Testosterone supplementation upregulates androgen receptor expression and translational capacity during severe energy deficit

Testosterone supplementation during energy deficit promotes whole body lean mass accretion, but the mechanisms underlying that effect remain unclear. To elucidate those mechanisms, skeletal muscle molecular adaptations were assessed from muscle biopsies collected before, 1 h, and 6 h after exercise and a mixed meal (40 g protein, 1 h postexercise) following 14 days of weight maintenance (WM) and 28 days of an exercise- and diet-induced 55% energy deficit (ED) in 50 physically active nonobese men treated with 200 mg testosterone enanthate/wk (TEST) or placebo (PLA) during the ED. Participants (n = 10/group) exhibiting substantial increases in leg lean mass and total testosterone (TEST) were compared with those exhibiting decreases in both of these measures (PLA). Resting androgen receptor (AR) protein content was higher and fibroblast growth factor-inducible 14 (Fn14), IL-6 receptor (IL-6R), and muscle ring-finger protein-1 gene expression was lower in TEST vs. PLA during ED relative to WM (P < 0.05). Changes in inflammatory, myogenic, and proteolytic gene expression did not differ between groups after exercise and recovery feeding. Mechanistic target of rapamycin signaling (i.e., translational efficiency) was also similar between groups at rest and after exercise and the mixed meal. Muscle total RNA content (i.e., translational capacity) increased more during ED in TEST than PLA (P < 0.05). These findings indicate that attenuated proteolysis at rest, possibly downstream of AR, Fn14, and IL-6R signaling, and increased translational capacity, not efficiency, may drive lean mass accretion with testosterone administration during energy deficit.

Trichoderma harzianum transcriptome in response to cadmium exposure

Cadmium (Cd) is a heavy metal present in the environment mainly as a result of industrial contamination that can cause toxic effects to life. Some microorganisms, as Trichoderma harzianum, a fungus used in biocontrol, are able to survive in polluted environments and act as bioremediators. Aspects about the tolerance to the metal have been widely studied in other fungi although there are a few reports about the response of T. harzianum. In this study, we determined the effects of cadmium over growth of T. harzianum and used RNA-Seq to identify significant genes and processes regulated in the metal presence. Cadmium inhibited the fungus growth proportionally to its concentration although the fungus exhibited tolerance as it continued to grow, even in the highest concentrations used. A total of 3767 (1993 up and 1774 down) and 2986 (1606 up and 1380 down) differentially expressed genes were detected in the mycelium of T. harzianum cultivated in the presence of 1.0 mg mL^-1 or 2.0 mg mL^-1 of CdCl₂, respectively, compared to the absence of the metal. Of these, 2562 were common to both treatments. Biological processes related to cellular homeostasis, transcription initiation, sulfur compound biosynthetic and metabolic processes, RNA processing, protein modification and vesicle-mediated transport were up-regulated. Carbohydrate metabolic processes were down-regulated. Pathway enrichment analysis indicated induction of glutathione and its precursor's metabolism. Interestingly, it also indicated an intense transcriptional induction, especially by up-regulation of spliceosome components. Carbohydrate metabolism was repressed, especially the mycoparasitism-related genes, suggesting that the mycoparasitic ability of T. harzianum could be affected during cadmium exposure. These results contribute to the advance of the current knowledge about the response of T. harzianum to cadmium exposure and provide significant targets for biotechnological improvement of this fungus as a bioremediator and a biocontrol agent.

β-Methylamino-L-alanine substitution of serine in SOD1 suggests a direct role in ALS etiology

Exposure to the environmental toxin β-methylamino-L-alanine (BMAA) is linked to amyotrophic lateral sclerosis (ALS), but its disease-promoting mechanism remains unknown. We propose that incorporation of BMAA into the ALS-linked protein Cu,Zn superoxide dismutase (SOD1) upon translation promotes protein misfolding and aggregation, which has been linked to ALS onset and progression. Using molecular simulation and predictive energetic computation, we demonstrate that substituting any serine with BMAA in SOD1 results in structural destabilization and aberrant dynamics, promoting neurotoxic SOD1 aggregation. We propose that translational incorporation of BMAA into SOD1 is directly responsible for its toxicity in neurodegeneration, and BMAA modification of SOD1 may serve as a biomarker of ALS.

Manganese causes neurotoxic iron accumulation via translational repression of amyloid precursor protein and H-Ferritin

For more than 150 years, it is known that occupational overexposure of manganese (Mn) causes movement disorders resembling Parkinson's disease (PD) and PD-like syndromes. However, the mechanisms of Mn toxicity are still poorly understood. Here, we demonstrate that Mn dose- and time-dependently blocks the protein translation of amyloid precursor protein (APP) and heavy-chain Ferritin (H-Ferritin), both iron homeostatic proteins with neuroprotective features. APP and H-Ferritin are post-transcriptionally regulated by iron responsive proteins, which bind to homologous iron responsive elements (IREs) located in the 5'-untranslated regions (5'-UTRs) within their mRNA transcripts. Using reporter assays, we demonstrate that Mn exposure repressed the 5'-UTR-activity of APP and H-Ferritin, presumably via increased iron responsive proteins-iron responsive elements binding, ultimately blocking their protein translation. Using two specific Fe2+ -specific probes (RhoNox-1 and IP-1) and ion chromatography inductively coupled plasma mass spectrometry (IC-ICP-MS), we show that loss of the protective axis of APP and H-Ferritin resulted in unchecked accumulation of redox-active ferrous iron (Fe2+ ) fueling neurotoxic oxidative stress. Enforced APP expression partially attenuated Mn-induced generation of cellular and lipid reactive oxygen species and neurotoxicity. Lastly, we could validate the Mn-mediated suppression of APP and H-Ferritin in two rodent in vivo models (C57BL6/N mice and RjHan:SD rats) mimicking acute and chronic Mn exposure. Together, these results suggest that Mn-induced neurotoxicity is partly attributable to the translational inhibition of APP and H-Ferritin resulting in impaired iron metabolism and exacerbated neurotoxic oxidative stress. OPEN SCIENCE BADGES: This article has received a badge for *Open Materials* because it provided all relevant information to reproduce the study in the manuscript. The complete Open Science Disclosure form for this article can be found at the end of the article. More information about the Open Practices badges can be found at https://cos.io/our-services/open-science-badges/.

Susceptibility of isolated myofibrils to in vitro glutathionylation: Potential relevance to muscle functions

In this study we investigated the molecular mechanism of glutathionylation on isolated human cardiac myofibrils using several pro-glutathionylating agents. Total glutathionylated proteins appeared significantly enhanced with all the pro-oxidants used. The increase was completely reversed by the addition of a reducing agent, demonstrating that glutathione binding occurs by a disulfide and that the process is reversible. A sensitive target of glutathionylation was alpha-actin, showing a different reactivity to the several pro-glutathionylating agents by ELISA. Noteworthy, myosin although highly sensitive to the in vitro glutathionylation does not represent the primary glutathionylation target in isolated myofibrils. Light scattering measurements of the glutathionylated alpha-actin showed a slower polymerisation compared to the non-glutathionylated protein and force development was depressed after glutathionylation, when the myofibrils were mounted in a force recording apparatus. Interestingly, confocal laser scanning microscopy of cardiac cryosections indicated, for the first time, the constitutive glutathionylation of alpha-cardiac actin in human heart. Due to the critical location of alpha-actin in the contractile machinery and to its susceptibility to the oxidative modifications, glutathionylation may represent a mechanism for modulating sarcomere assembly and muscle functionality under patho-physiological conditions in vivo.

MEPE's diverse effects on mineralization

Matrix extracellular phosphoglycoprotein (MEPE) is an inhibitor of mineralization in situ and in cell cultures where altered expression is associated with oncogenic osteomalacia and hypophosphatemic rickets. The purpose of this study was to determine whether the intact protein or the peptide(s) originating from this protein was responsible for the inhibition. The ability of the intact protein and the acidic, serine- and aspartate-rich MEPE-associated motif (ASARM) peptide to promote or inhibit de novo hydroxyapatite formation and growth of hydroxyapatite seed crystals, in both phosphorylated and dephosphorylated forms, was assessed at room temperature in a dynamic gel diffusion system at 3.5 and 5 days. The most effective nucleator concentration was also examined when associated with fibrillar type I collagen. The phosphorylated intact protein was an effective promoter of mineralization in the gelatin gel diffusion system, while the ASARM peptide was an effective inhibitor. When dephosphorylated both the intact protein and the ASARM peptide had no effect on mineralization. Associated with collagen fibrils, some of the effect of the intact protein was lost. This study demonstrates the importance of posttranslational modification for the site-specific activity of MEPE and its ASARM peptide.

Effects of cigarette smoke borne reactive nitrogen species on salivary alpha-amylase activity and protein modifications

Cigarette smoke (CS) is associated with a variety of human pathologies including cardiovascular disease and cancer. Oral squamous cell carcinoma (OSCC) is the most common malignancy of the head and neck. The major inducer of OSCC is exposure to tobacco. Recent studies demonstrated that oxidative and nitrosative stress contributes to the development of oral carcinogenesis through DNA damage. All salivary reactive nitrogen species (RNS) analyzed from OSCC patients are significantly higher in comparison with healthy subjects. Our findings show that CS and external RNS addition induced reduction in alpha-amylase activity and produced some excited carbonyl formation, but to a much less extant than CS. The addition of epigallocatechine-3-gallate (EGCG) to saliva produced no protective effect against damage to alpha-amylase activity. Our proposed mechanism for the decrease in alpha-amylase activity is the formation of adducts at SH groups of the alpha-amylase active site. In this case, EGCG was unable to counteract this phenomenon, as it does not reduce the concentration of disulfides, and does not alter the amount of protein-SH moieties. However, EGCG did reduce the levels of excited carbonyl formation. Our results indicate that although RNS are abundant in CS, a significant decrease in amylase activity is due to other components in CS, probably aldehydes, reacting with the thiol group of proteins by the Michael addition reaction.

Serum vitamin E and oxidative protein modification in hemodialysis: a randomized clinical trial

BACKGROUND

Patients with end-stage renal disease have increased circulating concentrations of oxidatively modified circulating proteins. Therefore, we examined the ability of vitamin E alpha (alpha-tocopherol) to alter levels of these modified proteins.

STUDY DESIGN

Randomized clinical trial.

SETTING & PARTICIPANTS

27 clinically stable patients treated by means of hemodialysis in 4 freestanding outpatient dialysis units.

INTERVENTION

Oral administration of 800 IU of vitamin E alpha or placebo daily.

OUTCOMES & MEASUREMENTS

Plasma levels of alpha- and gamma-tocopherol and oxidative protein modifications reflecting 2 pathways for protein-oxidant damage. The advanced glycation end product pentosidine reflects glycoxidation. The lipid peroxidation products iso[4]-levuglandin E(2), (E)-4-hydroxy-2-nonenal, and (E)-4-oxo-2-nonenal are formed through covalent adduction.

RESULTS

Circulating levels of all oxidative protein modifications were increased in patients with end-stage renal disease. Supplementation with alpha-tocopherol caused alpha-tocopherol levels to rise (13.2 +/- 3.7 to 27.3 +/- 14 mug/mL), but gamma-tocopherol levels to decrease (4.1 +/- 1.6 to 3.5 +/- 1.1 mug/mL). Control values were unchanged. There was no effect on oxidative protein modifications (placebo versus treatment; mean for pentosidine, 15.6 +/- 11.4 (SD): 95% confidence interval (CI), 8.2 to 23.1 versus 21.3 +/- 9.0 pg/mg protein; 95% CI, 16.1 to 26.6; iso[4]-levuglandin E(2), 8.31 +/- 2.55; 95% CI, 6.77 to 9.85 versus 8.46 +/- 2.37 nmol/mL; 95% CI, 7.09 to 9.84; (E)-4-hydroxy-2-nonenal, 0.51 +/- 0.11; 95% CI, 0.45 to 0.57 versus 0.51 +/- 0.08 nmol/mL; 95% CI, 0.46 to 0.56; (E)-4-oxo-2-nonenal, 189 +/- 44; 95% CI, 162 to 215 vs 227 +/- 72 pmol/mL; 95% CI, 183 to 271).

LIMITATIONS

Sample size was adequate to show changes in alpha- and gamma-tocopherol levels in response to treatment. However, power was insufficient to show an effect on oxidative protein modifications.

CONCLUSIONS

Intervention of oral supplementation with alpha-tocopherol did not result in changes in circulating oxidative protein modifications. A larger study may be required to show an effect in this clinical setting.

leptin-induced growth stimulation of breast cancer cells involves recruitment of histone acetyltransferases and mediator complex to CYCLIN D1 promoter via activation of Stat3

Numerous epidemiological studies documented that obesity is a risk factor for breast cancer development in postmenopausal women. Leptin, the key player in the regulation of energy balance and body weight control also acts as a growth factor on certain organs in both normal and disease state. In this study, we analyzed the role of leptin and the molecular mechanism(s) underlying its action in breast cancer cells that express both short and long isoforms of leptin receptor. Leptin increased MCF7 cell population in the S-phase of the cell cycle along with a robust increase in CYCLIN D1 expression. Also, leptin induced Stat3-phosphorylation-dependent proliferation of MCF7 cells as blocking Stat3 phosphorylation with a specific inhibitor, AG490, abolished leptin-induced proliferation. Using deletion constructs of CYCLIN D1 promoter and chromatin immunoprecipitation assay, we show that leptin induced increase in CYCLIN D1 promoter activity is mediated through binding of activated Stat3 at the Stat binding sites and changes in histone acetylation and methylation. We also show specific involvement of coactivator molecules, histone acetyltransferase SRC1, and mediator complex in leptin-mediated regulation of CYCLIN D1 promoter. Importantly, silencing of SRC1 and Med1 abolished the leptin induced increase in CYCLIN D1 expression and MCF7 cell proliferation. Intriguingly, recruitment of both SRC1 and Med1 was dependent on phosphorylated Stat3 as AG490 treatment inhibited leptin-induced recruitment of these coactivators to CYCLIN D1 promoter. Our data suggest that CYCLIN D1 may be a target gene for leptin mediated growth stimulation of breast cancer cells and molecular mechanisms involve activated Stat3-mediated recruitment of distinct coactivator complexes.

Aminoglycoside-induced translational read-through in disease: overcoming nonsense mutations by pharmacogenetic therapy

A third of inherited diseases result from premature termination codon mutations. Aminoglycosides have emerged as vanguard pharmacogenetic agents in treating human genetic disorders due to their unique ability to suppress gene translation termination induced by nonsense mutations. In preclinical and pilot clinical studies, this therapeutic approach shows promise in phenotype correction by promoting otherwise defective protein synthesis. The challenge ahead is to maximize efficacy while preventing interaction with normal protein production and function.

Secretion and lysophospholipase D activity of autotaxin by adipocytes are controlled by N-glycosylation and signal peptidase

Autotaxin (ATX) is a lysophospholipase D involved in synthesis of lysophosphatidic acid (LPA). ATX is secreted by adipocytes and is associated with adipogenesis and obesity-associated diabetes. Here we have studied the mechanisms involved in biosynthesis and secretion of ATX by mouse 3T3-F442A adipocytes. We found that inhibition of N-glycosylation with tunicamycin or by double point deletion of the amino-acids N53 and N410 of ATX inhibit its secretion. In addition, N-glycosidase treatment and point deletion of the amino-acid N410 inhibits the lysophospholipase D activity of ATX. Analysis of the amino-acid sequence of mouse ATX shows the presence of a N-terminal signal peptide. Treatment with the signal peptidase inhibitor globomycin inhibits ATX secretion by adipocytes. Transfection in Cos-7 cells of site-directed deleted ATX shows that ATX secretion is dependent on the hydrophobic core sequence of the signal peptide, not on the putative signal peptidase cleavage site sequence. Analysis of the amino-acid sequence of mouse ATX also reveals the presence of a putative cleavage site by the protein convertase furin. Treatment of adipocytes with the furin inhibitor decanoyl-Arg-Val-Lys-Arg-chloromethylketone does not modified secretion or lysophospholipase D activity of ATX. Transfection in Cos-7 cells of site-directed deleted ATX shows that the furin recognition site is not required for secretion or lysophospholipase D activity of ATX. In conclusion, the present work demonstrates the crucial role of N-glycosylation in secretion and activity of ATX. The present work also confirms the crucial role signal peptidase in secretion of ATX by adipocytes.

Structural and mechanistic features of protein O glycosylation linked to CD8+ T-cell apoptosis

CD8+ T-cell apoptosis is essential for the contraction phase of the immune response, yet the initiating signals and precise pathways involved are unresolved. The ST3Gal-I sialyltransferase is a candidate mechanistic component and catalyzes sialic acid addition to core 1 O-glycans during protein O glycosylation. ST3Gal-I inactivation or enzymatic removal of its product renders CD8+ T cells, but not CD4+ T cells, susceptible to apoptosis by differential cross-linking of O-glycoproteins in the absence of interleukin-2 and T-cell receptor (TCR) signaling. This results in caspase activation, DNA fragmentation, and phosphatidylserine externalization prior to cell death. We further show that ST3Gal-I function is regulated by a posttranscriptional mechanism operating distal to Golgi core 2 O glycosylation and is invariably linked to CD8+ T-cell contraction following viral (lymphocytic choriomeningitis virus) infection and bacterial (staphylococcal enterotoxin B) antigen immunization. The mechanism does not involve the ST3Gal-I substrate CD43 or core 2 O-glycan induction and overcomes the ability of Bcl-2 to inhibit the contraction phase in vivo. Loss of ST3Gal-I function further reduces Bim-deficient CD8+ T-cell accumulation without diminishing apoptotic sensitivity. We propose that an endogenous lectin activates an apoptotic pathway constructed in CD8+ T cells following TCR stimulation and enables contraction upon attenuation of immune signaling.

Increase in HLA-G1 proteolytic shedding by tumor cells: a regulatory pathway controlled by NF-κB inducers

HLA-G is expressed by tumors, in which it contributes to the evasion of immunosurveillance. NF-kappaB appears to be a candidate for regulating HLA-G expression, since it is considered to be a hallmark of cancer. We investigated the role of NF-kappaB in modulating HLA-G expression in HLA-G-positive tumor cells, JEG-3 (choriocarcinoma), FON (melanoma), and M8-HLA-G1 (HLAG1-transfected melanoma). The treatment of tumor cells with two NF-kappaB inducers, tumor necrosis factor-alpha and phorbol 12-myristate 13-acetate, decreased HLA-G1 cell surface expression but increased intracytoplasmic HLA-G proteins. Reduction in HLA-G1 cell surface expression is driven by NF-kappaB and involves a proteolytic shedding process dependent on metalloproteinase activity. In contrast, an increase in intracytoplasmic HLA-G proteins involves post-transcriptional mechanisms that are independent of NF-kappaB. These results, and the fact that soluble HLA-G1 reduces the cytotoxicity of the NKL cell line, lead us to propose a novel regulatory pathway for HLA-G expression by tumor cells that may have particular relevance in tumor escape.

A study of CD33 (SIGLEC-3) antigen expression and function on activated human T and NK cells: two isoforms of CD33 are generated by alternative splicing

The expression of CD33, a restricted leukocyte antigen considered specific for myeloid lineage, has been studied extensively on lymphoid cells. We demonstrated that wide subsets of mitogen- or alloantigen-activated human T and natural killer (NK) cells express CD33 at protein and nucleic acid levels. CD33+ and CD33- T and NK cell populations showed identical surface expression of activation markers such as CD25, CD28, CD38, CD45RO, or CD95. Myeloid and lymphoid CD33 cDNA were identical. However, lymphoid CD33 protein had lower molecular weight, suggesting cell type-specific, post-translational modifications. Additionally, reverse transcriptase-polymerase chain reaction and Northern blot analysis showed an unknown CD33 isoform (CD33m) expressed on all CD33+ cell lines or T cell clones tested. CD33m was identical to CD33 (CD33M) in the signal peptide, the immunoglobulin (Ig) domain C2, the transmembrane, and the cytoplasmic regions but lacked the extracellular ligand-binding variable Ig-like domain encoded by the second exon. CD33m mRNA was mostly detected on NKL and myeloid cell lines but poorly expressed on B cell lines and T lymphocytes. The CD33m extracellular portion was successfully expressed as a soluble fusion protein on transfected human cells, suggesting a functional role on cell membranes. Cross-linking of CD33 diminished the cytotoxic activity of NKL cells against K562 and P815 target cells, working as an inhibitory receptor on NK cells. These data demonstrate that CD33 expression is not restricted to the myeloid lineage and could exist as two different splicing variants, which could play an important role in the regulation of human lymphoid and myeloid cells.

Differences in the apical and basolateral pathways for glycosaminoglycan biosynthesis in Madin–Darby canine kidney cells

Serglycin with a green fluorescent protein tag (SG-GFP) expressed in epithelial Madin-Darby canine kidney cells is secreted mainly (85%) into the apical medium, but the glycosaminoglycan (GAG) chains on the SG-GFP protein core secreted basolaterally (15%) carry most of the sulfate added during biosynthesis (Tveit et al. (2005) J. Biol. Chem., 280, 29596-29603). Here we report further differences in apical and basolateral GAG synthesis. The less intensely sulfated chondroitin sulfate (CS) chains on apically secreted SG-GFP are longer than CS chains attached to basolateral SG-GFP, whereas the heparan sulfate (HS) chains are of similar lengths. When the supply of 3'-phosphoadenosine-5'-phosphosulfate (PAPS) is limited by chlorate treatment, the synthesis machinery maintains sulfation of HS chains on basolateral SG-GFP until it is inhibited at 50 mM chlorate, whereas basolateral CS chains lose sulfate already at 12.5 mM chlorate and become longer. Apically, incorporation of 35S-sulfate into CS is reduced to a lesser extent at higher chlorate concentrations than basolateral CS, although apical CS is less intensely sulfated than basolateral CS in control cells. Similar to what was found for basolateral HS, sulfation of apical HS was not reduced at chlorate concentrations below 50 mM. Also, protein-free, xyloside-based GAG chains secreted basolaterally are more intensely sulfated than their apical counterpart, supporting the view that separate apical and basolateral pathways exist for GAG synthesis and sulfation. Introduction of benzyl beta-d-xyloside (BX) to the GAG synthesis machinery reduces the apical secretion of SG-GFP dramatically and also the modification of SG-GFP by HS.

Modulation of HSP70 GlcNAc-directed lectin activity by glucose availability and utilization

It is well-accepted that protein quality control (occurring either after protein synthesis or after cell damage) is mainly ensured by HSP, but the mechanism by which HSP decides whether the protein will be degraded or not is poorly understood. Within this framework, it has been hypothesized that O-GlcNAc, a cytosolic and nuclear-specific glycosylation whose functions remain unclear, could take a part in the protection of proteins against degradation by modifying both the proteins themselves and the proteasome. Because the synthesis of O-GlcNAc is tightly correlated to glucose metabolism and Hsp70 was endowed with GlcNAc-binding property, we studied the relationship between GlcNAc-binding activity of both Hsp70 and Hsc70 (the nucleocytoplasmic forms of HSP70 family) and glucose availability and utilization. We thus demonstrated that low glucose concentration, inhibition of glucose utilization with 2DG, or inhibition of glucose transport with CytB led to an increase of Hsp70 and Hsc70 lectin activities. Interestingly, the response of Hsp70 and Hsc70 lectin activities toward variations of glucose concentration appeared different: Hsp70 lost its lectin activity when glucose concentration was >5 mM (i.e., physiological glucose concentration) in contrast to Hsc70 that exhibited a maximal lectin activity for glucose concentration approximately 5 mM and at high glucose concentrations. This work also demonstrates that HSP70 does not regulate its GlcNAc-binding properties through its own O-GlcNAc glycosylation.

Nuclear factor-kappaB induced by doxorubicin is deficient in phosphorylation and acetylation and represses nuclear factor-kappaB-dependent transcription in cancer cells

The primary goal of chemotherapy is to cause cancer cell death. However, a side effect of many commonly used chemotherapeutic drugs is the activation of nuclear factor-kappaB (NF-kappaB), a potent inducer of antiapoptotic genes, which may blunt the therapeutic efficacy of these compounds. We have assessed the effect of doxorubicin, an anthracycline in widespread clinical use, on NF-kappaB activation and expression of antiapoptotic genes in breast cancer cells. We show that doxorubicin treatment activates NF-kappaB signaling and produces NF-kappaB complexes that are competent for NF-kappaB binding in vitro. Surprisingly, these NF-kappaB complexes suppress, rather than activate, constitutive- and cytokine-induced NF-kappaB-dependent transcription. We show that doxorubicin treatment produces RelA, which is deficient in phosphorylation and acetylation and which blocks NF-kappaB signaling in a histone deacetylase-independent manner, and we show that NF-kappaB activated by doxorubicin does not remain stably bound to kappaB elements in vivo. Together these data show that NF-kappaB signaling induced by doxorubicin reduces expression of NF-kappaB-dependent genes in cancer cells.