Logo of narLink to Publisher's site
Nucleic Acids Res. Dec 15, 2002; 30(24): 5529–5538.
PMCID: PMC140061

Genome-wide analysis of mRNA decay in resting and activated primary human T lymphocytes


We used microarray technology to measure mRNA decay rates in resting and activated T lymphocytes in order to better understand the role of mRNA decay in regulating gene expression. Purified human T lymphocytes were stimulated for 3 h with medium alone, with an anti-CD3 antibody, or with a combination of anti-CD3 and anti-CD28 antibodies. Actinomycin D was added to arrest transcription, and total cellular RNA was collected at discrete time points over a 2 h period. RNA from each point was analyzed using Affymetrix oligonucleotide arrays and a first order decay model was used to determine the half-lives of approximately 6000 expressed transcripts. We identified hundreds of short-lived transcripts encoding important regulatory proteins including cytokines, cell surface receptors, signal transduction regulators, transcription factors, cell cycle regulators and regulators of apoptosis. Approximately 100 of these short-lived transcripts contained ARE-like sequences. We also identified numerous transcripts that exhibited stimulus-dependent changes in mRNA decay. In particular, we identified hundreds of transcripts whose steady-state levels were repressed following T cell activation and were either unstable in the resting state or destabilized following cellular activation. Thus, rapid mRNA degradation appears to be an important mechanism for turning gene expression off in an activation-dependent manner.


In diverse eukaryotic organisms ranging from yeast to humans, control of mRNA turnover plays a key role in regulating cellular responses to environmental stimuli (1,2). Following transcriptional activation, for example, the regulated decay of mammalian immediate early response gene transcripts, including c-fos, c-jun and c-myc, is crucial for normal cellular functions such as cell cycle progression, proliferation and apoptosis (3). Aberrant regulation of decay leads to oncogenic activation and malignancy (37). In T lymphocytes, T cell receptor (TCR) stimulation induces the expression of numerous early response genes. Many of these genes, including cytokine genes and proto-oncogenes, produce mRNA transcripts that exhibit rapid degradation, but subsets of these short-lived transcripts can undergo differential regulation. For example, CD28 co-stimulation of TCR-activated T lymphocytes leads to specific stabilization of cytokine transcripts, including interleukin-2 (IL-2), granulocyte-macrophage colony stimulating factor (GM-CSF), tumor necrosis factor (TNF)-α and interferon (IFN)-γ, while proto-oncogene transcripts such as c-myc remain unstable (8). Thus, the decay of an individual mRNA transcript can exhibit gene-specific, stimulus-dependent regulation that impacts the overall expression of the gene.

Although increasing information suggests that mRNA degradation is an important control point for regulating T lymphocyte gene expression, mRNA decay rates have been measured for only a small number of T lymphocyte mRNA transcripts. Recently developed microarray technology has revolutionized gene expression research, allowing the expression of thousands of genes to be simultaneously profiled in different cell types or different treatment conditions. The vast majority of experiments involving microarray technology have evaluated only steady-state mRNA levels. Recent work, however, suggested that microarray technology can be used to categorize mRNA transcripts based on their mRNA decay rates (9,10). In the present study, microarray technology was used to quantitatively measure on a genome-wide basis the decay rates of mRNA transcripts in resting and activated primary human T lymphocytes following transcriptional arrest. The half-life and 95% confidence interval (CI) was determined for each of approximately 6000 transcripts expressed in T lymphocytes. This approach allowed the identification of hundreds of T lymphocyte genes that are regulated at the level of mRNA degradation.


Purification of human T lymphocytes

Human T lymphocytes were purified as described previously (11). Briefly, human peripheral blood mononuclear cells were isolated through a Ficoll-Hypaque (Amersham Biosciences) cushion from buffy coat white blood cell packs (American Red Cross) and were then passed through T cell enrichment columns (R&D Systems). Purified cells consisted of 90–95% CD3+ T lymphocytes based on flow cytometry analysis.

T lymphocyte stimulation and RNA isolation following actinomycin D treatment

Purified T lymphocytes were cultured overnight in RPMI 1640 (Life Technologies Inc.) supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 100 U/ml penicillin G and 100 µg/ml streptomycin. Cells (25 000 000–50 000 000 cells/ group) were then stimulated for 3 h with medium alone or with immobilized monoclonal antibodies (1.0 µg/ml) directed against the CD3 component of the TCR complex (αCD3) (R&D Systems) or a combination of αCD3 and a monoclonal antibody directed against the CD28 co-stimulatory molecule (αCD28) (R&D Systems) as described previously (11). Actinomycin D (Act D) (Sigma Corp.) was then added to a final concentration of 10 µg/ml and total cellular RNA was isolated at discrete time points over a 2 h period using Trizol reagent (Life Technologies). Four independent experiments were performed. After addition of Act D, RNA was isolated at 0, 45, 90 and 120 min time points in two experiments, at 0 and 120 min time points in one experiment and at 0 and 90 min time points in one experiment.

Microarray hybridizations

cDNA was synthesized from 10–15 µg total RNA using the Superscript II RT cloning kit (Life Technologies). This cDNA was used to synthesize biotin-labeled cRNA in an in vitro transcription reaction using a commercially available kit (Enzo Diagnostics). cRNA was purified with the RNeasy Mini-kit (Qiagen) and 15 µg cRNA was used for hybridization to human U95Av2 arrays (Affymetrix Inc.), according to the manufacturer’s protocol. Quantitative scanning of arrays was done on an HP Agilent 2200 confocal scanner.

Microarray data analysis

Affymetrix Microarray Software Suite (v.4.0) was used to calculate hybridization intensities, referred to as average difference (AD), for each gene present on the microarray and to determine if transcripts corresponding to these genes were ‘present’ or ‘absent’. After scaling the average intensity of all arrays to 1000, the AD values for each target transcript on each array was normalized to values for GAPDH on the same array. The scaled, normalized AD values at each time point following addition of Act D were used to estimate the transcript half-life (t½) based on a first order decay model using the equation, y = β0 eβ1t + ε, where y is the normalized AD value at time t following the addition of Act D, β0 is the initial intensity, β1 is a decay parameter related to half-life (t½ = –ln2/β1) and ε is an error term. This first order decay model, combining data from all four experiments, was used to determine probability distributions for the initial hybridization intensity (β0) and for the half-life of each transcript under each stimulation condition. A detailed description of the statistical analysis can be found in Supplementary Material.

Northern blot and real time RT–PCR

Purified human T lymphocytes were stimulated for 3 h with medium or αCD3+αCD28, Act D was added, and total cellular RNA was harvested at 0, 45, 90 and 120 min time points. For northern blot analysis, 10 µg total RNA from each sample was separated by electrophoresis on a 1% glyoxal agarose gel using the NorthernMax™-Gly Glyoxal-Based System (Ambion). The RNA was blotted onto Brightstar-Plus membranes (Ambion), and membranes were crosslinked with UV energy and hybridized for 16–18 h with 32P-labeled GAPDH, MAD-3, TNF superfamily member 14 (TNFSF14) and p27kip1 probes. The GAPDH probe was generated using the DECAprime™ II Random Priming DNA Labeling Kit (Ambion) according to the manufacturer’s protocol. The MAD-3 probe was generated by end-labeling a DNA oligonucleotide containing the sequence 5′-GCCCCTTTGCACTCATAACGTCAGACGCTGGCCTCCAAACACACAGTCATCATAGGGC-3′) and the TNFSF14 probe was generated by end-labeling the DNA oligonucleotide 5′-GGCACCCTCTGAGTTCTCCACGTGTCAGACCCATGTCCAATGCACCACGCTCC-3′. End-labeling reactions were performed using a KinaseMax™ 5′ End-Labeling Kit (Ambion) according to the manufacturer’s directions. The p27kip1 probe was generated by RT–PCR. cDNA was generated using the ProSTAR™ Ultra HF RT–PCR System (Stratagene) using total RNA from purified human T cells. PCR was then performed using p27kip1 specific forward and reverse primers (5′-TTCAGACGGTTCCCCAAAT-3′ and 5′-AACGCTTTTAGAGGCAGATCA-3′). The PCR product was gel purified and used as a template for an additional PCR in the presence of [α-32P]ATP. The hybridized blots were washed, and then quantified using a phosphorimager (Molecular Dynamics). The hybridization intensity of each transcript was normalized to GAPDH and the normalized values were used to calculate half-lives.

For real time RT–PCR, cDNA was synthesized from total cellular RNA and used for PCR using the Brilliant™ Two-Step Quantitative RT–PCR Core Reagent Kit (Stratagene) according to the manufacturer’s instructions. Dye-labeled TaqMan probes were synthesized by PE Applied Biosystems and the oligonucleotide primers were synthesized by Integrated DNA Technologies Inc. The probe and primer sequences for each gene evaluated are listed below in the following order: sense primer, probe, antisense primer. IL-2: GAATCCCAAACTCACCAGGA, ACCTCTGGAGGAAGTGCTGAATTTAGCTCA, ATGGTTGCTGTCTCATCTGC; GM-CSF: CAG CCTCACCAAGCTCAAG, ACTTCCTGTGCAACCCAGATTATCACCTTT, AAGGGGATGACAAGCAGAAA; tat response element DNA-binding protein (TARDBP): GGGGATGTGATGGATGTCTT, TCATATATCCAATGCCGAACCTAAGCACAA, CCACCTGGATTACCACCAAA; c-myc: TCGGATTCTCTGCTCTCCTC, AGCGACTCTGAGGAGGAACAAGAAGATGAG, CTCTGACCTTTTGCCAGGAG; phospholipase C β2 (PLCB2): ACAACTCCCACATCCAGGAA, GAACAGATACGGGAGATGGAAAAGCAGTTC, CTTCACCTCTGCCTCCAGAC; hypoxanthine phosphoribosyl transferase (HPRT): GGTGAAAAGGACCCCACGAA, TGTTGGATTTGAAATTCCAGACAAGTTTGT, AGTCAAGGGCATATCCAACA. The internal oligonucleotides for IL-2, GM-CSF, TARDBP and PLCB2 TaqMan® probes were labeled with a 5′ reporter dye, 6-carboxyfluorescein (6FAM), and a 3′ quencher dye, 6-carboxytetramethyl rhodamine (TAMRA). The control HPRT probe was labeled with a 5′ reporter dye, tetrachloro-6-carboxyfluorescein (TET), and a 3′ quencher dye, TAMRA. PCR amplification was carried out using a Cepheid® Smart Cycler thermocycler and analyzed using Smart Cycler software. Standard curves for each gene were generated to determine the relative concentrations of amplified transcripts. The concentration of each transcript was then normalized to HPRT and the normalized values were used to calculate half-lives.


Steady-state mRNA levels in resting and activated T lymphocytes

Purified human T lymphocytes were stimulated in four independent experiments with medium alone (resting), with immobilized αCD3 or with a combination of immobilized αCD3 and αCD28 (αCD3+αCD28) antibodies. αCD3 activates the TCR complex, providing partial cellular activation, and αCD28 provides additional co-stimulation, allowing complete T cell activation. After 3 h of stimulation, Act D was added to arrest transcription and total cellular RNA was isolated at discrete time points over a 2 h period. This time course was designed to evaluate early changes in gene expression and the role of mRNA degradation in regulating early gene expression events. Transcripts were considered to be present at the initial time point (at the time of Act D addition) under each stimulation condition, if the Affymetrix software determined them to be present in three of the four experiments. Of 12 625 genes represented on the arrays, transcripts from 5296, 5541 and 5497 genes were present at this initial time point in the medium, αCD3 and αCD3+ αCD28 groups, respectively. A total of 6112 transcripts were present in one or more of these stimulation conditions. The correlation coefficient comparing initial AD values between two different experiments exceeded 0.98 within each of the stimulation groups.

For each expressed transcript, an initial hybridization intensity parameter called β0 was computed based on our first order decay model using data from all four experiments (see Materials and Methods). The correlation between the β0 median values and the initial Affymetrix AD values within an experiment exceeded 0.96. The initial hybridization intensity values (β0) for each stimulation condition were used to identify transcripts that were induced or repressed following cellular activation. The number of transcripts induced greater than 5-fold by αCD3 and αCD3+αCD28 were 81 and 98, respectively (P < 0.05 for each), whereas the number of transcripts repressed at least 5-fold by αCD3 and αCD3+ αCD28 were 107 and 95, respectively (P < 0.05 for each). We also identified 80 transcripts that were induced by CD28 co-stimulation (P < 0.05) by comparing the initial hybridization intensities (β0) after αCD3 versus αCD3+αCD28 stimulation. The transcripts induced by CD28 co-stimulation included cytokine genes (see Table Table1),1), in accord with previously reported results (8). Interestingly, we found almost no transcripts whose expression was repressed by CD28 co-stimulation in a statistically significant manner. Table Table33 shows a subset of the transcripts induced or repressed by αCD3+αCD28 stimulation (P < 0.05 for each). More complete listings of transcripts that were induced or repressed following T lymphocyte activation are included in Supplementary Material. The complete data set from this study can be found on the web at http://web.ahc.umn.edu/~bohjanen/.

Table 1.
Comparison of T lymphocyte transcript half-life values obtained by microarray analysis or northern blot
Table 3.
Short-lived transcripts that were induced (P < 0.05) or repressed (P < 0.05) upon αCD3+αCD28 stimulation

Identification of short-lived transcripts and stimulus-induced changes in mRNA decay

We next computed mRNA decay rates for the transcripts found to be present under each stimulation condition. The median half-life and 95% CI were calculated for each expressed transcript under each stimulation condition and can be found at http://web.ahc.umn.edu/~bohjanen/. We compared T lymphocyte transcript half-lives determined using microarrays to half-lives reported in the literature that were obtained using northern blots (see Table Table1)1) and found a good correlation. We also used northern blots to measure half-lives following Act D treatment of three transcripts that had not been previously reported in the literature and again found good correlation with our microarray data (Fig. (Fig.1).1). In addition, we analyzed five transcripts by real time RT–PCR and found that the half-lives correlated well with half-lives determined using microarrays (Table (Table2).2). Figure Figure2A2A shows the percentage of transcripts expressed under each stimulation condition that had median half-lives falling within the indicated ranges as determined using microarrays. The largest proportion of transcripts had long half-lives (>6 h) with a much smaller proportion of transcripts having very short half-lives. Partial listings of transcripts with short half-lives are shown in Tables Tables33 and and4,4, and a more complete listing is shown in Supplementary Material. Short-lived transcripts exhibited three dominant patterns of gene expression: activation-induced transcripts with short half-lives, short-lived transcripts that were repressed by activation, and stable transcripts that were destabilized and repressed by activation.

Figure 1
Comparison of transcript half-lives determined by northern blot or microarrays. Purified human T lymphocytes were stimulated for 3 h with medium or αCD3+αCD28. Act D was added and total cellular RNA was then isolated at the 0, ...
Figure 2
Profile of T lymphocyte transcript half-lives. (A) Purified human T lymphocytes were stimulated for 3 h with medium, αCD3 or αCD3+αCD28. Act D was added and total cellular RNA was isolated at discrete time points over ...
Table 2.
Comparison of mRNA decay rates determined using microarrays or real time RT–PCR
Table 4.
Transcripts that were repressed (P < 0.05) >2.5-fold and were destabilized (P < 0.05) or became absent upon stimulation with αCD3+αCD28

A large proportion of transcripts that were induced by T lymphocyte activation had very short half-lives (Fig. (Fig.2B).2B). We identified 81 and 98 activation-induced transcripts with half-lives of <60 min following αCD3 and αCD3+αCD28 stimulation, respectively (see Supplementary Material). The pattern of gene induction and mRNA decay of a subset of these short-lived transcripts is shown in Figure Figure33 (Induced). Overall, αCD3 stimulation and αCD3+αCD28 stimulation induced similar sets of transcripts that exhibited rapid decay. Many αCD3+αCD28-induced transcripts were more abundant, however, and some exhibited longer half-lives (Fig. (Fig.33 and Supplementary Material). Half-life data for several short-lived transcripts that were induced by αCD3+αCD28 stimulation are shown in Table Table33 (top). This set of transcripts included transcripts encoding important regulatory proteins such as cytokines, signal transduction regulators, transcription factors and regulators of apoptosis.

Figure 3
Short-lived transcripts whose steady-state levels were induced or repressed upon T cell activation. Purified human T lymphocytes were stimulated for 3 h with medium or αCD3+αCD28. Act D was added and total cellular RNA was isolated ...

A second predominant pattern of expression included short-lived transcripts whose steady-state levels were repressed following αCD3 or αCD3+αCD28 stimulation (see Fig. Fig.3,3, Repressed). Stimulation with αCD3 or αCD3+αCD28 led to a 2.5-fold or greater repression of 114 or 100 transcripts with half-lives of <60 min, respectively (P < 0.05 for each). The set of genes repressed by αCD3 and αCD3+αCD28 were very similar. A subset of the short-lived transcripts whose steady-state levels were repressed by αCD3+αCD28 stimulation is shown in Table Table33 (bottom). Many of these short-lived transcripts also encode important regulatory proteins, including transcription factors and cell cycle regulators. The rapid mRNA decay exhibited by these transcripts may be essential for their decreased expression upon cellular activation.

By comparing mRNA decay rates in resting and activated T lymphocytes, we identified 212 transcripts that were repressed at least 2-fold (P < 0.05) and also destabilized (P < 0.05) by αCD3+αCD28 stimulation. A very similar set of transcripts was also repressed and destabilized by αCD3 stimulation. In addition, 166 transcripts were identified that were relatively stable (half-life > 120 min) in resting cells but presumably became unstable following cellular activation because they became absent after αCD3 or αCD3+αCD28 stimulation (see Supplementary Material). Table Table44 shows a subset of the transcripts that appeared to be destabilized by αCD3+αCD28 stimulation. This group of transcripts also encoded important regulatory proteins, including cell surface receptors, signal transduction mediators, transcription regulators, and regulators of cell growth or death.

Cytokine transcripts were previously reported to be stabilized by CD28 co-stimulation compared to αCD3 stimulation alone (8). Our data showed that transcripts from αCD3-stimulated cells had similar mRNA decay profiles to αCD3+αCD28-stimulated cells, with notable exceptions. For example, IL-2 and GM-CSF transcripts were stabilized by CD28 co-stimulation in a statistically significant manner (Table (Table1),1), in agreement with previous reports (8). CD28 co-stimulation also led to statistically significant (P < 0.05) stabilization of 18 other activation-induced genes, including the chemokines exodus-1 and lymphotactin, the signal transduction regulators JAK-binding protein/TIP3 and Ras-like protein Tc4, the transcription factors NF-κB, ERF and CREM and the apoptosis regulator IPL (Table (Table5).5). Transcripts encoding other cytokines, such as IFN-γ and TNF-α, also showed a trend toward stabilization by CD28 co-stimulation, but this effect did not reach statistical significance (Table (Table11).

Table 5.
Transcripts that were induced by activation (P < 0.05) and were stabilized by αCD3+αCD28 stimulation in comparison to αCD3 stimulation alone

Identification of short-lived transcripts that contain AU-rich element-like sequences

Perhaps the best characterized example of a cis-element that controls mRNA degradation is the AU-rich element (ARE). AREs present in the 3′-untranslated regions (3′-UTRs) of cytokine and proto-oncogene transcripts mediate their rapid decay (1,2,1215). By comparing sequences from known ARE-containing transcripts, a minimal consensus ARE was recently defined as WWWUAUUUAUWWW (W = U or A) (16). These researchers searched GenBank for human transcripts that contained this consensus sequence in their 3′-UTRs and found 897 transcripts that were compiled into an ARE database (16). The ARE-like sequences present in >90% of the transcripts in the ARE database, however, have not been evaluated for function. We searched for the intersection between transcripts that we found to be expressed in T cells and transcripts present in the ARE database, and we found approximately 400 transcripts that were expressed in T cells and contained ARE-like sequences. Of these, ~25% (~100 transcripts) exhibited rapid decay (half-life <60 min) under at least one stimulation condition, 45% exhibited intermediate decay (half-life 120–180 min) under at least one condition but did not exhibit rapid decay under any condition, and 30% exhibited relatively slow decay (half-life >180 min). Half-life data for a subset of the transcripts that contained ARE-like sequences and exhibited rapid decay is shown in Table Table6,6, and a more complete listing is shown in Supplementary Material. Interestingly, some of these transcripts were induced by T cell activation while others were repressed. The ARE-like sequences present in these short-lived transcripts are good candidates for functional cis-elements that control decay, but experiments, including mutational analysis or expression of the cis-elements in a heterologous context, would need to be performed to verify their function. Overall, our findings suggest that a major subset of short-lived transcripts contain ARE-like sequences. Many other short-lived transcripts, however, do not contain defined ARE or ARE-like sequences and many transcripts that contain ARE-like sequences do not exhibit rapid decay, suggesting that other signals that have not been identified also regulate decay. A complete listing of the transcripts expressed in T cells that contain ARE-like sequences, along with data regarding their expression and half-lives, can be found at http://web.ahc.umn.edu/~bohjanen/.

Table 6.
Examples of short-lived T cell transcripts that contain ARE-like sequences and were either induced (P < 0.05) or repressed (P < 0.05) upon αCD3+αCD28 stimulation


We used microarray technology to measure decay rates of approximately 6000 T lymphocyte mRNA transcripts, allowing us to identify hundreds of genes that are regulated at the level of mRNA decay. In particular, we identified important regulatory genes that produce unstable transcripts or transcripts that were stable in resting cells but were destabilized in a stimulus-dependent manner. Since steady-state mRNA levels are determined by the rate of transcription as well as the rate of mRNA degradation, rapid mRNA degradation provides the cell with a general mechanism for rapidly turning off the expression of regulatory genes in response to changes in transcription. In addition, activation-induced mRNA destabilization provides another mechanism for turning gene expression off.

Activation-induced genes in T lymphocytes tend to be expressed during precise periods of time, and then their expression is turned off (1721). Our results, as well as the results of others (8,2229), suggest that many activation-induced genes produce transcripts with short half-lives. The precise program of transient gene expression following cellular activation requires precise coordination between transcription and mRNA decay. For example, certain T lymphocyte early response genes, including cytokine genes and proto-oncogenes, are induced transcriptionally but appear to be turned off, at least in part, through rapid mRNA decay (8). Our data suggest that many other regulatory transcripts are also induced transcriptionally and then turned off through mRNA degradation. We cannot, however, exclude the possibility that mRNA stabilization contributes to the induction of these transcripts since many are expressed at low or undetectable levels in the resting state and their half-lives cannot be measured. A combination of decreased transcription, rapid mRNA decay and rapid protein turnover may all contribute to turning off the expression of specific genes.

Turning off expression of activation-induced genes through rapid mRNA decay may be part of a homeostatic mechanism to stop T cell activation and thereby down-modulate an immune response. For example, turning off expression of cytokine genes (see Table Table1)1) through rapid mRNA decay would serve to down-modulate an immune response and prevent excessive inflammatory destruction. Down-modulation of co-stimulatory cell surface receptors such as CD69 (30) or signaling lymphocytic activation molecule (31) may contribute to limiting an immune response (see Table Table3,3, top). Also, turning off expression of induced pro-apoptotic regulators such as Fas ligand (32) would prevent excessive killing of target or bystander cells. Induction of the anti-apoptotic Bcl-2-related protein Bfl-1 (33,34) may allow T cells to initially survive and perform their effector functions and then turning off this gene at a later point through mRNA decay would promote apoptotic death of T cells and thereby limit further inflammation. Also, activation-induced genes (21,35) encoding signal transducers such as JAK-binding protein/TIP3, and MKKK8 or transcription factors such as c-Myc, c-Fos, Jun B, interferon regulatory factor-4 and activating transcription factor-3 (Tables (Tables11 and and3)3) appeared to be turned off, in part, through rapid mRNA degradation. Turning off expression of these genes may bring an activated T cell back toward an inactive state, further limiting an immune response. Failure to turn off activation-induced gene expression through mRNA degradation could lead to severe intracellular consequences. For example, abnormal stabilization of activation-induced transcripts such as c-fos or c-myc has been associated with malignancy (47).

Hundreds of transcripts were identified whose steady-state levels decreased following T lymphocyte activation. A subset of these repressed transcripts had short half-lives in the resting state (Table (Table3,3, bottom). The finding that steady-state levels of these transcripts were maintained in unstimulated cells, despite their rapid rate of decay, suggests that the rapid decay was balanced by rapid transcription. Apparently, the cell expends considerable metabolic energy to constantly synthesize and degrade these transcripts in order to maintain a steady state. This balanced state may exist so that these transcripts are poised to be regulated quickly. For example, cellular activation may lead to decreased transcription of these transcripts, shifting the balance toward decreased steady-state levels. Such a mechanism would allow genes to be turned off rapidly under the appropriate stimulation conditions. Another subset of activation-repressed transcripts exhibited activation-dependent mRNA destabilization (Table (Table4).4). Expression of these transcripts could potentially be turned off rapidly, even in the face of continued transcription.

It appears that cellular quiescence is an actively maintained state and turning off the expression of genes that actively maintain quiescence is a critical component of normal cellular activation (36,37). We found that transcripts encoding many important regulators of cell proliferation or cell fate were turned off following cellular activation, at least in part, through rapid mRNA degradation (see Table Table3,3, bottom, and Table Table4).4). These transcripts encoded cell growth regulators such as the ubiquitous Kruppel-like factor, cell cycle checkpoint control protein, anaphase promoting complex, cyclin D3, cyclin T2, cyclin-dependent kinase 8 and retinoblastoma-like protein 2, as well as apoptosis regulators such as autophagy 5-like protein (APG5), Kruppel-type zinc finger protein, Drak1 and caspase-6. We also found that transcripts encoding components of receptor signaling pathways (21,35,38), including vav1, janus kinase 2, phospholipase C β1, phospholipase C β3, MEK kinase and p56lck, were destabilized and repressed after activation, perhaps as part of a homeostatic feedback loop to shut off TCR-mediated signaling. Rapid mRNA degradation also contributed to the activation-induced shut-off of transcripts encoding cell surface molecules, including α5 integrin, α6B integrin, δ2 catenin, chemokine C-X3-C receptor-1, IL-10 receptor α and IL-10 receptor β. Perhaps at the site of inflammation, activated T cells no longer require signals from these receptors for homing, chemotaxis or differentiation (3942). Alternatively, turning off expression of these receptors could prevent additional T cell signaling and thereby help to down-modulate an immune response.

Although we identified approximately 100 transcripts that were stabilized by αCD3+αCD28 stimulation, few of these transcripts displayed a significant increase in steady-state level at the time point studied. We speculate that activation-induced stabilization could have effects on steady-state mRNA levels at later time points following αCD3+αCD28 stimulation. At least at the early time point studied, activation-induced mRNA destabilization produced much more dramatic effects on steady-state mRNA levels than did mRNA stabilization. We also identified numerous short-lived transcripts whose steady-state levels did not change after 3 h of activation with αCD3 or αCD3+αCD28. Perhaps these transcripts are poised to be rapidly regulated at other time points or under environmental conditions not examined in our experiments. Our experiments evaluated mRNA decay at only a single point in time, 3 h after T cell activation, and therefore dynamic changes in mRNA decay that occur at other time points could have been missed. Future experiments will be performed at a variety of time points following T cell activation in order to better profile the dynamic changes in mRNA decay that may occur.

TCR-mediated stimulation without CD28 co-stimulation results in defective proliferation and decreased survival (43,44). CD28 exerts its effect in large part by inducing increased expression of IL-2, a critical T cell growth factor (45,46). CD28 co-stimulation leads to increased IL-2 transcription as well as stabilization of IL-2 mRNA (8,47). We also found that CD28 co-stimulation led to stabilization of IL-2 mRNA (see Table Table1).1). In addition, we identified several other transcripts that were stabilized by CD28 co-stimulation, including transcripts encoding the chemokines exodus-1 and lymphotactin, the signal transduction regulators JAK-binding protein/TIP3 and Ras-like protein Tc4, the transcription factors NF-κB, ets-domain protein ERF, cAMP response element modulator, and the apoptosis regulator imprinted in liver and placenta (IPL). Stabilization of these transcripts could contribute to CD28-mediated cellular activation and proliferation.

Many of the short-lived transcripts that we identified contain AREs, well characterized mRNA sequences that target transcripts for rapid degradation (1,2,1215). AREs, found at the 3′ end of many rapidly degraded transcripts, function through their interaction with ARE-binding proteins to promote the deadenylation and degradation of mRNA (11,4856). ARE-binding proteins may regulate mRNA degradation through their interaction with the exosome, a multi-protein complex that has 3′→5′ exonuclease activity (57). A database of transcripts containing ARE or ARE-like sequences has been compiled (16) and many of the short-lived transcripts that we identified contain these sequence motifs (Table (Table6).6). Many short-lived transcripts, however, do not contain characterized ARE sequences, therefore other sequences in these transcripts appear to regulate their rapid decay. Also, many transcripts that contain ARE-like sequences do not exhibit rapid decay, suggesting that the presence of an ARE-like sequence is not sufficient for mediating rapid mRNA decay and that additional signals may be required. Identification of transcripts that are regulated at the level of mRNA decay is a first step toward identifying additional novel regulatory sequence elements and trans-acting proteins, with the goal of understanding the biochemical mechanisms that regulate rapid or stimulus-dependent mRNA decay.


Supplementary Material is available at NAR Online.

[Supplementary Material]


We thank Marc Jenkins, Ashley Haase, Vivek Kapur and Arkady Khodursky for critically reading this manuscript. This work was supported by grant R01-AI49494 from the NIH and by an award to the University of Minnesota Medical School under the Research Resources Program of the Howard Hughes Medical Institute.


1. Malter J.S. (1998) Posttranscriptional regulation of mRNAs important in T cell function. Adv. Immunol., 68, 1–49. [PubMed]
2. Ross J. (1995) mRNA stability in mammalian cells. Microbiol. Rev., 59, 423–450. [PMC free article] [PubMed]
3. Pearson P.L. and Van der Luijt,R.B. (1998) The genetic analysis of cancer. J. Intern. Med., 243, 413–417. [PubMed]
4. Raymond V., Atwater,J.A. and Verma,I.M. (1989) Removal of an mRNA destabilizing element correlates with the increased oncogenicity of proto-oncogene fos. Oncogene Res., 5, 1–12. [PubMed]
5. Meijlink F., Curran,T., Miller,A.D. and Verma,I.M. (1985) Removal of a 67-base-pair sequence in the noncoding region of protooncogene fos converts it to a transforming gene. Proc. Natl Acad. Sci. USA, 82, 4987–4991. [PMC free article] [PubMed]
6. Aghib D.F., Bishop,J.M., Ottolenghi,S., Guerrasio,A., Serra,A. and Saglio,G. (1990) A 3′ truncation of MYC caused by chromosomal translocation in a human T-cell leukemia increases mRNA stability. Oncogene, 5, 707–711. [PubMed]
7. Bernasconi N.L., Wormhoudt,T.A. and Laird-Offringa,I.A. (2000) Post-transcriptional deregulation of myc genes in lung cancer cell lines. Am. J. Respir. Cell Mol. Biol., 23, 560–565. [PubMed]
8. Lindsten T., June,C.H., Ledbetter,J.A., Stella,G. and Thompson,C.B. (1989) Regulation of lymphokine messenger RNA stability by a surface-mediated T cell activation pathway. Science, 244, 339–343. [PubMed]
9. Lam L.T., Pickeral,O.K., Peng,A.C., Rosenwald,A., Hurt,E.M., Giltnane,J.M., Averett,L.M., Zhao,H., Davis,R.E., Sathyamoorthy,M. et al. (2001) Genomic-scale measurement of mRNA turnover and the mechanisms of action of the anti-cancer drug flavopiridol. Genome Biol., 2, RESEARCH0041. [PMC free article] [PubMed]
10. Wang Y., Liu,C.L., Storey,J.D., Tibshirani,R.J., Herschlag,D. and Brown,P.O. (2002) Precision and functional specificity in mRNA decay. Proc. Natl Acad. Sci. USA, 99, 5860–5865. [PMC free article] [PubMed]
11. Raghavan A., Robison,R.L., McNabb,J., Miller,C.R., Williams,D.A. and Bohjanen,P.R. (2001) HuA and tristetraprolin are induced following T cell activation and display distinct but overlapping RNA binding specificities. J. Biol. Chem., 276, 47958–47965. [PubMed]
12. Chen C.Y. and Shyu,A.B. (1995) AU-rich elements: characterization and importance in mRNA degradation. Trends Biochem. Sci., 20, 465–470. [PubMed]
13. Jacobson A. and Peltz,S.W. (1996) Interrelationships of the pathways of mRNA decay and translation in eukaryotic cells. Annu. Rev. Biochem., 65, 693–739. [PubMed]
14. Shaw G. and Kamen,R. (1986) A conserved AU sequence from the 3′ untranslated region of GM-CSF mRNA mediates selective mRNA degradation. Cell, 46, 659–667. [PubMed]
15. Xu N., Chen,C.Y. and Shyu,A.B. (1997) Modulation of the fate of cytoplasmic mRNA by AU-rich elements: key sequence features controlling mRNA deadenylation and decay. Mol. Cell. Biol., 17, 4611–4621. [PMC free article] [PubMed]
16. Bakheet T., Frevel,M., Williams,B.R., Greer,W. and Khabar,K.S. (2001) ARED: human AU-rich element-containing mRNA database reveals an unexpectedly diverse functional repertoire of encoded proteins. Nucleic Acids Res., 29, 246–254. [PMC free article] [PubMed]
17. Granucci F., Castagnoli,P.R., Rogge,L. and Sinigaglia,F. (2001) Gene expression profiling in immune cells using microarray. Int. Arch. Allergy Immunol., 126, 257–266. [PubMed]
18. Ellisen L.W., Palmer,R.E., Maki,R.G., Truong,V.B., Tamayo,P., Oliner,J.D. and Haber,D.A. (2001) Cascades of transcriptional induction during human lymphocyte activation. Eur. J. Cell Biol., 80, 321–328. [PubMed]
19. Ollila J. and Vihinen,M. (1998) Stimulation of B and T cells activates expression of transcription and differentiation factors. Biochem. Biophys. Res. Commun., 249, 475–480. [PubMed]
20. Crabtree G.R. (1989) Contingent genetic regulatory events in T lymphocyte activation. Science, 243, 355–361. [PubMed]
21. Ullman K.S., Northrop,J.P., Verweij,C.L. and Crabtree,G.R. (1990) Transmission of signals from the T lymphocyte antigen receptor to the genes responsible for cell proliferation and immune function: the missing link. Annu. Rev. Immunol., 8, 421–452. [PubMed]
22. Borger P., Kauffman,H.F., Vijgen,J.L., Postma,D.S. and Vellenga,E. (1996) Activation of the cAMP-dependent signaling pathway downregulates the expression of interleukin-3 and granulocyte-macrophage colony-stimulating factor in activated human T lymphocytes. Exp. Hematol., 24, 108–115. [PubMed]
23. Cerdan C., Martin,Y., Courcoul,M., Mawas,C., Birg,F. and Olive,D. (1995) CD28 costimulation up-regulates long-term IL-2R beta expression in human T cells through combined transcriptional and post-transcriptional regulation. J. Immunol., 154, 1007–1013. [PubMed]
24. Geginat J., Clissi,B., Moro,M., Dellabona,P., Bender,J.R. and Pardi,R. (2000) CD28 and LFA-1 contribute to cyclosporin A-resistant T cell growth by stabilizing the IL-2 mRNA through distinct signaling pathways. Eur. J. Immunol., 30, 1136–1144. [PubMed]
25. Lake R.A., O’Hehir,R.E., Verhoef,A. and Lamb,J.R. (1993) CD28 mRNA rapidly decays when activated T cells are functionally anergized with specific peptide. Int. Immunol., 5, 461–466. [PubMed]
26. Malter J.S., Reed,J.C. and Kamoun,M. (1988) Induction and regulation of CD2 mRNA in human lymphocytes. J. Immunol., 140, 3233–3236. [PubMed]
27. Santis A.G., Lopez-Cabrera,M., Sanchez-Madrid,F. and Proudfoot,N. (1995) Expression of the early lymphocyte activation antigen CD69, a C-type lectin, is regulated by mRNA degradation associated with AU-rich sequence motifs. Eur. J. Immunol., 25, 2142–2146. [PubMed]
28. Umland S.P., Razac,S., Shah,H., Nahrebne,D.K., Egan,R.W. and Billah,M.M. (1998) Interleukin-5 mRNA stability in human T cells is regulated differently than interleukin-2, interleukin-3, interleukin-4, granulocyte/macrophage colony-stimulating factor and interferon-gamma. Am. J. Respir. Cell Mol. Biol., 18, 631–642. [PubMed]
29. Wingett D., Reeves,R. and Magnuson,N.S. (1991) Stability changes in pim-1 proto-oncogene mRNA after mitogen stimulation of normal lymphocytes. J. Immunol., 147, 3653–3659. [PubMed]
30. Ziegler S.F., Ramsdell,F. and Alderson,M.R. (1994) The activation antigen CD69. Stem Cells, 12, 456–465. [PubMed]
31. Nichols K.E., Koretzky,G.A. and June,C.H. (2001) SAP: natural inhibitor or grand SLAM of T cell activation? Nature Immunol., 2, 665–666. [PubMed]
32. Siegel R.M., Chan,F.K., Chun,H.J. and Lenardo,M.J. (2000) The multifaceted role of Fas signaling in immune cell homeostasis and autoimmunity. Nature Immunol., 1, 469–474. [PubMed]
33. D’Souza B., Rowe,M. and Walls,D. (2000) The bfl-1 gene is transcriptionally upregulated by the Epstein-Barr virus LMP1 and its expression promotes the survival of a Burkitt’s lymphoma cell line. J. Virol., 74, 6652–6658. [PMC free article] [PubMed]
34. Zong W.X., Edelstein,L.C., Chen,C., Bash,J. and Gelinas,C. (1999) The prosurvival Bcl-2 homolog Bfl-1/A1 is a direct transcriptional target of NF-kappaB that blocks TNFalpha-induced apoptosis. Genes Dev., 13, 382–387. [PMC free article] [PubMed]
35. Crabtree G.R. and Clipstone,N.A. (1994) Signal transmission between the plasma membrane and nucleus of T lymphocytes. Annu. Rev. Biochem., 63, 1045–1083. [PubMed]
36. Thompson C.B., Rathmell,J.C., Frauwirth,K.A., Lindsten,T., Rudin,C.M., Opferman,J.T., Ashton-Rickardt,P.G., Harris,M.H., Chandel,N.S., Schumacker,P.T. and Vander Heiden,M.G. (1999) What keeps a resting T cell alive? Cold Spring Harb. Symp. Quant. Biol., 64, 383–387. [PubMed]
37. Buckley A.F., Kuo,C.T. and Leiden,J.M. (2001) Transcription factor LKLF is sufficient to program T cell quiescence via a c-Myc–dependent pathway. Nature Immunol., 2, 698–704. [PubMed]
38. Raab M., Pfister,S. and Rudd,C.E. (2001) CD28 signaling via VAV/SLP-76 adaptors: regulation of cytokine transcription independent of TCR ligation. Immunity, 15, 921–933. [PubMed]
39. Bonecchi R., Bianchi,G., Bordignon,P.P., D’Ambrosio,D., Lang,R., Borsatti,A., Sozzani,S., Allavena,P., Gray,P.A., Mantovani,A. and Sinigaglia,F. (1998) Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J. Exp. Med., 187, 129–134. [PMC free article] [PubMed]
40. Epler J.A., Liu,R. and Shimizu,Y. (2000) From the ECM to the cytoskeleton and back: how integrins orchestrate T cell action. Dev. Immunol., 7, 155–170. [PMC free article] [PubMed]
41. Akdis C.A. and Blaser,K. (2001) Mechanisms of interleukin-10-mediated immune suppression. Immunology, 103, 131–136. [PMC free article] [PubMed]
42. Joss A., Akdis,M., Faith,A., Blaser,K. and Akdis,C.A. (2000) IL-10 directly acts on T cells by specifically altering the CD28 co-stimulation pathway. Eur. J. Immunol., 30, 1683–1690. [PubMed]
43. Jenkins M.K., Chen,C.A., Jung,G., Mueller,D.L. and Schwartz,R.H. (1990) Inhibition of antigen-specific proliferation of type 1 murine T cell clones after stimulation with immobilized anti-CD3 monoclonal antibody. J. Immunol., 144, 16–22. [PubMed]
44. Boise L.H., Minn,A.J., Noel,P.J., June,C.H., Accavitti,M.A., Lindsten,T. and Thompson,C.B. (1995) CD28 costimulation can promote T cell survival by enhancing the expression of Bcl-XL. Immunity, 3, 87–98. [PubMed]
45. Norton S.D., Zuckerman,L., Urdahl,K.B., Shefner,R., Miller,J. and Jenkins,M.K. (1992) The CD28 ligand, B7, enhances IL-2 production by providing a costimulatory signal to T cells. J. Immunol., 149, 1556–1561. [PubMed]
46. Jenkins M.K., Taylor,P.S., Norton,S.D. and Urdahl,K.B. (1991) CD28 delivers a costimulatory signal involved in antigen-specific IL-2 production by human T cells. J. Immunol., 147, 2461–2466. [PubMed]
47. Fraser J.D., Irving,B.A., Crabtree,G.R. and Weiss,A. (1991) Regulation of interleukin-2 gene enhancer activity by the T cell accessory molecule CD28. Science, 251, 313–316. [PubMed]
48. Bohjanen P.R., Petryniak,B., June,C.H., Thompson,C.B. and Lindsten,T. (1991) An inducible cytoplasmic factor (AU-B) binds selectively to AUUUA multimers in the 3′ untranslated region of lymphokine mRNA. Mol. Cell. Biol., 11, 3288–3295. [PMC free article] [PubMed]
49. Bohjanen P.R., Petryniak,B., June,C.H., Thompson,C.B. and Lindsten,T. (1992) AU RNA-binding factors differ in their binding specificities and affinities. J. Biol. Chem., 267, 6302–6309. [PubMed]
50. Carballo E., Lai,W.S. and Blackshear,P.J. (1998) Feedback inhibition of macrophage tumor necrosis factor-alpha production by tristetraprolin. Science, 281, 1001–1005. [PubMed]
51. Fan X.C. and Steitz,J.A. (1998) Overexpression of HuR, a nuclear-cytoplasmic shuttling protein, increases the in vivo stability of ARE-containing mRNAs. EMBO J., 17, 3448–3460. [PMC free article] [PubMed]
52. Ford L.P., Watson,J., Keene,J.D. and Wilusz,J. (1999) ELAV proteins stabilize deadenylated intermediates in a novel in vitro mRNA deadenylation/degradation system. Genes Dev., 13, 188–201. [PMC free article] [PubMed]
53. Jain R.G., Andrews,L.G., McGowan,K.M., Pekala,P.H. and Keene,J.D. (1997) Ectopic expression of Hel-N1, an RNA-binding protein, increases glucose transporter (GLUT1) expression in 3T3-L1 adipocytes. Mol. Cell. Biol., 17, 954–962. [PMC free article] [PubMed]
54. Levy N.S., Chung,S., Furneaux,H. and Levy,A.P. (1998) Hypoxic stabilization of vascular endothelial growth factor mRNA by the RNA-binding protein HuR. J. Biol. Chem., 273, 6417–6423. [PubMed]
55. Peng S.S., Chen,C.Y., Xu,N. and Shyu,A.B. (1998) RNA stabilization by the AU-rich element binding protein, HuR, an ELAV protein. EMBO J., 17, 3461–3470. [PMC free article] [PubMed]
56. Laroia G., Cuesta,R., Brewer,G. and Schneider,R.J. (1999) Control of mRNA decay by heat shock-ubiquitin-proteasome pathway. Science, 284, 499–502. [PubMed]
57. Chen C.Y., Gherzi,R., Ong,S.E., Chan,E.L., Raijmakers,R., Pruijn,G.J., Stoecklin,G., Moroni,C., Mann,M. and Karin,M. (2001) AU binding proteins recruit the exosome to degrade ARE-containing mRNAs. Cell, 107, 451–464. [PubMed]
58. Paillard F. and Vaquero,C. (1991) Down-regulation of lck mRNA by T cell activation involves transcriptional and post-transcriptional mechanisms. Nucleic Acids Res., 19, 4655–4661. [PMC free article] [PubMed]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...