| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Laboratory of Immunology, National Institute on Aging, National Institutes of Health, Baltimore, MD 21224;
Mathematical and Statistical Computing Laboratory, Center of Information Technology, National Institutes of Health, Bethesda, MD 20892; and
DNA Array Unit, National Institute on Aging, National Institutes of Health, Baltimore, MD 21224
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
200 cDNA clones whose expression levels changed after
activation and suggest that the level of expression of up-regulated
genes is a molecular mechanism that differentiates the response of
memory from naive CD4+ T cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The well-documented differences between human memory and naive
CD4+ T cells include the requirements for
activation and the magnitude of the subsequent cellular responses
(13, 14, 15, 16). Ag-naive CD4+ T cells
require the engagement not only of TCR as signal 1, but also a
costimulatory receptor, such as CD28, as signal 2 for a complete
activation leading to proliferation and differentiation into helper T
cells (15, 17, 18, 19). Stimulation with signal 1 in the
absence of signal 2 may lead naive T cells into an unresponsive or
anergic state (20). In contrast, memory
CD4+ T cells appear to have a lower activation
threshold than naive T cells. Depending on the in vitro system, memory
CD4+ T cells can be either fully activated
(13) or partially activated by signal 1 alone (15, 17, 21, 22). Once activated, memory and naive
CD4+ T cells display quite different cellular
responses (6, 23). Memory CD4+ T
cells produce a broader spectrum and a greater quantity of cytokines
including IL-1
, IL-2, IL-4, IL-5, IL-6, IL-17, IFN-
, TNF-, and
GM-CSF, whereas naive CD4+ T cells produce only
IL-1, IL-2, IFN-, and TNF- (24, 25, 26, 27, 28, 29).
At the molecular level, information regarding the difference between memory and naive CD4+ T cell response is limited (30). With the development of DNA microarray technology, the systematic analysis of gene expression has become feasible (31, 32). Two recent reports have analyzed gene expression in murine T cells and human CD4+ T cell clones using DNA arrays (33, 34). Teague et al. (33) studied murine T cells after activation with a superantigen in vivo. They reported that resting T cells express about the same number of genes as activated T cells and that activation via superantigen induced expression of 51–280 genes in murine T cells. Rogge et al. (34) analyzed human CD4+ Th1 and Th2 clones in vitro and showed that Th1 and Th2 clones expressed different sets of genes that could in turn modulate effector functions. However, analysis of the gene expression profile of memory CD4+ T cells and its comparison with naive CD4+ T cells at the genome scale has not been reported.
Here we report a genome-scale analysis of gene expression in human
memory and naive CD4+ T cells after analyzing
54,768 unique cDNA clones. We demonstrate that memory and naive
CD4+ T cells have a similar pattern of gene
expression at rest and after in vitro stimulation with anti-CD3 mAb
alone or with anti-CD3 plus anti-CD28 (anti-CD3/CD28) mAbs.
At the individual gene level, we have identified 14 cDNA clones that
were highly expressed in memory CD4+ T cells
relative to naive CD4+ T cells and 200 cDNA
clones whose expression levels were changed in memory
CD4+ T cells after in vitro anti-CD3/CD28
stimulation. Although the mRNA levels of the down-regulated genes are
diminished rather uniformly, the levels of up-regulated genes are
higher in memory than in naive cells and are higher after stimulations
with anti-CD3/CD28 than with anti-CD3 alone. Taken together,
these results provide a general assessment of activation-induced
changes in gene expression in memory and naive
CD4+ T cells and suggest that the increased
expression of certain genes defines memory CD4+ T
cell response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human memory and naive CD4+ T cells were isolated from peripheral blood by immunomagnetic separation as previously described (12). In brief, blood was obtained from normal donors of the National Institutes of Health Blood Bank. PBMC were isolated by Ficoll gradient centrifugation (Organon Teknika, Durham, NC), and then were incubated with a panel of mouse mAbs, including mAbs against CD8 (B9.8), CD19 (FMC63), CD11b (NIH11b-1), CD14 (63D3), CD16 (3G8), MHC class II (IVA12), erythrocytes (glycophorin, 10F7) and platelets (37F9-E7). mAbs against CD45RA (FMC71) or CD45R0 (UCHL-1) were added reciprocally for isolation of memory or naive CD4+ T cells. Ab-bound cells were subsequently removed by incubation with anti-mouse IgG-conjugated magnetic beads (Polysciences, Warrington, PA). The purity of isolated memory and naive CD4+ T cells was generally 90–95% determined by FACS analysis. Among the contaminating cells, CD45RA+CD45RO+ made up 3–5%, and each of CD8+, CD14+, CD19+, CD40+, and CD83+ cells made up <1%.
Stimulation of memory and naive CD4+ T cells in vitro
The procedures for stimulation of memory and naive CD4+ T cells were previously described (35). Anti-CD3 (OKT3) alone or anti-CD3/CD28 (9.3) mAb-conjugated magnetic beads (Dynal, Lake Success, NY) were gifts from C. June (University of Pennsylvania, Philadelphia, PA). Memory and naive CD4+ T cells were resuspended at 2–5 x 106 cells/ml in RPMI 1640 medium (Life Technologies, Rockville, MD) supplemented with 10% FBS (Gemini Biological Products, Calabasas, CA) and 1x penicillin-streptomycin (Life Technologies), mixed with Ab-conjugated beads at a 1:1 cell/bead ratio and were incubated for 16 h or other defined time points before harvest.
Proliferation assay
Memory and naive CD4+ T cells were cultured at 5 x 105 cells/well in quadruplicate in 96-well flat-bottom plates (Falcon, Franklin Lake, NJ) in 0.2 ml RPMI 1640 medium/10% FBS/1x penicillin-streptomycin and under the stimulation conditions as described above over a 5-day period. Then 1 µCi of [3H]thymidine was added to each well at 0, 24, 48, 72, and 96 h after stimulation. After incubation for another 20–22 h, the cells were then harvested at day 1, 2, 3, 4, and 5 by a 96-well plate harvester (Tomtec, Orange, CT), and the incorporation of [3H]thymidine into DNA was quantified using a liquid scintillation counter (Beckman Coulter, Fullerton, CA).
Analysis of cell division by cell tracking dye (CFSE)
Memory and naive CD4+ T cells were cultured with 10 µM CFSE (Molecular Probes, Eugene, OR) in RPMI 1640 at 2 x 106 cells/ml at 37°C for 10 min. Cells were washed with 10 volumes of cold PBS/0.1%BSA, resuspended in RPMI 1640 medium, and incubated at 37°C for 30 min before stimulation with anti-CD3/CD28 as described above for 5 days. Cells were collected from a FACScan (BD Biosciences, San Jose, CA), and the CFSE profiles were analyzed by CellQuest (BD Biosciences) and ModFit (Verity Software House, Topsham, ME).
RNA isolation and cDNA probe preparation
Total RNA was extracted from memory and naive
CD4+ T cells using STAT-60 RNA isolation solution
(Tel-Test, Friendswood, TX). Messenger RNA was isolated from the total
RNA by oligo(dT) beads (Dynal) and used for making cDNA probe. In
general, mRNA (0.6 µg) was mixed with 1 µl 50 µM
oligo(dT)18 (Research Genetics, Huntsville, AL),
incubated at 70°C for 10 min, frozen on dry ice, and lyophilized to
dryness. The dried mRNA and oligo(dT) mixtures were dissolved in
reserve transcription solution containing 1x Moloney murine leukemia
virus buffer, 571 µm each of dATP, dGTP, and dTTP, 40 µCi
[-33P]dCTP (2000 Ci/mmol, 10 µCi/µl),
and 240 U of Moloney murine leukemia virus reverse transcriptase
(Promega, Madison, WI) and incubated at 42°C for 2 h. To degrade
RNA, 39.5 µl water and 12.5 µl 1 N NaOH were added to the reaction
mixture and incubated at 37°C for 10 min and then neutralized by
adding 12.5 µl 1 M Tris-HCl (pH 8.0) and 10 µl 1 N HCl. The
unincorporated nucleotides were removed from the probe by a G-50 spin
column (Bio-Rad, Hercules, CA). The labeled cDNA probes were heated at
100°C for 3 min, quickly cooled on ice, and used immediately.
cDNA microarray experiments using commercial filters
The gene discovery array (GDA)2 human filters version 1.3 consisting of two 22 x 22-cm and one 11 x 22-cm nylon filter were purchased from Incyte Genomics (St. Louis, MO). The filters contain a total of 45,878 nonredundant human cDNA clones spotted in duplicate. Hybridization was conducted according to manufacturer’s instruction. In brief, filters were prehybridized in 35 ml (for 22 x 22-cm filters) or 17.5 ml (for 11 x 22-cm filter) Quickhyb hybridization solution (Stratagene, La Jolla, CA) at 43°C for 2 h in roller bottles and hybridized in the presence of sheared salmon sperm DNA (115 µg/ml) (Research Genetics), Cot1 DNA (0.1 µg/ml) (Life Technologies), and cDNA probes at 43°C for 12 h. cDNA probes prepared from 1.8 and 0.9 µg mRNA were used for the 22 x 22-cm and 11 x 22-cm filters, respectively. Filters were washed twice with 200 ml 2x SSC/0.1% SDS at 22°C for 30 min, once with 1x SSC/0.1% SDS at 65°C for 30 min, and twice with 0.6x SSC/0.1% SDS at 65°C for 30 min. The filters were then exposed to PhosphorImager screens (Molecular Dynamics, Sunnyvale, CA) for 20 h, and the images were collected by a PhosphorImager scanner (Storm 860; Molecular Dynamics). The human GeneFilters (GF200, GF201, and GF202) containing a total of 16,056 cDNA clones in each filter were purchased from Research Genetics. Hybridization was conducted according to the manufacturer’s instruction and was under similar conditions to those described above. cDNA probes prepared from 0.6 µg mRNA were used for each hybridization based on the size of the filters.
Custom-made cDNA microarray filters
A total of 2,878 cDNA clones were selected for constructing the custom-made cDNA microarray. cDNA clones were obtained from Incyte Genomics and Research Genetics and were cloned in this laboratory. Plasmids were isolated from bacteria with an alkaline lysis kit (Edge BioSystems, Gaithersburg, MD). The insert of cDNA clones were amplified by PCR using a pair of common primers, universal forward (5'-CTGCAAGGCGATTAAGTTGGGTAAC-3') and universal reverse (5'-GTGAGCGGATAACAATTTCACACAGGAAACAGC-3'), precipitated with ethanol and resuspended in Tris-EDTA buffer. The PCR products were denatured by NaOH (0.1 N) and spotted onto nylon membrane (Nytran Supercharge membrane; Schleicher & Schull, Keene, NH) by GMS 417 Arrayer (Affy- metrix, Santa Clara, CA). The pin size (diameter) of the arrayer is 300 µm, and the space between the centers of two spots is 665 µm. Each cDNA clone was spotted in duplicate onto each membrane. The printed filters were treated with UV cross-linking (Stratagene) for immobilizing DNA to the membrane.
Analysis of microarray results by P-SCAN
Image files were collected from the PhosphorImager in "gel" format and were processed using the P-SCAN analysis software as described (36) and freely available at http://abs.cit.nih.gov/pscan. Briefly, the hybridization spots on the image of the microarray were located, and the average image intensity was determined for the pixels within a circle of fixed radius around the spot. These numerical intensities were saved in a file along with array coordinates and merged with the "gene list." The analysis steps included a normalization of spot intensity by dividing by the median spot intensity for the entire image and the determination of relatively over- and underexpressed genes between two experimental conditions. For the GDA filters, a 2-fold cutoff was established to select interesting spots whose intensities have changed between two images; one of which must also show an intensity above the background level (approximated by the median intensity). Resulting lists of spots were then reviewed graphically to verify that the apparent spot intensities had, in fact, changed between the two original images.
The custom-made cDNA microarrays consisting of three filters were analyzed using the same methods as above but using the "custom array" option of P-SCAN. Each array contained two spots corresponding to each clone, an internal replicate that was used for analyzing the consistency of the results. A database of microarray data was created using Matlab (version 5.2; Mathworks, Natick, MA) for results of the custom-made microarray. This database contained a GenBank identifier for each clone, the location on the array of each clone, and the numerical measured intensity for each experiment using these microarrays. It also contained coded information about the experiment type (experiment number, cell type, cell pretreatment, time after treatment, replicate number, internal replicate number, etc.). Using this database, it was possible to quickly test a variety of normalization strategies and to select subsets of clones that met certain criteria. An initial subset was determined by requiring that the median change in normalized intensity of each clone from two independent hybridization experiments was >1.6-fold compared with the unstimulated condition. These selected clones were then presented in a color-coded (red = intensity increase, green = intensity decrease, white = no change) array, and inconsistently colored clones (indicating inconsistent expression ratios) were removed from the study.
Northern blot analysis
Northern blotting was conducted as described previously
(37). The probes were PCR products of selected cDNA clones
and labeled by random priming using Ready to Go DNA labeling beads
(Amersham Pharmacia Biotech, Piscataway, NJ) in the presence of
[-32P]dCTP (New England Nuclear, Boston,
MA). Prehybridization and hybridization were conducted at 55°C for 1
and 16 h, respectively. The washing includes twice with 2x
SSC/0.1% SDS at 22°C for 15 min, once with 1x SSC/0.1% SDS at
65°C for 30 min, and twice with 0.5x SSC/0.1% at 65°C for 30 min.
The filters were exposed to PhosphorImager screens, and the time of
exposure was optimized for each gene. The images were collected from
the PhosphorImager scanner (Storm 860; Molecular Dynamics).
Western blot analysis
Cells were lysed in Kyriakis lysis buffer (20 mM HEPES, pH 7.4,
50 mM 
-glycerophosphate, 10 mM sodium fluoride, 1% Triton X-100,
10% glycerol, 1 mM sodium orthovanadate, 10 µg/ml leupeptin, 10
µg/ml aprotinin, and 100 µg/ml
[4-(2-aminoethyl)-benzenesulfonylfluoride hydrochloride]). Lysates
from 1 x 106 cells were loaded to each
well and separated by SDS-PAGE. Proteins were transferred to
Immobilon-P membranes (Millipore, Bedford, MA). The membranes were
probed with anti-actin Ab (University of Iowa, Iowa City,
IA), washed three times, and incubated with donkey
anti-rabbit Ig HRP-linked whole Ab (Amersham Pharmacia Biotech).
Signals were detected using the ECLPlus detection
system (Amersham Pharmacia Biotech) according to manufacturer’s
instructions. The membranes were stripped and probed again with
anti-ZAP70 Ab (a gift from Ron Wange, National Institute on Aging,
National Institutes of Health, Baltimore, MD).
Analysis of F-actin by flow cytometry
Freshly isolated and stimulated memory and naive CD4+ T cells were fixed in 3.7% formaldehyde in PBS at 4°C overnight (or up to 1 wk). The fixed cells were washed twice in PBS at room temperature and incubated with 0.1% Triton X-100 in PBS for 5 min, then washed three times with PBS, and resuspended in 500 µl PBS containing 1% BSA (ICN Biomedicals, Costa Mesa, CA). To quantify the polymerized actin (F-actin) level, cells were stained with Alexa Fluor 488 phalloidin (Molecular Probes) and analyzed by flow cytometry
Cytokine assays
Supernatants were collected from in vitro-stimulated memory and
naive CD4+ T cells. IL-2, IL-4, IL-5, IFN-,
TNF-, and TNF- concentrations were determined using ELISA
immunoassay kit (BioSource International, Camarillo, CA, and R&D
Systems, Minneapolis, MN) according to the manufacturer’s
instructions.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our strategy to assess the transcriptional nature of memory and
naive CD4+ T cells is shown in Fig. 1
. We have applied the cDNA microarray
method with commercial filters for the initial genome-scale screening
and custom-made filters for detailed analysis of activation-induced
gene expression changes. We then used conventional methods (Northern
and Western blots) for confirmation of array results. The commercial
filters consisted of a total of 54,768 nonredundant cDNA clones that
were obtained from two sources: 1) GDA human filters (GDA version 1.3;
Incyte Genomics; 45,968 clones), and 2) GeneFilters (GF200, GF201, and
GF202; Research Genetics; 16,056 clones including 7,166 overlapping
clones with the GDA filters). The custom-made filter arrays consisted
of 2,878 cDNA clones selected from either the commercial filters or
genes with known immunological functions that were not on the
commercial filters.
|
A). The profiles of the CFSE
experiment showed that memory and naive CD4+ T
cells divided a maximum of six times following anti-CD3/CD28
stimulation for 5 days (Fig. 2B). Almost all memory (98.6%)
and naive cells (99.8%) entered cell cycle after stimulation. However,
memory cells had a greater number of cells in division 5 and 6 than
naive cells (Fig. 2B).
|
C). No CD25 expression was detected in either memory
or naive CD4+ T cells after anti-CD3
stimulation (data not shown). After anti-CD3/CD28 stimulation, the
kinetics of CD69 expression was similar to that after anti-CD3
stimulation, and the percentage of CD25 expressing cells approached a
plateau 16 h after stimulation (Fig. 2C). Therefore, we
selected 16 h after stimulation for the initial analysis of
activation-induced genes in memory CD4+ T cells.
Interestingly, although memory and naive CD4+ T
cells exhibited a similar percentage of surface expression of CD69 and
CD25 after stimulation, the intensity levels of CD69 and CD25 were
higher in memory than in naive CD4+ T cells (Fig. 2D). Genome-scale gene expression profiles of memory and naive CD4+ T cells at rest and after in vitro stimulation
At the genome-scale level, memory and naive CD4+ T cells expressed similar numbers of genes at rest (22 and 20%), after stimulation with anti-CD3 alone (28 and 27%), and with anti-CD3/CD28 (19 and 20% of total cDNA clones) based on a conservative estimate that the expressed gene must be at least 2-fold higher than the median intensity of all clones on the filter and clearly visible over the background. The results were derived from two independent experiments with quadruplicate data for each clone and are listed in http://www.grc.nia.nih.gov/branches/li/weng/arraydata.htm. Comparative analyses of gene expression using commercial and subsequently custom-made filters revealed that the difference in gene expression between freshly isolated memory and naive CD4+ T cells was very small and that significant changes of gene expression were found after stimulation in both memory and naive CD4+ T cells.
Genes highly expressed in memory CD4+ T cells
After a sequential analysis of gene expression using the
commercial and the custom-made filters, we identified 14 cDNA clones
(11 known genes and 3 expressed sequence tags (ESTs)) that were
expressed at consistently higher levels in memory cells than in naive
cells (Fig. 3A). Among the 11
known genes, CD69, cyclin-dependent kinase inhibitor 1B, and natural
killer cell protein 4 are involved in T cell activation and cell cycle
regulation (38, 39, 40). Integrin 5,
proteoglycan 1, and annexin A1 function in cell-cell and cell-matrix
interaction, which could contribute to lymphocyte migration and
activation processes (41, 42, 43). The functional importance
of the other genes and ESTs in memory cell function is not clear and
will require further study. Upon in vitro stimulation with
anti-CD3/CD28 for 16 h, the expression of four cDNA clones was
up-regulated (proteoglycan 1, protein kinase inhibitor, cytochrome
c, and one EST) (Figs. 3B and
4A) and three clones were
down-regulated (annexin A1, cyclin-dependent kinase inhibitor 1B, and
one EST) (Figs. 3B and 4B) in both memory and
naive CD4+ T cells. The remaining seven clones
did not change their expression levels significantly after stimulation
in both memory and naive CD4+ T cells (Fig. 3B).
|
After the sequential analyses using the commercial and custom-made
filters, we have identified 200 cDNA clones whose expressions are
changed in memory CD4+ T cells after
anti-CD3/CD28 stimulation (Fig. 4, A, B, and
D). The reproducibility of the changes of these selected
clones were analyzed between two independent experiments (Fig. 4C) and were confirmed by Northern analysis from a randomly
selected 16 cDNA clones (data not shown). Among the up-regulated cDNA
clones, 130 are known genes and 5 are ESTs (Fig. 4A). In
general, memory cells expressed higher levels of transcripts of these
up-regulated genes than did naive cells, and cells stimulated with
anti-CD3/CD28 expressed higher levels of these same transcripts
than cells stimulated with anti-CD3 alone (Fig. 4D). The
average log-ratios of these up-regulated known genes (±SE) are
0.065 ± 0.005 (anti-CD3-treated naive cells), 0.133 ±
0.005 (anti-CD3-treated memory cells), 0.260 ± 0.007
(anti-CD3/CD28-treated naive cells), and 0.339 ± 0.006
(anti-CD3/CD28-treated memory cells).
|
, B and D). In
contrast to up-regulated genes, the majority of the activation
down-regulated genes (70%) had a similar degree of reduction in mRNA
levels in both memory and naive CD4+ T cells
stimulated with either anti-CD3 or anti-CD3/CD28 (Fig. 4D). The average log-ratios (±SE) of these down-regulated
genes are –0.219 ± 0.007 (anti-CD3-treated naive cells),
–0.206 ± 0.010 (anti-CD3-treated memory cells), –0.295
± 0.007 (anti-CD3/CD28-treated naive cells), and
–0.296 ± 0.008 (anti-CD3/CD28-treated memory cells). Functional characteristics of up-regulated genes in memory and naive CD4+ T cells
To further analyze the function of genes that were up-regulated
after anti-CD3/CD28 stimulation, we divided them into five
functional groups and a "miscellaneous" group: 1) transcriptional
regulation, 2) receptor and signal transduction, 3) cytokines and
receptors, 4) cell cycle, and 5) structure and metabolism (Fig. 4A).
There are several distinct factors that contribute to the
transcriptional regulation in memory and naive
CD4+ T cells. The basic transcription factors,
including transcription factor 6-like 1, LSF, and SL1, may
regulate the basal level of transcription during lymphocyte activation.
Histone deacetylase 2 and nonhistone chromosomal proteins HMG14 and
HMG17 are known for regulating chromatin structure during transcription
(44), which has been reported in the regulation of
cytokine gene expression (45). v-Myc, IL enhancer binding
factor 2 (NF-ATp45), and NF-
B are involved in mediating signal
transduction and cell proliferation (46) as well as in
regulating transcription of IL-2, IL-4, and other genes involved in T
cell activation and differentiation (47, 48).
The engagement of TCR and costimulatory receptors that activate
tyrosine kinases and their downstream targets does not require new gene
expression (49). However, a sustained activation event
does require transcription of these signaling-related genes. Indeed,
many activation up-regulated genes appeared to function in signaling,
such as H-RYK receptor-like tyrosine kinase and HAX-1, a protein
associated with the Src family tyrosine kinase substrate HS1 that
functions in promoting cell survival in activated
CD4+ T cells (Fig. 4A)
(50). In addition, a large group of genes, including RhoC,
nucleolar GTPase, Ras-related GTP-binding protein RAGA, GTP-binding
protein NGB, guanine nucleotide exchange factor, and GTP-binding
protein RAB-2, were significantly induced by TCR stimulation.
Interestingly, calcium/calmodulin kinase II (CAMKII), which plays a
critical role in T cell activation and in generation of memory cells
(51), was significantly increased after activation.
Although the majority of memory and naive CD4+ T
cells have not entered the S phase of cell cycle after 16 h of
anti-CD3/CD28 stimulation measured by
[3H]thymidine incorporation, genes involved in
cell division were already up-regulated when compared with resting
cells. These induced genes include cyclins, cyclin-dependent kinases,
DNA repair enzymes, and chromosomal proteins involved in DNA
replication and chromosome segregation (Fig. 4A). In
addition, genes related to cell cytoskeleton, including actin,
actin-related protein 3, and -tubulin, were also significantly
increased along with genes involved in basic cellular functions and
energy metabolism such as GAPDH, glycogen synthase, ATP synthase, and
ribosomal proteins (Fig. 4A).
The ultimate consequence of activation of memory and naive
CD4+ T cells is the production of cytokines,
which influence the outcome of an immune response. Remarkably, we
observed that both memory and naive CD4+ T cells
expressed a large group of cytokine genes, including IL-2, IL-9,
IFN-, pro-B cell colony-enhancing factor, and lymphotoxin
(TNF-) (Fig. 4A). The expression of IL-4 and IL-5 was
extremely low and was detected by Northern blot in activated memory
CD4+ T cells (see Fig. 6). In addition, we have
observed enhanced expression of IL-1R and its accessory protein and
IL-15R in both memory and naive CD4+ T cells
after activation. Because some common cytokine receptors such as CD25
and IL-4R were not on the commercial or custom-made filters, we were
not able to assess the expression status of these genes.
|
The functional roles of genes that are highly expressed in the
resting T cells and down-regulated after activation have begun to be
understood (52). Their functions range from prohibiting
cellular proliferation to preventing apoptosis (33, 53).
To facilitate further analysis, we have separated these down-regulated
genes into four groups: 1) transcriptional regulation, 2) receptor and
signal transduction, 3) antiproliferation, and 4) cell survival with
the remaining genes placed in a "miscellaneous" group (Fig. 4B).
We identified several genes that function in prohibiting cellular
proliferation, including B cell translocation gene 1 (54),
IFN-induced transmembrane protein 1 (55), RhoGDP
dissociation inhibitor (56), and latent TGF binding
protein 1 (57). We also identified genes that are involved
in cell survival and antiapoptosis, such as ubiquination factor E4,
which is known to maintain cell viability and survival under stress
conditions (58), and neuronal apoptosis inhibitory
protein, which is capable of protecting motor-neuron from apoptosis
(59). The expression of two important IL receptors,
IL-7 (60) and IL-10 (61), were
decreased as well. In addition to these functionally grouped activation
down-regulated genes, a gene involved in DNA recombination and repair
(Ku Ag) was also identified (Fig. 4B).
Increase of actin expression in activated memory and naive CD4+ T cells
Actin polymerization plays a critical role in TCR signaling
(62). To assess changes in actin expression, we have
performed Northern analysis and confirmed that mRNA levels of actin and
its upstream regulator, the small guanosine triphosphatase RhoC
(63), increased in both memory and naive
CD4+ T cells after stimulation (Fig. 5A). To further characterize
actin changes at the protein level, we examined the actin monomer and
polymers in memory and naive CD4+ T cells after
in vitro stimulation. Both monomer and polymerized actin were
significantly increased after stimulation with anti-CD3 alone or
with anti-CD3/CD28 (Fig. 5, B and C). In
agreement with the levels of mRNA, memory cells expressed higher levels
of actin protein than naive cells under both stimulation conditions,
and anti-CD3/CD28 induced higher levels of actin than did
anti-CD3 alone (Fig. 5, B and C).
|
, TNF-, and TNF proteins than naive
CD4+ T cells
The essential function of CD4+ T cells is to
produce cytokines and to "help" the effector cells in an immune
response. Our cDNA microarray and/or Northern blot analyses showed that
several lymphokine genes, including IL-2, IL-4, IL-5, IFN-, TNF-,
and TNF, were up-regulated differentially in memory and naive
CD4+ T cells after stimulation (Figs. 4A and 6A). To test
whether changes in the mRNA content resulted in corresponding changes
in protein levels, and to determine the kinetics of cytokine
production, we measured IL-2, IL-4, IL-5, IFN-, TNF-, and TNF
proteins in the supernatants of stimulated memory and naive
CD4+ T cells over a 5-day period. Consistent with
the mRNA expression, IL-2, IL-4, IL-5, IFN-, TNF-, and TNF
proteins were not detected in freshly isolated lymphocytes but were
significantly induced after stimulation (Fig. 6, B and
C). The kinetics of secreted IL-2, IL-4, IL-5, IFN-,
TNF-, and TNF were also quite different between the two subsets
of T cells (Fig. 6, B and C). After anti-CD3
stimulation, all six cytokines were undetectable in the supernatant of
naive cells, but low levels of IL-2, IFN-, TNF-, and TNF were
detected in the supernatant of memory cells (Fig. 6B). After
anti-CD3/CD28 stimulation, both naive and memory
cells secreted high levels of IL-2, IFN-, TNF-, and TNF (Fig. 6C). However, memory cells secreted significantly greater
amounts of these cytokines than did naive cells over a 5-day period.
Interestingly, IL-4 and IL-5 were only detected from memory cells but
not naive cells (Fig. 6C). Thus, memory
CD4+ T cells should induce a stronger immune
response through production of a larger quantity and greater variety of
cytokines than naive CD4+ T cells.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
200 cDNA clones whose
expression was changed after in vitro stimulation with
anti-CD3/CD28. We present results showing that the expression
levels of those up-regulated genes are higher in memory cells than in
naive CD4+ T cells after stimulation, suggesting
that the level of expression of these genes is a molecular mechanism
that differentiates the response of memory from naive
CD4+ T cells.
Despite recent advances in the phenotypical and functional
characterizations of memory CD4+ T cells,
information regarding the molecular features of these cells is limited.
The comparative analysis in this report has revealed some interesting
differences as well as similarities between freshly isolated memory and
naive CD4+ T cells. Memory cells express several
genes including CD69, integrin 5, proteoglycan
1, annexin A1, and NK cell protein 4, that are involved in activation,
cell adhesion, and migration (38, 40, 41, 42, 43). These processes
are critically important for memory cell functions. In addition, memory
cells also express higher levels of CDK inhibitor 1B that prevent cells
from entering the cell cycle (39). Thus, these small sets
of genes provide a glimpse of the delicate balance between expression
of genes in activation/migration and in maintenance of the
"resting" status of memory cells. Nevertheless, the difference in
gene expression between freshly isolated memory and naive
CD4+ T cells is small. There are several possible
explanations that are not mutually exclusive. First, the difference
between memory and naive CD4+ T cells is small at
the gene transcript level but is significant at the protein
(functional) level. Second, the current microarray method is not
sensitive enough to detect subtle changes of the transcripts between
memory and naive CD4+ T cells. It is particularly
difficult to detect differences in genes that are expressed at low copy
number, such as IL-4 and IL-5 genes. Third, recent studies suggest that
memory and naive CD4+ T cells isolated based on
CD45 isoforms can be further divided into functional subsets (64, 65). Analysis of memory cell subsets isolated based on
additional cell surface markers, such as the chemokine receptor CCR7,
might reveal a greater difference in gene expression between subsets of
memory and naive CD4+ T cells (28).
Further studies with improved cDNA microarray methods are needed to
determine the subtle differences in gene expression profiles using
subsets of memory CD4+ T cells.
Memory and naive CD4+ T cells also share significant similarities in gene expression. A number of genes that function in maintenance of the "resting" status of T cells, such as genes involved in antiproliferation and antiapoptosis (53, 66), are expressed in freshly isolated memory and naive CD4+ T cells. These genes participate in transcription, signaling, antiproliferation, and cell survival. The expression of these genes was significantly down-regulated after activation, suggesting that the "resting" status of memory and naive CD4+ T cells is likely to be actively maintained through regulated expression of these antiproliferation and survival-related genes.
Among the cDNA clones that are highly expressed in memory CD4+ T cells vs naive CD4+ T cells, four were significantly up-regulated after stimulation with anti-CD3/CD28, suggesting that the higher expression levels of these activation-related genes/ESTs may contribute to the fast and strong response of memory CD4+ T cells upon stimulation. The expression kinetics of CD69 in resting and activated CD4+ T cells is quite interesting. Freshly isolated memory cells express CD69 mRNA yet have low to undetectable surface expression of CD69 protein. However, 16 h after stimulation, memory cells expressed high levels of CD69 protein, although the mRNA levels did not change. This suggests that CD69 mRNA has a high turnover rate (38) and that there are different mechanisms regulating CD69 mRNA transcription and protein translation.
The consequences of stimulating T cells through the TCR alone or through both TCR and costimulatory receptors have been well documented (3, 67). Here we compared quantitative and qualitative changes in gene expression in memory and naive CD4+ T cells stimulated with anti-CD3 alone or with anti-CD3/CD28. After anti-CD3 stimulation, both memory and naive CD4+ T cells down-regulated several groups of genes including antiproliferative genes and up-regulated many genes functioning in activation and proliferation. Although memory CD4+ T cells responded to anti-CD3 stimulation by up-regulating some activation-related genes including cytokine genes, these changes are not sufficient to induce complete activation and proliferation of memory cells.
In contrast, anti-CD3/CD28 stimulation induced significant changes
of gene expression in both memory and naive CD4+
T cells. Further comparative analysis indicated that memory and naive
CD4+ T cells exhibited a similar pattern of gene
expression for both up- and down-regulated genes, although expression
of the up-regulated genes was significantly higher in memory cells
after anti-CD3/CD28 stimulation (Fig. 4D). Thus, the
function of costimulation via CD28 can be viewed as an amplifier of
activation signals that enhance the expression of those up-regulated
genes. The precise mechanisms of CD28 signaling in mediating T cell
activation, particularly in up-regulation of activation-induced genes,
will require further studies. Taken together, these results indicate
that the induction of higher levels of activation-related genes in
memory CD4+ T cells after activation is a
molecular basis for the enhanced cellular response of memory cells.
Based on results from gene expression and cellular changes in memory
and naive CD4+ T cells at rest and after in vitro
stimulation, we propose a gene expression dosage gradient model of
memory and naive CD4+ T cell response (Fig. 7). In this model, memory and naive
CD4+ T cells express antiproliferation, cell
survival, and other genes to maintain the "resting" status.
Compared with naive cells, memory T cells may express fewer or lower
levels of these antiproliferation-related genes and higher levels of
activation-related genes. Stimulation through the TCR alone
(anti-CD3 alone) induces down-regulation of antiproliferation and
other genes that are required for maintaining the resting condition and
up-regulation of genes that are required for activation. However, the
expression level of those up-regulated genes is not high enough to lead
to cytokine production and proliferation in naive cells and can only
produce low levels of a limited number of cytokines and undergo partial
proliferation in memory cells. In contrast, stimulation through the TCR
and costimulatory receptor CD28 (anti-CD3/CD28) induces a similar
down-regulation of antiproliferation genes but higher expression levels
of up-regulated genes that are sufficient to completely activate both
memory and naive CD4+ T cells. However, even in
the presence of costimulatory stimulation, memory cells express higher
levels of those activation up-regulated genes and secrete more
cytokines than naive cells. Thus, our model proposes that activation of
CD4+ T cells requires two sequential molecular
events: down-regulation of genes required for maintenance of the
resting status and up-regulation of genes required for activation and
proliferation, and that the quantitative difference in expression
levels of those activation up-regulated genes is a molecular basis for
the qualitative difference between memory and naive
CD4+ T cell response (Fig. 7).
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: GDA, gene discovery array; EST, expressed sequence tag; FI, fluorescence intensity.
Received for publication October 20, 2000. Accepted for publication April 12, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Immunol. 162:3790.1 binding protein: a component of the large latent complex of TGF-1 with multiple repeat sequences. Cell 61:1051.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Y. Yang, J. An, and N.-p. Weng Telomerase Is Involved in IL-7-Mediated Differential Survival of Naive and Memory CD4+ T Cells J. Immunol., March 15, 2008; 180(6): 3775 - 3781. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K. McKinstry, S. Golech, W.-H. Lee, G. Huston, N.-P. Weng, and S. L. Swain Rapid default transition of CD4 T cell effectors to functional memory cells J. Exp. Med., September 3, 2007; 204(9): 2199 - 2211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fann, J. M. Godlove, M. Catalfamo, W. H. Wood III, F. J. Chrest, N. Chun, L. Granger, R. Wersto, K. Madara, K. Becker, et al. Histone acetylation is associated with differential gene expression in the rapid and robust memory CD8+ T-cell response Blood, November 15, 2006; 108(10): 3363 - 3370. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bosco, K. L. McKenna, C. J. Devitt, M. J. Firth, P. D. Sly, and P. G. Holt Identification of novel th2-associated genes in T memory responses to allergens. J. Immunol., April 15, 2006; 176(8): 4766 - 4777. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Goda, T. Kanaji, S. Kanaji, G. Tanaka, K. Arima, S. Ohno, and K. Izuhara Involvement of IL-32 in activation-induced cell death in T cells Int. Immunol., February 1, 2006; 18(2): 233 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Li, W. Zhi, P. Wareski, and N.-p. Weng IL-15 Activates Telomerase and Minimizes Telomere Loss and May Preserve the Replicative Life Span of Memory CD8+ T Cells In Vitro J. Immunol., April 1, 2005; 174(7): 4019 - 4024. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hess, Y. Yang, S. Golech, A. Sharov, K. G. Becker, and N.-p. Weng Kinetic assessment of general gene expression changes during human naive CD4+ T cell activation Int. Immunol., December 1, 2004; 16(12): 1711 - 1721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Liu and S. I. Abrams Alterations in Fas Expression Are Characteristic of, But Not Solely Responsible for, Enhanced Metastatic Competence J. Immunol., June 15, 2003; 170(12): 5973 - 5980. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Liu and S. I. Abrams Coordinate Regulation of IFN Consensus Sequence-Binding Protein and Caspase-1 in the Sensitization of Human Colon Carcinoma Cells to Fas-Mediated Apoptosis by IFN-{gamma} J. Immunol., June 15, 2003; 170(12): 6329 - 6337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Palena, J. Schlom, and K.-Y. Tsang Differential Gene Expression Profiles in a Human T-cell Line Stimulated with a Tumor-associated Self-peptide versus an Enhancer Agonist Peptide Clin. Cancer Res., May 1, 2003; 9(5): 1616 - 1627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. D. Dixit, R. Sridaran, M. A. Edmonsond, D. Taub, and W. E. Thompson Gonadotropin-Releasing Hormone Attenuates Pregnancy-Associated Thymic Involution and Modulates the Expression of Antiproliferative Gene Product Prohibitin Endocrinology, April 1, 2003; 144(4): 1496 - 1505. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Manel, S. Kinet, J.-L. Battini, F. J. Kim, N. Taylor, and M. Sitbon The HTLV receptor is an early T-cell activation marker whose expression requires de novo protein synthesis Blood, March 1, 2003; 101(5): 1913 - 1918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Liu, M. Catalfamo, Y. Li, P. A. Henkart, and N.-p. Weng IL-15 mimics T cell receptor crosslinking in the induction of cellular proliferation, gene expression, and cytotoxicity in CD8+ memory T cells PNAS, April 30, 2002; 99(9): 6192 - 6197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Liu, M. Catalfamo, Y. Li, P. A. Henkart, and N.-p. Weng IL-15 mimics T cell receptor crosslinking in the induction of cellular proliferation, gene expression, and cytotoxicity in CD8+ memory T cells PNAS, April 30, 2002; 99(9): 6192 - 6197. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
