Genome-wide DNA hydroxymethylation identifies potassium channels in the nucleus accumbens as discriminators of methamphetamine addiction and abstinence.

Epigenetic consequences of exposure to psychostimulants are substantial but the relationship of these changes to compulsive drug taking and abstinence is not clear. Here, we used a paradigm that helped to segregate rats that reduce or stop their methamphetamine (METH) intake (nonaddicted) from those that continue to take the drug compulsively (addicted) in the presence of footshocks. We used that model to investigate potential alterations in global DNA hydroxymethylation in the nucleus accumbens (NAc) because neuroplastic changes in the NAc may participate in the development and maintenance of drug-taking behaviors. We found that METH-addicted rats did indeed show differential DNA hydroxymethylation in comparison with both control and nonaddicted rats. Nonaddicted rats also showed differences from control rats. Differential DNA hydroxymethylation observed in addicted rats occurred mostly at intergenic sites located on long and short interspersed elements. Interestingly, differentially hydroxymethylated regions in genes encoding voltage (Kv1.1, Kv1.2, Kvb1 and Kv2.2)- and calcium (Kcnma1, Kcnn1 and Kcnn2)-gated potassium channels observed in the NAc of nonaddicted rats were accompanied by increased mRNA levels of these potassium channels when compared with mRNA expression in METH-addicted rats. These observations indicate that changes in differentially hydroxymethylated regions and increased expression of specific potassium channels in the NAc may promote abstinence from drug-taking behaviors. Thus, activation of specific subclasses of voltage- and/or calcium-gated potassium channels may provide an important approach to the beneficial treatment for METH addiction.

Epigenetics and addiction.

Addictions are public health menaces. However, despite advances in addiction research, the cellular or molecular mechanisms that cause transition from recreational use to addiction remain to be elucidated. We have recently suggested that addiction may be secondary to long-term epigenetic modifications that determine the clinical course of substance use disorders. A better understanding of epigenetic mechanisms in animal models that mimic human conditions should help to usher in a new area of drug development against addiction.

Long-term protective effects of methamphetamine preconditioning against single-day methamphetamine toxic challenges.

Methamphetamine (METH) use is associated with neurotoxic effects which include decreased levels of dopamine (DA), serotonin (5-HT) and their metabolites in the brain. We have shown that escalating METH dosing can protect against METH induced neurotoxicity in rats sacrificed within 24 hours after a toxic METH challenge. The purpose of the current study was to investigate if the protective effects of METH persisted for a long period of time. We also tested if a second challenge with a toxic dose of METH would cause further damage to monoaminergic terminals. Saline-pretreated rats showed significant METH-induced decreases in striatal DA and 5-HT levels in rats sacrificed 2 weeks after the challenge. Rats that received two METH challenges showed no further decreases in striatal DA or 5-HT levels in comparison to the single METH challenge. In contrast, METH-pretreated rats showed significant protection against METH-induced striatal DA and 5-HT depletion. In addition, the METH challenge causes substantial decreases in cortical 5-HT levels which were not further potentiated by a second drug challenge. METH preconditioning provided almost complete protection against METH -induced 5-HT depletion. These results are consistent with the idea that METH pretreatment renders the brain refractory to METH-induced degeneration of brain monoaminergic systems.

Transcriptional responses to reinforcing effects of cocaine in the rat hippocampus and cortex.

The psychostimulant effects of cocaine are thought to result from its ability to block dopamine (DA) uptake and increase DA levels in ventral striatum. In addition, cocaine causes biochemical changes in the brain areas involved in learning and memory, including hippocampus and cortex, whose role in drug reinforcement is now being actively investigated. Thus, we studied molecular events in the hippocampus and frontal cortex of rats treated with cocaine conditioned place preference (CPP) paradigm. After exposure to cocaine conditioning (cocaine paired), cocaine alone (cocaine non-paired) or saline rats were tested for place conditioning. Cocaine (10 mg/kg) caused increases in time spent in the drug-paired compartment. By using microarray analyses, we examined gene expression in the hippocampi and frontal cortices of cocaine-paired rats, cocaine non-paired and saline-treated controls. Our study revealed that 214 transcripts were differentially regulated in the hippocampi of cocaine-paired rats. These include genes that play roles in protein phosphorylation, RNA processing and protein synthesis, ubiquitin-dependent protein degradation and cytoskeleton organization. In contrast, 39 genes were differently expressed in the frontal cortex. Our data support the possibility that molecular changes in the hippocampus might participate in the formation and maintenance of memory patterns induced by cocaine in the brain. Differences in the transcriptional responses in the hippocampus and cortex suggest the primary importance of the hippocampus for recent memory processing associated with cocaine-induced CPP.

Methamphetamine causes differential regulation of pro-death and anti-death Bcl-2 genes in the mouse neocortex.

Bcl-2, an inner mitochondrial membrane protein, inhibits apoptotic neuronal cell death. Expression of Bcl-2 inhibits cell death by decreasing the net cellular generation of reactive oxygen species. Studies by different investigators have provided unimpeachable evidence of a role for oxygen-based free radicals in methamphetamine (METH) -induced neurotoxicity. In addition, studies from our laboratory have shown that immortalized rat neuronal cells that overexpress Bcl-2 are protected against METH-induced apoptosis in vitro. Moreover, the amphetamines can cause differential changes in the expression of Bcl-X splice variants in primary cortical cell cultures. These observations suggested that METH might also cause perturbations of Bcl-2-related genes when administered to rodents. Thus, the present study was conducted to determine whether the use of METH might indeed be associated with transcriptional and translational changes in the expression of Bcl-2-related genes in the mouse brain. Here we report that a toxic regimen of METH did cause significant increases in the pro-death Bcl-2 family genes BAD, BAX, and BID. Concomitantly, there were significant decreases in the anti-death genes Bcl-2 and Bcl-XL. These results thus support the notion that injections of toxic doses of METH trigger the activation of the programmed death pathway in the mammalian brain.

Temporal profiling of methamphetamine-induced changes in gene expression in the mouse brain: evidence from cDNA array.

Methamphetamine (METH) is a neurodegenerative drug of abuse. Its toxicity is characterized by destruction of monoaminergic terminals and by apoptosis in cortical and striatal cell bodies. Multiple factors appear to control METH neurotoxicity, including free radicals and transcription factors. Here, using cDNA arrays, we show the temporal profile of gene expression patterns in the cortex of mice treated with this drug. We obtained two patterns of changes from 588 genes surveyed. First, an early pattern is characterized by upregulation of transcription factors, including members of the jun family. Second, a delayed pattern includes genes related to cell death and to DNA repair. A number of trophic factors were also activated at the later timepoint. These observations suggest that METH can activate a multigene machinery that participates in the production of its toxic effects. The resulting degenerative effects of the drug are thus the result of a balance between protoxic and antiapoptotic mechanisms triggered by its administration to these animals. These observations are of clinical relevance because of the recent identification of degenerative changes in the brains of METH abusers.

Profile of genes showing increased expression following dopamine and methamphetamine exposure in an immortalized neuronal cell line.

Methamphetamine (METH) is a drug of abuse with well-described neurodegenerative effects. Some of the METH-induced degenerative manifestations are thought to be due to increased dopamine (DA) release in the cytoplasm of nerve terminals and subsequent extravasation in the synaptic cleft. Using an immortalized neural cell line, we have made use of the comprehensive cDNA array technology in order to compare and contrast the molecular effects of DA and METH. We found that the two compounds do have many similar but also different effects. Since these neural cells produce no DA, these results demonstrate that many of the METH-induced responses attributed to DA might, in fact, be intrinsic to METH itself. More biochemical studies are needed to investigate DA-independent METH deleterious events in the central nervous system.

Dual mechanism of Fas-induced cell death in neuroglioma cells: a role for reactive oxygen species.

ApoI/Fas belongs to the tumor necrosis factor receptor (TNFR) superfamily and mediates cell death in various cell types. A dual mode of Fas-triggered cell death has been reported depending on cell types used in the experiments. The present study was carried out to test the possible role of reactive oxygen species in this dual mechanism in neuroglioma cells. Anti-Fas antibody caused dose-dependent and time-dependent increase in cell death measured by lactate dehydrogenase (LDH) release in control neuroglioma cells and in cells that were transfected with catalase cDNA. However, cells transfected with copper/zinc superoxide dismutase (Cu/ZnSOD) cDNA showed marked attenuation of Fas-induced LDH release. Moreover, flow cytometry and confocal microscopy revealed that Fas-induced cell death in control cells occur mostly through an apoptotic process. This process was also completely abrogated in cells overexpressing catalase or copper/zinc superoxide dismutase (Cu/ZnSOD). Further experiments revealed that Fas-induced cell death was associated with increased formation of superoxide anions in control neuroglioma cells and in cells overexpressing catalase. These increases were significantly suppressed by Cu/ZnSOD overexpression. These data indicate that Fas-mediated cell death in neuroglioma cells occur, in part, through the production of reactive oxygen species (ROS). These observations also suggest that Fas-induced cell death in these cells occur through apoptosis and necrosis. Thus overexpression of Cu/ZnSOD caused the suppression of both types of Fas-induced cell death whereas catalase prevented apoptotic but not necrotic cell death. These observations are discussed in terms of their support for a role for both peroxides and superoxide radicals in Fas-induced cell death.

VASE-containing N-CAM isoforms are increased in the hippocampus in bipolar disorder but not schizophrenia.

The neural cell adhesion molecule (N-CAM) is a cell recognition molecule that is involved in cellular migration, synaptic plasticity, and CNS development. In schizophrenia, a 105- to 115-kDa N-CAM protein is increased in CSF and in the hippocampus and prefrontal cortex. The variable alternatively spliced exon (VASE) of N-CAM is developmentally regulated and can be spliced into any of the major 120-, 140-, and 180-kDa N-CAM isoforms. We determined that the variable alternative spliced exon of N-CAM (VASE) also is increased in bipolar disorder by quantitative Western immunoblot. VASE immunoreactive proteins (triplet bands around 140 kDa and a single band around 145 kDa) were identified in soluble and membrane brain extracts and quantified in the hippocampus. Soluble VASE 140 kDa was increased in the hippocampus of patients with bipolar disorder as compared to controls, patients with schizophrenia, and suicide cases. Membrane-extracted VASE 140 and 145 kDa were unchanged in the same groups. Multiple 145-kDa VASE-immunoreactive proteins that also reacted to an N-CAM antibody were separated by isoelectric focusing and electrophoresis followed by western immunoblotting; however, the VASE 140-kDa proteins were only weakly N-CAM immunoreactive. By immunohistochemistry, VASE colocalized with GFAP-positive astrocytes in the hippocampus. VASE immunostaining was also observed in the cytoplasm of CA4 pyramidal neurons that were positive for phosphorylated high molecular weight neurofilament and synaptophysin terminals. Thus no differences in VASE were found in patients with schizophrenia, but there was a marked increase of VASE immunoreactive proteins in bipolar disorder. It is possible that abnormal regulation of N-CAM proteins results in differing patterns of abnormal expression in neuropsychiatric disorders.

Stereoselective dysmorphogenicity of the enantiomers of the valproic acid analogue 2-N-propyl-4-pentynoic acid (4-yn-VPA): cross-species evaluation in whole embryo culture.

We previously reported the in vitro differential stereoselective dysmorphogenic potential of the R(+) and S(-) enantiomers of 2n-propyl-4-pentynoic acid (4-yn-VPA) in mice. To determine whether this stereoselectivity is species specific, we evaluated the dysmorphogenic potential of these isomers as well as valproic acid (VPA) to gestational day 9 rat embryos using whole embryo culture (WEC). Aqueous solutions of the sodium salts of R-4-yn-VPA, S-4-yn-VPA, 50%R/ 50%S-4-yn-VPA or VPA were added to the culture medium to give 0, 0.075, 0.15, 0.3, 0.6, or 1.2 mmol/L and embryos were evaluated 48 hr later. The S-4-yn-VPA enantiomer gave clear concentration-dependent dysmorphology as well as effects on developmental score, somite number, crown rump length, and head length. Effects on rotation and defects of the neural tube, somites and heart were observed. Embryolethality was observed only at 1.2 mmol/L concentration. The R-4-yn-VPA enantiomer was neither embryo toxic nor dysmorphogenic at any concentration. VPA significantly reduced all parameters and was dysmorphogenic at the highest concentration but was not embryo lethal. The 50/50 mixture of R- and S-isomers appeared to elicit a degree of embryolethality and dysmorphology similar to VPA. The potency order for the four chemicals was S(-) > S(-)/R(+) = VPA > > > R(+), comparable to that observed in mice by either in vivo or in vitro exposure. These data demonstrate that the stereoselective dysmorphology for these enantiomers can be observed across species and is not related to maternal metabolism.

Overexpression of superoxide dismutase and catalase in immortalized neural cells: toxic effects of hydrogen peroxide.

Hydrogen peroxide (H2O2) is a known toxicant which causes its damage via the production of hydroxyl radicals. It has been reported to cause both necrotic and apoptotic cell death. The present study was undertaken to evaluate the mode of H2O2-induced cell death and to assess if overexpression of catalase could protect against its toxicity. H2O2 causes cell death of immortalized CSM 14.1 neural cells in a dose-dependent manner. H2O2-induced death was associated with DNA laddering as shown by agarose gel electrophoresis. Stable overexpression of catalase by transfection of a vector containing human cDNA into these cells markedly attenuated H2O2-induced toxic effects. Transfection of a vector containing a SOD cDNA afforded no protection. These results indicate that H2O2 can lead to the activation of endonuclease enzyme that breaks DNA into oligosomes. These cells which overexpress catalase or SOD will help to determine the specific role of H2O2 or O2- in the deleterious effects of a number of toxins.

Species- and brain region-specific dopamine transporters: immunological and glycosylation characteristics.

Dopamine transporters (DATs) from the caudate nucleus of four species (rat, mouse, dog, and human) and four regions of rat brain (striatum, nucleus accumbens, prefrontal cortex, and midbrain) were photoaffinity labeled and analyzed by immunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis for cross-reactivity to four epitope-specific rat antipeptide antibodies. Each of these antibodies varied in its efficiency at recognizing DAT. The DATs from the rat brain regions exhibited the same degree of recognition by each of the four sera, a result compatible with these proteins being the product of a single gene. The DATs from the different species were recognized by all four sera but with different efficiencies, possibly relating to amino acid sequence differences within the immunizing epitope. All of the photolabeled, immunoprecipitated DATs migrated with a molecular mass of approximately 80 kDa, and no lower molecular mass forms were found. The DATs from all species and brain regions tested were shown by enzymatic deglycosylation to contain N-linked carbohydrates and sialic acids in amounts comparable with rat striatal DATs. The finding that no photolabeled DAT forms < 80 kDa were isolated from membranes indicates that partially or incompletely glycosylated forms are not present, even in the midbrain cell bodies where immature forms might be expected to be found. These findings verify the utility of these anti-rat antibodies as biochemical tools for studying DATs from other species and extend our knowledge of biochemical characteristics of DATs from these species and brain regions.

Profile of procarbazine-induced embryotoxicity in an embryo hepatocyte co-culture system and after in utero glutathione depletion.

Procarbazine (PCZ) is an antineoplastic agent useful in the treatment of Hodgkin's disease, brain tumors, and chronic leukemia. PCZ is dysmorphogenic to developing embryos exposed in vivo or cultured in the serum of PCZ-treated rats. However, embryos directly cultured with PCZ (up to 400 micrograms/ml) or PCZ plus S-9 liver fractions are unaffected. Since intact liver cells provide several advantages over hepatic subcellular fractions for in vitro bioactivation, we exposed rat embryos to PCZ in an embryo/hepatocyte co-culture system. Gestation day (GD) 9.5 rat embryos exposed to 0, 200, 300, or 400 micrograms PCZ/ml in the presence of untreated or phenobarbital induced male rat hepatocytes failed to display toxicity. However, in a companion study GD 9.5 rat embryos cultured in the serum from PCZ-treated rats exhibited developmental deficiencies. Studies have shown that the formation of toxic metabolites can result from glutathione (GSH) conjugation of toxicants in the liver. Therefore, in a second set of experiments, rat embryos were cultured in serum from rats pretreated with two GSH depleters (phorone and buthionine sulfoximine) and subsequently dosed with PCZ. Effects on development were enhanced when embryos were cultured in the serum from PCZ-treated/GSH depleted rats. These data indicate that PCZ requires in vivo activation to be dysmorphogenic and further suggest that the metabolite(s) responsible for procarbazine embryo-toxicity are formed readily under conditions of low GSH levels. This argues against a glutathione conjugate as the ultimate toxicant.

Toxicokinetics and structure-activity relationships of nine para-substituted phenols in rat embryos in vitro.

The purpose of this study was to examine the toxicokinetics of embryo uptake following exposure to a variety of chemically related phenols in rat embryo culture. The uptake of nine radiolabeled para-substituted phenols by day 10 (9-13 somite stage) rat embryos in vitro was determined from 1 to 42 hrs after being placed in culture media containing various phenols. Uptake was rapid, having a half-life of 3 hr or less, with 7 of the nine compounds having uptake half-times of less than one hour. The equilibrium concentration in the embryo ranged from 53 to 136% of the media concentration, indicating only a factor of 2 in maximum discrimination against the compound for any of the phenols studied. The fraction of radioactivity remaining unbound in the media decreased with increasing log P (octanol/water partition coefficient). The binding was calculated to be 50% for log P = 1.77 from the fitted regression equation and decreased by a factor of 5.9 for every decade increase in P. When hepatocytes were also present in the media the equilibrium concentration in the embryos was less than when hepatocytes were absent. With the limited data, four of the phenols appeared to have no (i.e., zero) equilibrium level when hepatocytes were present. Thus the metabolites produced by the hepatocytes appeared to have less affinity for the embryo than the parent phenol. Toxicodynamic information as given by the effective concentration of the phenol in the embryo to cause somite or tail teratological effects was best predicted by the measured unbound fraction.

Developmental effects of methyl benzimidazolecarbamate following exposure during early pregnancy.

Methyl 2-benzimidazolecarbamate (MBC) and its parent compound benomyl are used as agricultural fungicides. Both chemicals are embryotoxic if administered during organogenesis, and benomyl is teratogenic. Based on a previous study indicating a lack of maternal effects of MBC following exposure during early pregnancy, the current experiments were designed to evaluate the effect of exposure to MBC during early pregnancy on developmental parameters of offspring. Rats were administered MBC at 0, 100, 200, 400, or 600 mg/kg/day during Days 1-8 of pregnancy and killed on Day 11 or Day 20 of gestation. On Day 11, embryos were assessed for survival rate, growth parameters, and anomalies. On Day 20, standard developmental toxicity evaluations were performed. Doses of 200 to 600 mg/kg/day MBC reduced embryonic survival by Day 11; exposure to MBC at 100 to 600 mg/kg/day reduced the number of fetuses surviving on Day 20. Evidence of developmental delay was apparent on Day 11 at all doses, and fetal weight was reduced by Day 20. MBC produced a dose-dependent increase in developmental defects seen on Day 11 and in several malformations observed on Day 20. MBC exposure during the first week of pregnancy was shown to be embryotoxic, resulting in embryonic death, growth retardation, and developmental abnormalities when evaluated on Days 11 or 20 of gestation.

In vitro embryotoxicity of a series of para-substituted phenols: structure, activity, and correlation with in vivo data.

The embryotoxicity of phenol and twelve para-substituted congeners on mid-gestation rat embryos was evaluated in vitro. Through application of correlative procedures and stepwise regression, equations describing the relationship between physical-chemical properties and various measures of activity were developed. Embryotoxicity was quantified by the log of the reciprocal of the potency estimates for reduction in selected growth parameters and induction of four morphological defects. In general, co-cultured hepatocytes ameliorated embryotoxicity; only phenol-induced embryotoxicity was enhanced by the presence of hepatocytes. In the absence of hepatocytes, measures of growth retardation were positively correlated with molar refractivity of the phenols. With hepatocytes, lipophilicity became important for describing the potential to induce growth deficits. The structural defects had varying correlation patterns in both culture systems. Potencies of these congeners in vitro were also compared to maternal and developmental potencies observed in vivo (Kavlock, Teratology, 41:43-59, '90). Two of the congeners were very toxic in both systems. For the remaining congeners, one maternal toxicity measure was found to be positively correlated with embryotoxicity for growth and development in vitro without hepatocytes. With hepatocytes, a broad spectrum of correlations, both positive and negative, were observed between in vivo developmental toxicity endpoints and in vitro embryotoxicity. Data from preliminary dosimetry studies suggest that phenol congeners may accumulate in embryos exposed in vitro more readily than with in vivo exposure. Potency calculations based on dosimetry information may demonstrate better correlations between data and allow additional relationships between chemical structure and activity to be developed.

Interleukin 2, interleukin 2 receptor, and interferon-gamma synthesis and mRNA expression in phorbol myristate acetate and calcium ionophore A23187-stimulated T cells from elderly humans.

The levels of interleukin 2 (IL-2), interleukin 2 receptor (IL-2R), and interferon-gamma (IFN-gamma) specific mRNA and their gene products were examined in phorbol myristate acetate (PMA) and calcium ionophore A23187-costimulated purified T cells from young and elderly humans. In addition, the number of high-affinity IL-2R per activated cell, the high-affinity IL-2R density, and the proliferative response of the cells were measured. Among PMA/A23187-stimulated T cells, there was no statistically significant age-related difference in IL-2 or IL-2R specific mRNA accumulation or in the amount of IL-2 or IL-2R synthesized. IFN-gamma specific mRNA was increased significantly in T cells from elderly individuals and the amount of IFN-gamma synthesized by PMA/A23187-activated T cells was nearly double that produced by cells from young individuals. Quantification of the number of high-affinity IL-2R by [125I]IL-2 binding demonstrated there was no decrease in either the mean number or the dissociation constant of the high-affinity IL-2R on activated T cells of the elderly. Despite producing large amounts of IL-2 and having comparable numbers of both low- and high-affinity IL-2R. PMA/A23187-stimulated T cells from elderly subjects still proliferated less vigorously than did T cells from young persons. The addition of exogenous IL-2 to the cultured cells did not fully correct this age difference. Our findings that the expression of the IL-2, IL-2R, and IFN-gamma genes are not constitutionally defective in the elderly support the hypothesis that the age-related decline in proliferation observed in mitogen-stimulated T cells of the elderly is most likely attributable to alterations in the transmission of signals from the cell membrane to the nucleus.

Abundant alkali-sensitive sites in DNA of human and mouse sperm.

The DNA of human and mouse sperm cells was analyzed by single-cell microgel electrophoresis, by agarose gel electrophoresis, and by alkaline elution--three techniques that can detect single-strand DNA breaks and/or labile sites. Under these conditions a surprisingly large number of single-strand DNA breaks, approximately 10(6) to 10(7) per genome, were detected in human and mouse sperm but not in human lymphocytes or in mouse bone marrow cells. These breaks were also present in chicken erythrocyte DNA, which is also highly condensed. These breaks were not observed under neutral pH conditions nor under denaturing conditions not involving alkali, suggesting that these sites are alkali-sensitive and do not represent preexisting single-strand breaks. The high frequency of such sites in sperm from healthy mouse and human donors suggests that they represent a functional characteristic of condensed chromatin rather than DNA damage.

Expression of interleukin 2 and the interleukin 2 receptor in aging rats.

Lymphocytes of aged animals exhibit a marked decrease in proliferative capacity in response to mitogen stimulation when compared to those of younger animals. In humans and mice the decreased proliferation is due at least in part (i) to the inability of lymphocytes to synthesize sufficient interleukin 2 (IL-2) and (ii) to decreased expression of IL-2 receptors (IL-2R) on the surface of aged lymphocytes. We compared proliferative abilities, IL-2 production, and IL-2R expression in splenocyte cultures of 4- to 5- and 22- to 24-month-old Fischer 344 rats stimulated with either concanavalin A (Con A) or A23187 and phorbol myristate acetate (PMA). Proliferation was significantly decreased in aged lymphocytes (30-50%) with both treatment protocols. However, unlike mice and humans we observed no difference in IL-2 activity, IL-2 mRNA levels, or IL-2R cell surface expression of lymphocytes from young and aged rats stimulated with either Con A or A23187 and PMA. These results indicate that factors other than decreased expression of IL-2 and IL-2R are responsible for the diminished proliferative capacity of aged rat lymphocytes following mitogen stimulation.

Decreased proliferation, interleukin 2 synthesis, and interleukin 2 receptor expression are accompanied by decreased mRNA expression in phytohemagglutinin-stimulated cells from elderly donors.

Using cDNA probes to human interleukin 2 (IL2) and interleukin 2 receptor (IL2R), the amount of IL2 and IL2R mRNA produced by PHA stimulated peripheral blood mononuclear cells from young (less than 40 yr) and old (greater than 60 yr) donors was quantitated. Stimulated cell cultures from each individual were also examined for proliferative ability, expression of membrane IL2R, membrane IL2R density, and for the amount of IL2R shed into the culture supernatant. Induction of IL2 and IL2R mRNAs were decreased in cells from elderly individuals, as were the levels of IL2 secretion, the percentage of IL2R+ T cells and the density of membrane IL2R per cell. The results suggest that decreased expression of both IL2 and IL2R mRNA contributes to the low synthesis of IL2 and membrane IL2R, respectively, and is partially responsible for the diminished proliferative activity observed in lymphocytes from the elderly.