without fibronectin, IL-8 plus fibronectin, and IL-1 beta with or without fibronectin. Hypoxia/reoxygenation in and of itself significantly increased DAG formation vs levels seen in the presence or absence of IL-8, TNF-alpha, or IL-1 beta with or without fibronectin. Changes in early DAG production during the alterations in oxygen tension correlated best with corresponding changes in O 2 production in intact cells, whereas late DAG production correlated best with NADPH oxidase activation assayed in the cell-free system. Thus, changes in oxygen tension can directly modulate the extent of the PMN response to stimulation by IL-8, TNF-alpha, or IL-1 beta and the G PLC-receptor pathway is particularly regulated by physiologically relevant periods of hypoxia/reoxygenation.

Altered oxygen tension modulates cytokine-induced signal transduction in polymorphonuclear leukocytes: regulation of the GPLD pathway.

We investigated the effect of alterations in buffer oxygen tensions from normoxia (PO2 = 180-200 mm/Hg) to hypoxia (PO2 < 30 mm/HG) and then reoxygenation (PO2 > 140 mmHg) on the GPLD-pathway by measuring phosphatidylethanol formation in the presence of ethanol and subsequent NADPH oxidase activation and O2-production in polymorphonuclear leukocytes (PMN). Experiments were performed with PMN stimulated with either interleukin (IL)-8, tumor necrosis factor (TNF)-alpha, or IL-1 beta in the presence or absence of fibronectin. Hypoxia exerted a downregulating effect on this pathway and reoxygenation restored GPLD activation to levels seen during normoxia; however, supraphysiological concentrations of cytokines were able to reverse this pattern. Changes in GPLD activation correlated best with changes in O2-production during the hypoxia to hypoxia/reoxygenation transition induced by TNF-alpha-Fn and IL-1 beta +/- Fn. Thus, changes in oxygen tension can directly modulate the extent of the PMN response to stimulation by IL-8, TNF-alpha, or IL-1 beta, and activation of the GPLD-pathway appears to be highly sensitive to hypoxia and hypoxia/reoxygenation.

Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours.

Apoptosis is a genetically encoded programme of cell death that can be activated under physiological conditions and may be an important safeguard against tumour development. Regions of low oxygen (hypoxia) and necrosis are common features of solid tumours. Here we report that hypoxia induces apoptosis in oncogenically transformed cells and that further genetic alterations, such as loss of the p53 tumour-suppressor gene or overexpression of the apoptosis-inhibitor protein Bcl-2, substantially reduce hypoxia-induced cell death. Hypoxia also selects for cells with defects in apoptosis, because small numbers of transformed cells lacking p53 overtake similar cells expressing wild-type p53 when treated with hypoxia. Furthermore, highly apoptotic regions strongly correlate with hypoxic regions in transplanted tumours expressing wild-type p53, whereas little apoptosis occurs in hypoxic regions of p53-deficient tumours. We propose that hypoxia provides a physiological selective pressure in tumours for the expansion of variants that have lost their apoptotic potential, and in particular for cells acquiring p53 mutations.

Attenuated response of p53 and p21 in primary cultures of human prostatic epithelial cells exposed to DNA-damaging agents.

The multifocal origin of prostate cancer suggests a pan-organ defect in a tumor suppressor pathway. Although structural mutations in the p53 gene have been implicated in late-stage prostate cancer, little is known about the p53 response to genotoxic stress in normal human prostatic epithelial cells from which adenocarcinomas originate. We found that the majority (10 of 12) of epithelial cell cultures derived from histologically normal tissues of radical prostatectomy specimens failed to exhibit p53 accumulation in response to ionizing radiation. Epithelial cell cultures derived from benign prostatic hyperplasia and a primary prostatic adenocarcinoma also failed to accumulate p53 in response to ionizing radiation. In contrast, cultures of prostatic stromal cells derived from normal, benign prostatic hyperplasia, or adenocarcinoma tissues exhibited a 3-9-fold induction of p53 within 1-3 h after irradiation. Since p53 regulates a cell cycle checkpoint through the induction of the cyclin-cdk inhibitor p21, we examined p21 accumulation and cell cycle arrest following exposure to ionizing radiation. With one exception, epithelial cells that did not display increased p53 or p21 induction did not demonstrate a significant G1-S arrest in response to ionizing radiation, whereas stromal cells that accumulated p53 and p21 exhibited a large cell cycle arrest. These results indicate a functional difference between the DNA damage response of epithelial and stromal prostatic cells and suggest a possible mechanism for the increased susceptibility of prostatic epithelial cells to accumulate genetic alterations.

Hypoxia induces accumulation of p53 protein, but activation of a G1-phase checkpoint by low-oxygen conditions is independent of p53 status.

It has been convincingly demonstrated that genotoxic stresses cause the accumulation of the tumor suppressor gene p53. One important consequence of increased p53 protein levels in response to DNA damage is the activation of a G1-phase cell cycle checkpoint. It has also been shown that G1-phase cell cycle checkpoints are activated in response to other stresses, such as lack of oxygen. Here we show that hypoxia and heat, agents that induce cellular stress primarily by inhibiting oxygen-dependent metabolism and denaturing proteins, respectively, also cause an increase in p53 protein levels. The p53 protein induced by heat is localized in the cytoplasm and forms a complex with the heat shock protein hsc70. The increase in nuclear p53 protein levels and DNA-binding activity and the induction of reporter gene constructs containing p53 binding sites following hypoxia occur in cells that are wild type for p53 but not in cells that possess mutant p53. However, unlike ionizing radiation, the accumulation of cells in G1 phase by hypoxia is not strictly dependent on wild-type p53 function. In addition, cells expressing the human papillomavirus E6 gene, which show increased degradation of p53 by ubiquitination and fail to accumulate p53 in response to DNA-damaging agents, do increase their p53 levels following heat and hypoxia. These results suggest that hypoxia is an example of a "nongenotoxic" stress which induces p53 activity by a different pathway than DNA-damaging agents.