| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
T Cell Subsets Defines Distinct Immunoregulatory Phenotypes and Unexpected Gene Expression Profiles
* Veterinary Molecular Biology, Montana State University, Bozeman, MT 59717; and
Department of Veterinary Pathobiology, University of Minnesota, St. Paul, MN 55108
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells using serial analysis
of gene expression (SAGE). Approximately 20,000 SAGE tags were
generated from each library. A comparison of the two libraries
demonstrated 297 and 173 tags representing genes with 5-fold
differential expression in GD3.5+ and GD3.5-
T cells, respectively. Consistent with their localization into
sites of inflammation, GD3.5+ T cells appeared
transcriptionally and translationally more active than
GD3.5- cells. GD3.5- T cells
demonstrated higher expression of the cell proliferation inhibitor BAP
37, which was associated with their less activated gene expression
phenotype. The immune regulatory and apoptosis-inducing molecule,
galectin-1, was identified as a highly abundant molecule and was higher
in GD3.5+ T cells. Surface molecules attributed to
myeloid cells, such as CD14, CD68, and scavenger receptor-1, were
identified in both populations. Furthermore, expression of B
lymphocyte-induced maturation protein, a master regulator of B cell and
myeloid cell differentiation, was identified by SAGE analysis and was
confirmed at the RNA level to be selectively expressed in T
cells vs 
T cells. These results provide new insights into the
inherent differences between circulating T cell
subsets. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells were first identified 16 years ago (1, 2),
and since then considerable effort has been devoted to understanding
their function and importance in human and animal health. Although low
in number in the peripheral blood, these cells are maintained in every
animal with a complex immune system emphasizing their importance as a
functional entity. A consensus concerning their role, however, has not
been established. An interesting aspect of these cells that has come
from studies in mice, humans, and cattle is that discrete subsets of
these cells, defined by their TCR usage, as well as certain
differentiation markers accumulate in different tissues
(3, 4, 5, 6, 7, 8, 9, 10). Recent findings demonstrate that tissue-specific
subsets have different functions, with some being proinflammatory and
others being anti-inflammatory (11, 12). Also, subsets
respond to infectious challenge differently. For example, removal of
the V1 subset improves the ability of mice to clear Listeria
monocytogenes infection, whereas depleting all T cells has
the opposite effect (13). Conversely, in coxsackie virus
infection, depletion of the V1 subset results in an exacerbation of
the infection-induced myocarditis. On the other hand, removal of V4
cells reduces the inflammatory damage caused by infection
(14). It is likely that some of the inconsistency in the
literature concerning T cells is due to the analysis of mixed
populations of cells that contain subsets with opposing activities.
Therefore, to fully understand T cell populations as a whole,
experiments must include analyses of individual subsets.
Some differences in T cell subsets isolated from tissues are
caused by influences of local growth and activating factors. However,
many differences are due to inherent differences in the subsets
themselves. Various tissue-specific T cell subsets can be found
in the circulation, thus some of the confounding effects of tissue
microenvironments on T cells can be avoided by analyzing these cells.
However, the analysis of circulating T cells can be difficult
due to their low numbers in the peripheral blood of most mammals
(usually 1–10%). In contrast, newborn ruminants, such as 1- to
6-mo-old calves, have 30–70% circulating T cells
(15); making them a useful animal model for these types of
studies. As in other mammals, the majority of bovine T cells are
negative for CD4 and CD8. However, one T cell subset is of
particular interest because of its coexpression of CD8 and CD2, both
functionally important accessory molecules, and lack of GD3.5 Ag.
Bovine CD8+, GD3.5-
T cells are found in low numbers in blood and in high numbers in
mucosal sites and the red pulp of the spleen (8).
Conversely, CD8-, GD3.5+
T cells are found predominantly in the white pulp of the spleen,
peripheral lymphoid tissues, sites of inflammation, and the blood
(16). The unique distribution of these subsets suggests
that their functions are different, although a comprehensive comparison
has not been performed.
High throughput sequencing of expressed sequence tags (ESTs) and/or
hybridization of gene arrays allows for a comprehensive analysis of
global gene expression in cells and can provide insights into novel
cellular functions. Serial analysis of gene expression
(SAGE)3 or Affymetrix
high density oligonucleotide arrays have been applied to the study of
different T cell subsets, including T cells. Fahrer and
colleagues (17) compared gene expression patterns of
tissue-specific, resting intraepithelial lymphocytes (IELs) vs
cells following Yersinia infection, and identified
unexpected regulated genes important in lipid metabolism, cholesterol
homeostasis, and physiology. Recently, using SAGE (18),
Shires et al. (19) characterized an activated, yet resting
phenotype of murine IELs, due to the high expression of
molecules, such as granzymes and Fas ligand, and the low expression of
conventional cytokines and their receptors. To date, there have been no
reported functional genomic analyses of circulating T cells. In
this report we describe a SAGE analysis of circulating
GD3.5+ and GD3.5- bovine
T cells, which are also distinguished by CD2/CD8 expression, to
analyze their functional potential before being recruited to a specific
tissue site. Results showed a surprising number of differentially
expressed genes that provide new insights into the functional
differences of these T cell subsets and the relationship of
these cells to myeloid cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Holstein calves were purchased from local producers and housed at the Montana State University large animal facilities at the Veterinary Molecular Biology Laboratory. Cattle used in this study were bull calves from 1–4 mo of age. Peripheral blood was collected into sodium heparin anticoagulant tubes by venipuncture, and PBMC were purified by Histopaque 1077 (Sigma-Aldrich, St. Louis, MO) gradient centrifugation.
Abs and FACS analysis
The following mouse mAbs, whose staining patterns on bovine
T cells have been previously characterized (8, 16, 20, 21), were used in these studies. GD3.8 recognizes TCR
(8). GD3.5 recognizes nearly all
CD8- T cells, but not
CD8+ T cells (8, 16, 20).
CC58 recognizes bovine CD8, and CC42 recognizes bovine CD2
(21). MHM23 recognizes bovine CD18 (DAKO, Glostrup,
Denmark). H58A (anti-bovine MHC class I), CACT116A (anti-bovine
IL-2R), and BAT31A (anti-bovine CD44) were obtained from VMRD
(Pullman, WA). Some mAbs were directly conjugated to FITC, PE, or
biotin for multicolor flow cytometric analysis. Second-stage reagents
included PE-conjugated anti-mouse IgG (Jackson ImmunoResearch
Laboratories, West Grove, PA) and avidin-conjugated CyChrome (BD
PharMingen, San Diego, CA).
Multicolor flow cytometric analysis was performed as follows. One
hundred microliters of hybridoma supernatant fluid or purified mAb at a
final concentration of 20–50 µg/ml was incubated with cells for 30
min on ice and then washed from the cells with PBS containing 5% horse
serum (Sigma-Aldrich; PBS-HS). PE-labeled second-stage, diluted 1/250
in PBS-HS was then added and incubated for 30 min on ice. The samples
were washed with PBS-HS, incubated in 10% mouse serum in PBS for 15
min, and washed again. As an example of a three-color stain,
biotin-labeled GD3.8 (anti-pan T cell mAb) and
FITC-conjugated CC58 (anti-CD8) were added, and the cells were
incubated on ice for 30 min. Cells were washed in PBS-HS and incubated
with avidin-CyChrome diluted 1/2000 in PBS-HS. After 30 min on ice, the
cells were washed in PBS-HS and analyzed using a FACSCalibur (BD
Biosciences, Mountain View, CA). The 488-nm laser and FL1 (FITC), FL2
(PE), and FL3 (CyChrome) detectors were used. The FACSCalibur was
calibrated using Calibright beads (BD Biosciences). Compensation was
set manually using single-color stains of the various fluorochromes.
Data from up to 50,000 cells were acquired. Negative controls included
1) single-color stains, 2) irrelevant isotype-matched Ab stains, and 3)
second-stage reagent controls. For statistical analysis, markers were
placed just above the upper limit of background staining of
control Abs.
Cell sorting
For high speed cell sorting of CD8+ and
CD8- T cells, two different staining
protocols were used based on our previous studies (8, 16).
In one, GD3.8 (anti- T cell) and CC58 (anti-CD8) were
used in a two-color stain to detect the two populations. In the other,
GD3.8 was combined with GD3.5, which stains CD8-
T cells (8, 16, 20). CD8+
T cells express lower levels of TCR, and this characteristic was
also used in restricting the sort gates to ensure the purest population
possible (16) (see Results). To obtain sorted
T cells vs enriched T cells, PBMC were stained using
GD3.8 Ab (total T cells) vs an FITC-labeled CC42 Ab
(CD2+ cells) (20). All
GD3.8-positive cells were included in the T cell sort, whereas
the T cells included only CD2+
GD3.8- cells (all bovine T cells express
CD2). High speed cell sorting was performed on a Vantage SE cell sorter
equipped with Turbo Sort (BD Biosciences). Sort rates ranged from
10,000–20,000 cells/s. The purity of each sort was confirmed by
analyzing samples on a FACSCalibur, as described above.
CD8+ T cells ranged from 1–5% of the
starting population in each sort and were the limiting cell population.
Cell yields ranged from 1 x 106 to 2
x 107 cells/sort from 
100 ml of blood.
RNA isolation and RT-PCR analysis
Sorted cells were seeded in 24- or 6-well culture plates
(Costar, Cambridge, MA) at a concentration of 1 x
106/ml and were cultured overnight at 37°C in
5% CO2 in RPMI containing 10% FCS. The
following day, cells were stimulated for 3.5 h with 20 ng/ml of
PMA (Sigma-Aldrich) and 0.5 µg/ml of ionomycin (Sigma-Aldrich). Cells
were then washed with HBSS and pelleted by centrifugation, and total
RNA was isolated using the TRIzol reagent (Life Technologies,
Gaithersburg, MD), according to the manufacturer’s protocol. The RNA
from at least three sorts of blood from different calves (at least
1 x 106 sorted cells for each subset from
each animal) was combined for further analysis. The pooled RNA was
treated with DNase, extracted again with phenol/chloroform, and used in
real-time PCR analysis. RT was performed with Superscript RT, random
primers (Invitrogen, San Diego, CA), and 300 ng sample
(CD8+ or CD8- T
cell) RNA according to the manufacturer’s protocol. Relative specific
mRNA in the CD8+ and CD8-
T cells was quantified by measuring SYBR green incorporation
during real-time quantitative PCR using the relative standard curve
method. Bovine sequences were analyzed using Primer Express software
(PE Applied Biosystems, Foster City, CA) to design optimal real-time
PCR primers (Table I
). Primers specific
for bovine 18S RNA were used as the endogenous control. Standard curves
were constructed using serially diluted, similarly extracted, total
bovine PBL RNA. One microliter of each RT reaction was used in the 25
µl real-time PCR reactions, which were performed in triplicate. The
PCR was set up and cycled, data were collected on the GeneAmp 5700
Sequence Detection System (PE Applied Biosystems), and calculations
were performed as described in the manufacturer’s protocol and in User
Bulletin 2 for the ABI PRISM 7700 Sequence Detection System.
Statistical significance of differential expression was determined for
each primer set using Student’s t test. PCR products
(ornithine decarboxylase, IL-1, selenoprotein T,
Bcl-xL, JunB, B lymphocyte-induced maturation
protein-1 (BLIMP-1), and CD14) were confirmed by sequence analysis.
|
. Primers were based on bovine EST data
from the National Center for Biotechnology Information data bank
as well as based on published sequences (24, 25). SAGE library construction and analysis
Due to the relatively low numbers of sorted
CD8+ T cells that could be obtained, SAGE
libraries were constructed by template switching and PCR amplification,
as previously described (26). Briefly, a biotinylated
oligo(dT) primer
(biotin-5'-AAGCAGTGGTAACAACGCAGAGTAC(T)30VN-3',
where V = A, C, or G, and N = T, C, G, or A) and
Superscript II reverse transcriptase (Invitrogen) were used to make
first-strand cDNA from 100 ng of total RNA. Second-strand synthesis and
cDNA amplification were completed by PCR using Advantage2 Polymerase
Mix (Clontech, Palo Alto CA) and a switching primer
(5'-AAGCAGTGGTAACAACGCAGAGTACGCGGG-3') in combination with the original
biotinylated oligo(dT). SAGE library construction was completed using 5
µg of the ds-cDNA following standard protocols (19, 27, 28). Large-scale sequencing of concatermarized di-tags was
completed at the University of Minnesota sequencing facility
(Veterinary Pathobiology, Minneapolis, MN) and the U.S. Department of
Agriculture Meat Animal Research Center (Clay Center, NE). Tag
frequencies were analyzed using the SAGE software package (version
4.05) (19) and a relational database (ACCESS; Microsoft,
Redmond, WA). Genes represented by tag sequences from the
analysis were identified by direct comparison with annotated
3'-specific tag sequences extracted from the available Bos
taurus ESTs. To provide current sequence annotation, bovine ESTs
were first BLASTed locally against the nonredundant nucleotide and
protein databanks obtained from National Center for Biotechnology
Information (http://www.ncbi.nlm.nih.gov).
RNase protection assay
Total T cells and -enriched lymphocytes were sorted
as described above. RNA was isolated from 4-h PMA/ionomycin-stimulated,
unsorted PBMC; total T cells; and -depleted,
-enriched lymphocytes (CD2+,
GD3.8-). Bovine BLIMP-1 primers were designed
based on sequence homology of a bovine EST to human and mouse BLIMP-1
(National Center for Biotechnology Information accession no. NP-031574;
gb BE483183.1) (5'BLIMP-1, CCAGTGCTGTGAAGGTTCCA; 3'BLIMP-1,
AGCTCCCCTCTGGAATAG; PCR size, 420 bp), and an
RsaI/Sau3a restriction digest for directional
cloning into the SmaI/BamHI-digested pBS vector
(Stratagene, La Jolla, CA) was performed, generating a 236-bp fragment.
The pBS-BLIMP-1 construct was linearized with EcoRI for in
vitro transcription of an antisense RNA probe using T3 RNA polymerase
(Roche, Indianapolis, IN) in a 20-µl reaction volume with 1x
transcription buffer (10 mM DTT, 0.3 mM cold rNTP mix (rATP, rCTP,
rGTP), 2.5 µM cold rUTP, 50 µCi of
[-32P]UTP, and 0.6 µl of RNasin) at 37°C
for 1 h. For DNA digest, 25 µg of yeast RNA and 5 µl of RQ1
DNase (Promega, Madison, WI) were added and incubated for 30 min at
37°C. The reaction was stopped with 150 µl of TES (10 mM Tris (pH
7.5), 5 mM EDTA, and 1% SDS) and unincorporated nucleotides were
removed using a 5-ml Sephadex G-50 fine column. As a standard curve,
declining concentrations of 4-h PMA/ionomycin-activated bovine PBMC-RNA
(10 to 0.1 µg) were used and compared with 1 µg of and
T cell RNA. For hybridization, yeast RNA was added to the
standard and sample RNAs for a final amount of 50 µg of RNA (e.g., 1
µg of sample RNA plus 49 µg of yeast RNA). Yeast RNA alone was used
as a negative control in the hybridization reactions. Five femtomoles
of labeled antisense BLIMP-1 RNA probe was used in the hybridization
procedure. Probe and samples were mixed, lyophilized, and then
hybridized in 30 µl of hybridization buffer (80% deionized
formamide, 40 mM PIPES (pH 7.0), 400 mM NaCl, 1 mM EDTA, in
0.1x TES) at 55°C for 18–24 h. For RNase digestion, tubes were
centrifuged, and 390 µl of RNase mix (10 µg/ml of RNase A (Roche)
and 0.35 U/µl of RNase T1 (Roche)) in 1x RNase digestion buffer (10
mM Tris (pH 7.5), 5 mM EDTA, and 300 mM NaCl) was added and incubated
for 1 h at 37°C. RNase digestion was terminated by adding 25
µl of proteinase K mix (1 µg/µl of proteinase K and 0.5 µg/µl
of yeast RNA in 10x TES buffer) and was incubated for 15 min at
37°C. Samples were phenol/chloroform-extracted, ethanol-precipitated,
and separated on a 5% denaturing acrylamide gel. The gel was dried and
exposed to x-ray film for 1–3 days.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cell subsets and SAGE library
construction
Using the different staining procedures outlined in
Materials and Methods, we found that the best purity of
CD8+ T cells was achieved when using GD3.8
and GD3.5 Ab staining. This is because CD8 surface expression on
T cells was too low to achieve a clean sort at rates >10,000 cells/s
(purities were <70% when sorting using anti-CD8; data not shown).
Using the higher intensity of the GD3.5 stain combined with the stain
for TCR provided a sorting effectiveness approaching 100% (Fig. 1A). Phenotypic analysis of
the sorted populations showed that the GD3.5-,
low TCR cells were CD2+ and >90%
CD8+, and the GD3.5+ cells
lacked CD2 and CD8 (data not shown) (8, 16, 20, 29).
Because of our past (16) and current focus on CD8, we
refer to these subsets as being either CD8+ or
CD8- T cells.
|
,
lymphotactin, TNF-, and IL-4 (N. Meissner and M. A. Jutila,
unpublished observations). Amplified cDNA was generated, and
differential gene expression of L-selectin, which we have previously
shown to be selectively expressed on CD8-
T cells (16), was measured by RT-PCR of the same sample as
that used for SAGE library construction. As shown in Fig. 1B, a greater amount of L-selectin PCR product was amplified
from the CD8- vs the CD8+
T cell library.
A total of 42,910 SAGE tags from both libraries were generated for this
initial evaluation. This provided 10,460 and 6,634 tags with
unique sequences for the CD8+ and
CD8- libraries, respectively (Table II). At this level of coverage, many
sequences were represented by only one tag. Of all unique tags, 1118
had a frequency of more than five tags. More than 60% of these tags
matched at least one bovine EST present in either the TIGR, UNIGENE, or
a bovine EST database available at National Center for Biotechnology
Information (data not shown). A surprising number of genes were
differentially expressed in activated CD8+ and
CD8- T cells (Table II). There were 173
and 297 distinct sequences differentially expressed by 5-fold or more
in CD8+ and CD8- T
cell subsets, respectively. These numbers were reduced to 34 and 77,
respectively when considering differences >10-fold. Overall,
2,000 unique tags from each library matched at least one EST;
however, far more tags had no match in the genome databases. Thus,
there was no reason at this time to sequence deeper into the
libraries, and the current level of coverage was used
to compare the two subsets. Complete listings of all tags and
corresponding blast results can be found at:
http://vmbmod10.msu.montana.edu/vmb/jutila-lab/SageBovine.htm.
|
).
Although the selective expression of selenoprotein T predicted by SAGE
was seen in the real-time RT-PCR, the difference was not significant
(p < 0.5; Fig. 2). All other confirmed
differences were significant (p < 0.05 or
p < 0.01). Semiquantitative RT-PCR results were
consistent for nocturnin (CCR4), IL-2R, galectin-1, IL-, and
selenoprotein T, but not Bcl-xL (data not shown).
Table III shows a comparison of SAGE tag
frequency to the levels of MHC I, CD44, CD18, IL-2R -chain, and TCR
on T cells as estimated by FACS analysis, again consistent with
the SAGE analysis. Protein expression profiles are shown from
unactivated and 24-h PMA/ionomycin-stimulated cells (minimal surface
changes were detected at 3.5 h; data not shown). Our staining for
CD18, IL-2R, and TCR were consistent with our previous studies
(16, 24) (E. Wilson, unpublished observations). MHC
I has been previously shown to be homogeneously expressed on bovine
lymphocytes (30). These results demonstrate that SAGE is a
reliable predictor of gene regulation in this system, particularly for
differences in SAGE that are 
5-fold.
|
|
T cells vs CD8+ T cells as predicted by SAGE
With few exceptions, the most abundant tags were general
housekeeping genes, coding for ribosomal proteins, histones, and
transcription and translation initiation factors (Table IV). One striking exception was connexin
43, a gap junction protein. Although both cell subsets were stimulated
in the same fashion, out of 100 genes identified by SAGE involved in
cellular activation, 37 genes were 5-fold higher in
CD8- T cells as opposed to only five
genes in CD8+ T cells (Table V). Furthermore, the expression profile
of CD8- T cells included genes involved
in cell cycle regulation such as cyclin-dependent kinase-5, cell cycle
check point proteins such as ornithine decarboxylase, and protein
translation-associated ribosomal proteins and translation elongation
factors. Additional evidence of the activated phenotype of
CD8- T cells was the higher expression of
the transcription factor NF-
B and its anti-apoptotic target
genes Bcl-xL and Bcl-2, and the IL-2R (Table V).
In contrast to CD8- T cells,
CD8+ T cell showed a 6-fold greater
expression (47 vs 8 tags) of the prohibitin-related protein, BAP 37, a
molecule initially described as a B cell receptor-associated molecule
that is involved in inhibition of cell proliferation (31).
BAP 37 expression is consistent with the less activated/proliferative
phenotype of CD8+ T cells. Differential
expression of ornithine decarboxylase, selenoprotein, nocturnin, IL-2R,
and Bcl-xL was confirmed by RT-PCR or FACS (Fig. 2, Table III, and data not shown).
|
|
Genes for a number of different cell surface molecules on immune
cells, such as MHC Ags, adhesion molecules, cytokine receptors, and
TCRs, were identified, and some were differentially expressed between
the two cell types (Table VI). Examples
of differentially expressed genes included CD44, semaphorin 4D,
scavenger receptor 1, TCR -chain, and IL-2R, all of which, except
CD63, were higher in CD8- T cells. As
expected, based on previous studies (9, 32), TCR
transcripts were represented in both T cell subset SAGE
libraries (Table VI).
|
). Genes that
were differentially expressed included galectin-1, prolactin, and
IgE-dependent histamine-releasing factor, which were >5-fold higher in
CD8- cells, and epidermal growth factor, IL-10,
and Gro-, which were >5-fold higher in CD8+
cells. Other genes, such as IL-1, were also differentially expressed,
but below the 5-fold level threshold (Table VI). Confirmation
experiments were performed for IL-1, CD44, TCR, CD18, MHC class I, and
galectin 1. Real-time RT-PCR confirmed the slightly higher expression
of IL-1 on CD8+ T cells and the
significantly higher expression of galectin-1 on
CD8- T cells (Fig. 2). FACS analysis of
resting and PMA/ionomycin-stimulated cells showed that CD44 and TCR
were more highly expressed on CD8- T
cells, and CD18 and MHC class I were the same on the two subsets,
consistent with the SAGE results (Table III).
Overall, these analyses suggest that T cells express
surface Ags and produce soluble factors that influence not only the
immune system, but the nervous, endocrine, and coagulation systems
as well.
T cells express genes normally associated with myeloid cells
A striking observation was the finding that myeloid-associated
genes were identified in both subsets by the SAGE analysis. For
example, myeloid surface Ags, such as scavenger receptor CD68, CD14
(confirmed by real-time RT-PCR; Fig. 2), and scavenger receptor 1, as
well as cytokines known to support myeloid cell differentiation and
activation, such as GM-CSF and G-CSF, were detected (Table VI). Another
molecule identified was the IgE-dependent, histamine-releasing factor,
a novel cytokine associated with activation and subsequent histamine
release in basophils and mast cells in the context of allergic
reactions. Of particular interest was the finding that the
transcriptional repressor BLIMP-1, which is considered an important
regulator of myeloid cell and B cell differentiation, was expressed in
both subsets, although CD8- T cells had
higher tag numbers (Table V and Fig. 2). Since BLIMP-1 expression is
thought to be restricted to myeloid cells and B cells, confirmation of
this observation was tested by two approaches. First, BLIMP-1
transcripts were detected by real-time RT-PCR in both subsets; however
CD8- T cells expressed higher levels
(Fig. 2). As another means of confirming these observations, an RNase
protection assay using a 236-bp BLIMP-1 antisense RNA probe was
performed. BLIMP-1 expression was compared in T cells and
-depleted, -enriched lymphocytes. As shown in Fig. 3, a significantly stronger BLIMP-1
signal was detected in 4-h PMA/ionomycin-stimulated T cell RNA,
as opposed to in the T cell-enriched RNA sample (CD2-positive,
non- T cells). Furthermore, the BLIMP-1 signal of 1 µg of
RNA compared with a signal between 1 and 3 µg of PBMC
standard RNA. Thus, T cells contributed the largest amount of
BLIMP-1 transcript in the RNA from total PBMCs.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cell subsets isolated from peripheral blood. Until now, the
only genomic studies of T cells have focused on tissue cells
isolated from the intraepithelial lining of the murine gut mucosa
(17, 19). By focusing on cells within the blood and
subsets known to localize to distinct tissues, functional differences
that relate primarily to the T cells themselves were identified, vs
those differences that are generated and maintained once the cell
enters into a tissue and responds to that tissue environment. Based on
this initial analysis, new insights into the functional potential of
T cells as a whole cell class as well as insights into
considerable differences in gene expression in circulating T
cell subsets have been gained. Although the antigenic characteristics
of the two subsets are not absolute, for the most part the sorted
GD3.5+ and GD3.5- cells
are defined by expression or lack of expression of CD8
(16). As such, we focus our discussion on
CD8+ and CD8- cells, with
the caveat that conclusions concerning the
CD8+ cells probably hold for the
GD3.5-, CD8-,
CD2+ subset described by others
(33), which was a minor population in the animals used in
this study.
Although a standard method of maximizing gene expression using
PMA/ionomycin was used, the comparisons in this study were focused on
differences between the two cell populations, not the role of cellular
activation. A separate study is being conducted that addresses the role
of mitogen (Con-A/IL-2) activation on gene expression in T
cells. Despite the low overall percentage of identified tags (due to
working in the bovine system), we gained information on 2000 or more
identified genes and/or ESTs for each library. Over 460 unique tags
were represented at least 5-fold higher in one or the other subset. A
considerable number of expected genes, based on earlier studies, were
identified, but the level of coverage in this first study was not
sufficient to identify some genes known to be differentially regulated
at the protein level in CD8+ and
CD8- T cells, such as CD8, CD2, WC1, and
CD6. This suggests that as we generate more SAGE tags, and as bovine
databases are expanded, many additional, regulated genes will be
identified.
One striking observation that came from the SAGE analysis was the
difference in the overall functional and/or activation status of the
circulating CD8- vs CD8+
T cells. CD8- T cells showed
significantly higher tag numbers for genes involved in transcriptional
and translational regulation, proliferation, and inhibition of
apoptosis. These differences were evident with or without the PMA
stimulation. For example, IL-2R was much higher on both unactivated and
PMA-stimulated CD8- vs
CD8+ T cells. Furthermore, these findings
are in agreement with and complement a recently performed cDNA array
analysis of Con A/IL-2-activated CD8+ vs
CD8- T cells, which showed the same
activation differences of the two subsets (52). In their
recent study Shires et al. (19) described IEL
compared with IELs in the mouse to be in a resting yet activated
state, meaning that T cells express in high abundance molecules
involved in cytotoxicity such as Fas ligand and granzymes, but low
amounts of cytokines. It may be that circulating
CD8- T cells are in a preactivated or
activated/resting state as well, which allows these cells to more
rapidly respond to external stimuli. Indeed, our previous studies have
found that bovine CD8- T cells respond
rapidly to acute inflammatory and mitogenic stimuli (16).
They also show a high level of random migration in in vitro migration
assays (E. A. Wilson and M. A. Jutila, unpublished
observations). Even though they express less CxCR4 receptor than
CD8+ T cells, CD8-
cells respond equally well, if not better, to the CxCR4 ligand SDF-1
(34) (E. Wilson and M. A. Jutila, unpublished
observations).
Based on the SAGE analysis, one gene that could contribute to the
differences in the activation status of CD8- vs
CD8+ T cells is BAP 37, a
prohibitin-related molecule involved in inhibition of cell
proliferation and RNA translation efficiency (31, 35).
SAGE analyses showed BAP 37 to be significantly higher in
CD8+ T cells as opposed to the
CD8- T cell subset. BAP 37 was originally
defined in B cells, and its expression in T cells suggests a
wider role for this molecule in regulating immune cell activity.
Another novel finding in our SAGE analysis was the differential
expression of galectin-1, originally described to be expressed on
epithelial and vascular endothelial cells (36, 37). Based
on SAGE, real-time RT-PCR, and semiquantitative RT-PCR (data not
shown), galectin-1 is expressed in T cells and 2- to 5-fold
higher in the CD8- T cell population.
Galectin-1 is a soluble, secreted lectin that has been shown to be
involved in the induction of apoptosis of thymocytes and activated T
cells. Its expression in thymic epithelial cells and endothelial cells
induces death of adherent T cells in a carbohydrate-dependent manner
(36, 37). Galectin-1 expression by
CD8- T cells suggests a novel role in the
regulation, particularly down-regulation, of the adaptive immune
response by these cells.
Of considerable interest, several genes normally associated with
myeloid cells were expressed in, and some were differentially expressed
between, the two T cell subsets. CD14, CD68, and macrophage
scavenger receptor 1 were represented by significant numbers of SAGE
tags in the two libraries. CD14 transcripts are expressed by bovine
T cells, as shown here by RT-PCR and SAGE, and
CD8- T cells bind FITC-labeled
Escherichia coli through a polyinosinic acid inhibitable
receptor, a characteristic of scavenger receptors (N. Meissner,
unpublished observations). This finding is not restricted to bovine
cells. T cells from two patients with Crohn’s disease express
CD14 (38), and we have found that some in vitro expanded
human T cells can be induced to expressed CD14 as well (J.
Hedges and M. Jutila, unpublished observations). SAGE also predicted
the expression of cytokines produced by and others that act on myeloid
cells. An example included IgE-dependent histamine-releasing factor, a
B cell growth factor and basophil activator. This basophil activator is
expressed in myeloid cells and is involved in delayed-type allergic
responses (39). Connexin 43, the most abundant of all SAGE
tags in this study, may also indicate a connection with macrophages,
since activated macrophages express this gap junction protein
(40). Connexin 43 may allow T cells to form gap
junctions with epithelial and/or endothelial cells.
An additional association with the myeloid lineage revealed by our
study was the finding that T cells express the myeloid cell
transcription and differentiation factor BLIMP-1. Due to this link, we
concentrated on this observation and confirmed BLIMP-1 expression in
bovine T cells by real-time RT-PCR and a quantitative RNase
protection assay. Again, this finding does not appear to be restricted
to bovine cells. Fahrer and colleges (17) using cDNA
arrays demonstrated BLIMP-1 expression in the intraepithelial T
cells of mice, but not in T cells isolated from the same
location. These findings are surprising, since the expression of
BLIMP-1 was initially described as an important transcriptional
repressor in B cells, driving B cell differentiation into a plasma
cell. In addition, BLIMP-1 expression has been shown to be a trigger
for differentiation of the myeloid lineage. Two important targets of
this repressor have been so far described: c-Myc and CIITA, a
coactivator for MHC II complex (41, 42, 43). BLIMP-1
expression in cells is repressed by BCL-6, which also represses CD44
expression (44). The observation that CD44 is highest in
CD8- T cells, in correlation with higher
BLIMP-1 expression in this subset, suggests that the regulation of
BLIMP-1 in T cells might be similar to that described in B cells
and myeloid cells. Therefore, BLIMP-1 might be functionally important
during terminal differentiation of at least some subsets of T
cells, and current studies are focused on defining the control elements
involved in its expression.
These new observations support the idea that T cells are an
important bridge between the innate and acquired immune systems
(45). Not only do these cells localize in tissues similar
to myeloid cells and produce factors that direct the activity of
myeloid cells, but they also have apparently retained many of the
functional attributes of neutrophils and monocytes. They are recruited
to sites of inflammation via adhesion molecules similar to those used
by myeloid cells. Unlike naive T cells, all bovine
CD8- T cells in newborns express
E-selectin ligands and migrate to acute and chronic sites of
inflammation, like myeloid cells (46, 47). Bovine T
cells are potentially phagocytic via scavenger receptor I and might
function as professional APC (48, 49, 50). Recently, Richards
and Nelson (51) postulated, based on a phylogenetic
analysis, that the TCR sequences are the most ancient of the Ag
recognition molecules of lymphocytes, and that Ag recognition
properties of cells (TCR) and B cells (Ig) arose from an ancient
T cell. Our findings support this hypothesis and provide
additional direct examples of potential functional links between innate
(myeloid-derived) and adaptive (lymphoid-derived) immunity provided by
T cells, which warrant further investigation.
Finally, the results of this study as well as results from the only
other genomic analyses performed on T cells to date (17, 18) must be considered in the context of the animal systems in
which they were performed (ruminants and rodents). The evolutionary
conservation of the T cell population suggests that common
findings between these animal groups will probably extend to humans,
but it will be important to perform similar types of studies with human
cells. Our results show that analysis of circulating cells, vs
isolation of tissue cells, can provide unique insights into the
functional differences of T cell subsets.
|
Acknowledgments |
|---|
T cell subsets, Diane Cockrell for preparing T
cell RNA, and Jim Thompson and Kerri Rask for collection of all blood
samples. We also thank Dr. James Fox for providing sequencing space and
organizing the sequencing of the CD8+ SAGE library
at U.S. Department of Agriculture Meat Animal Research Center (Clay
Center, NE), as well as Dr. Ed Schmidt for technical advice and
support. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Mark A. Jutila, Veterinary Molecular Biology, Montana State University, Bozeman, MT 59717. E-mail address: uvsmj{at}montana.edu
3 Abbreviations used in this paper: SAGE, serial analysis of gene expression; BLIMP-1, B lymphocyte-induced maturation protein-1; EST, expressed sequence tag; HS, horse serum; IEL, intraepithelial lymphocyte.
Received for publication June 6, 2002. Accepted for publication November 4, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
antigen receptor genes of Thy-1+ dendritic epidermal cells. Cell 55:837.[Medline]
cells with disrupted primary V gene usage. Science 279:1729. T cells in allergic airway inflammation. Science 280:1265.3+ fetal murine thymocytes home to the skin and give rise to proliferating dendritic epidermal T cells. J. Immunol. 146:2536.[Abstract]
T cell subsets defined by anti-T cell receptor for antigen antibodies. Immunol. Lett. 64:39.[Medline]
thymocyte subset with homogeneous T-cell receptors to mucosal epithelia. Nature 343:754.[Medline]
1+ T cell adherence and tissue retention. J. Immunol. 155:1117.[Abstract]
and interleukin-4 in response to Th1- and Th2-stimulating pathogens by T cells in vivo. Nature 373:255.[Medline]
T cells cytotoxic for activated macrophages. J. Exp. Med. 191:2145. T cell subset can increase host resistance to a bacterial infection. J. Immunol. 165:6472.1+ T cells suppress and V4+ T cells promote susceptibility to coxsackievirus B3-induced myocarditis in mice. J. Immunol. 165:4174. T cells in the ruminant immune system. Immunol. Today 12:30.[Medline]
T cell subset, which is found in large numbers in the spleen, accumulates inefficiently in an artificial site of inflammation: correlation with lack of expression of E-selectin ligands and L-selectin. J. Immunol. 162:4914. intraepithelial lymphocytes as suggested by their transcriptional profile. Proc. Natl. Acad. Sci. USA 98:10261.+ and TCR+ intraepithelial lymphocytes provided by serial analysis of gene expression (SAGE). Immunity 15:419.[Medline]
T cell-specific monoclonal antibody (GD3.5): biochemical comparisons of GD3.5 antigen with the previously described Workshop Cluster 1 (WC1) family. J. Immunol. 156:3772.[Abstract]
T cell subsets express distinct patterns of chemokine responsiveness and adhesion molecules: a mechanism for tissue-specific T cell subset accumulation. J. Immunol. 169:4970.-chain mRNA and protein in / T-cells from euthymic and athymic rats: implications for T-cell lineage divergence. Dev. Immunol. 8:19.[Medline]
/ T cells with unique surface phenotype and tissue distribution. Immunology 92:340.[Medline]
on T cell rolling and tight adhesion to monolayers of activated endothelial cells. J. Immunol. 164:5035. T-cell bridge: linking innate and acquired immunity. Nat. Med. 4:764.[Medline]
/ T cells. J. Immunol. 153:3917.[Abstract]
T cell interactions with E-, P-, and L-selectin: characterization of lymphocyte on lymphocyte rolling and the effects of O-glycoprotease. J. Immunol. 156:289.[Abstract]
T cells present antigen to CD4+ T cells. J. Leukocyte Biol. 63:707.[Abstract]
T cells. Immunology 82:229.[Medline]
T
lymphocyte subsets defines unique tissue-specific functions. J.
Leukocyte Biol. In
press.This article has been cited by other articles:
|
|
 |
|
  T. L Davis and J. L Pate Bovine Luteal Cells Stimulate Proliferation of Major Histocompatibility Nonrestricted Gamma Delta T Cells Biol Reprod, December 1, 2007; 77(6): 914 - 922. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K. Lahmers, J. F. Hedges, M. A. Jutila, M. Deng, M. S. Abrahamsen, and W. C. Brown Comparative gene expression by WC1+ {gamma}{delta} and CD4+ {alpha}{beta} T lymphocytes, which respond to Anaplasma marginale, demonstrates higher expression of chemokines and other myeloid cell-associated genes by WC1+ {gamma}{delta} T cells. J. Leukoc. Biol., October 1, 2006; 80(4): 939 - 952. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Berndt, J. Pieper, and U. Methner Circulating {gamma}{delta} T Cells in Response to Salmonella enterica Serovar Enteritidis Exposure in Chickens Infect. Immun., July 1, 2006; 74(7): 3967 - 3978. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Graff, M. Behnke, J. Radke, M. White, and M. A. Jutila A comprehensive SAGE database for the analysis of {gamma}{delta} T cells Int. Immunol., April 1, 2006; 18(4): 613 - 626. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Chtanova, R. Newton, S. M. Liu, L. Weininger, T. R. Young, D. G. Silva, F. Bertoni, A. Rinaldi, S. Chappaz, F. Sallusto, et al. Identification of T Cell-Restricted Genes, and Signatures for Different T Cell Responses, Using a Comprehensive Collection of Microarray Datasets J. Immunol., December 15, 2005; 175(12): 7837 - 7847. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Hedges, K. J. Lubick, and M. A. Jutila {gamma}{delta} T Cells Respond Directly to Pathogen-Associated Molecular Patterns J. Immunol., May 15, 2005; 174(10): 6045 - 6053. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. N. Rogers, D. G. VanBuren, E. E. Hedblom, M. E. Tilahun, J. C. Telfer, and C. L. Baldwin {gamma}{delta} T Cell Function Varies with the Expressed WC1 Coreceptor J. Immunol., March 15, 2005; 174(6): 3386 - 3393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Reverter, S. M. McWilliam, W. Barris, and B. P. Dalrymple A rapid method for computationally inferring transcriptome coverage and microarray sensitivity Bioinformatics, January 1, 2005; 21(1): 80 - 89. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Skeen, M. M. Freeman, and H. K. Ziegler Changes in peritoneal myeloid populations and their proinflammatory cytokine expression during infection with Listeria monocytogenes are altered in the absence of {gamma}/{delta} T cells J. Leukoc. Biol., July 1, 2004; 76(1): 104 - 115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Xu, C.-T. Shu, E. Purdom, D. Dang, D. Ilsley, Y. Guo, J. Weber, S. P. Holmes, and P. P. Lee Microarray Analysis Reveals Differences in Gene Expression of Circulating CD8+ T Cells in Melanoma Patients and Healthy Donors Cancer Res., May 15, 2004; 64(10): 3661 - 3667. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Hedges, J. C. Graff, and M. A. Jutila Transcriptional Profiling of {gamma}{delta} T Cells J. Immunol., November 15, 2003; 171(10): 4959 - 4964. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
