| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Section of Immunobiology, Yale University School of Medicine, and Howard Hughes Medical Institute, New Haven, CT 06511
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Memory CD4 T cells differ from naive CD4 T cells in their expression of
several cell surface molecules. Naive CD4 T cells that express high
levels of L-selectin and CD45RB and low levels of CD44
(L-selectinhigh, CD45RBhigh,
CD44low) (2–6; reviewed in Refs. 1 and 7) modulate their
surface phenotype upon activation, leading to the reciprocal pattern of
expression (L-selectinlow, CD45RBlow,
CD44high) as well as an up-regulation of LFA-1, ICAM-1,
CD43, and the 
4ß1 and
4ß7 integrins (8, 9; reviewed in Refs. 1
and 7). To gain insight into whether the differential expression of
these molecules contributes to the differences between primary and
memory responses, it is important to know whether these phenotypic
changes represent stages of differentiation of CD4 T cells, or if the
changes reflect a transient state of activation resulting from
recent contact with Ag. The L-selectinlow,
CD45RBlow, CD44high phenotype on resting
memory cells has been demonstrated in vivo and in vitro and is
supported by the observation that the memory phenotype is not seen in
early life, but increases with age (10, 11, 12). However, while some
studies suggest that the phenotype of resting memory cells is stable,
other studies indicate that a reversion from memory to naive phenotype
can occur in CD4 T cells. Studies in irradiated humans suggest that
there is a progressive reaccumulation of CD45RA+ (naive
phenotype) CD4 T cells with time (13), as have studies in rats
demonstrating reversion of memory phenotype, CD45RC-, to a
naive phenotype, CD45RC+, following adoptive transfer in
the absence of Ag (14). Further studies suggest that the reversion can
be prevented by the presence of Ag (15). Similarly, it has been shown
in mice that a significant proportion of CD4 T cells, dividing in
response to Mlsa Ags, carries the naive phenotype (16).
Thus, the stability of naive and memory phenotype and the role of Ag in
resting CD4 T cells remain contentious issues.
While the above studies indicate that a reversion from memory to naive phenotype can occur, there is evidence that effector CD4 T cells can express phenotypic markers that are expressed on naive CD4 T cells (17) and, further, that there are no known phenotypic differences between primary effector and secondary effector CD4 T cells (7). This suggests the possibility that memory phenotype CD4 T cells that appear to reacquire the naive phenotype may, in fact, be recently activated effector cells. In addition, in the murine Mlsa system, it is unclear whether cells responding to Mlsa reacquired the naive phenotype or did not change phenotype following contact with Mlsa Ag.
In this study the long term phenotypes of resting naive and memory CD4
T cells in mice are analyzed over time in an adoptive transfer model,
using both TCR transgenic
(Tg)3 and nontransgenic
donors, to determine under what circumstances the phenotype will change
and how this reflects the functional capabilities of the transferred
CD4 T cells. Adoptive transfer of CD4 T cells from TCR Tg mice allows
the monitoring of naive and memory phenotypes in CD4 T cells expressing
the transgenic TCR-ß pair, which have little potential to respond
to environmental Ag (18), and CD4 T cells bearing polyclonal TCRs,
which do have the potential to respond to an environmental Ag. Our
findings indicate that as a population, memory phenotype CD4 T cells,
whether a polyclonal population from normal mice or an Ag-specific
population from TCR Tg mice, can retain the memory phenotype over time
with no evidence of permanent reversion. Transferred populations of
naive CD4 T also retain the naive phenotype over time, and a memory
phenotype is only observed in CD4 T cells that have the potential for
antigenic exposure. Furthermore, the phenotypically converted cells
function as memory CD4 T cells by producing the mature memory cytokines
IL-4 and IFN-
. In addition, we find that the presence of Ag is not
necessary for maintaining a memory phenotype.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The AND TCR-ß transgenic mice (TCR specific for pigeon
cytochrome c) were derived from a heterozygous mouse
obtained from J. Kaye (Scripps Institute, La Jolla, CA) (19).
Subsequent generations of TCR transgenic mice were maintained as
heterozygotes on a (B10.A(5R) x B6)F1 background.
B6.PL-Thy1a/Cy mice were obtained from The Jackson
Laboratory (Bar Harbor, ME), and B6-LY5.2/Cr mice were obtained from
the National Institutes of Health (Frederick, MD). Both were bred and
maintained as (B10.A(5R) x B6)F1 in animal facilities at
Yale.
Preparation of naive and memory CD4 T cells for adoptive transfer
Unless otherwise stated, mAbs used in cell preparations were purified from supernatants of hybridomas maintained in this laboratory using standard protein A or protein G affinity chromatography. CD4 T cells were isolated from the spleens and lymph nodes of donor animals by negative selection, using mAbs against CD8 (clones 53-6.72 and 2.43; American Type Culture Collection, Manassas, VA), MHC class II I-A (clone 212.A1) (20), and FcR (clone 2.4-G2; American Type Culture Collection), followed by incubation with anti-mouse and anti-rat Ig-coated magnetic beads (Collaborative Research, Bedford, MA). The CD4 T cells were then labeled with anti-CD45RB-biotin Ab (clone 16A) (3) (PharMingen, San Diego, CA), followed by incubation with streptavidin-magnetic microbeads (Miltenyi Biotech, Sunnyvale, CA). Naive CD4 T cells were positively selected on MACS separation columns (Miltenyi Biotech). In separate preparations, memory CD4 T cells were negatively selected on MACS columns.
Generation of TCR transgenic memory CD4 T cells for adoptive transfer
Splenic CD4 T cells were isolated from AND transgenic mice by
negative selection and were cultured in vitro with APCs to make
effector cells. APCs were syngeneic T-depleted spleen cells prepared by
incubating them with mAbs against Thy1 (clone Y-19) (21), CD4 (clone
GK1.5; American Type Culture Collection), and CD8 (clone 53-6.72)
followed by complement-mediated lysis (JRH Biosciences, Lenexa, KS) and
concurrent treatment with mitomycin C (50 mg/ml; Boehringer Manheim,
Indianapolis, IN). The T cells were cultured with APCs at a 1:2 ratio
of T:APC for 4 days in the presence of 5 µg/ml pMCC (residues 81–103
of tobacco hornworm moth cytochrome c), 10 U/ml murine rIL-2
(Boehringer Mannheim), 200 U/ml murine rIL-4 (Collaborative Biomedical
Products, Bedford, MA), and 10 µg/ml anti-IFN- Ab (clone XMG
1.2; American Type Culture Collection) in EHAA medium (Life
Technologies, Grand Island, NY) supplemented with 5% FBS (Irvine
Scientific, Santa Ana, CA). The effector T cells were washed
extensively, and 1–1.3 x 107 cells were adoptively
transferred into sublethally irradiated (650 rad) euthymic (B10.A(5R)
x B6.PL-Thy1a/Cy)F1 mice. After 4–6 wk memory
cells were recovered from these recipient mice as described above,
using MACS column separation, and transferred into new hosts for
phenotypic analysis.
Adoptive transfers
Recipient mice used in phenotypic analyses and functional studies underwent a surgical thymectomy at 5–6 wk of age and were allowed to recover for at least 2 wk before being used in experiments. On the day of transfer, recipient mice were lethally irradiated (900 rad) and reconstituted within 8 h with 5 x 106 syngeneic T-depleted bone marrow cells and 1.5 x 107 naive or 1–1.5 x 107 memory CD4 T cells. Quantitation of cell recovery was estimated at various time points for individual animals as follows. Single cell suspensions of spleen and a pool of mesenteric, inguinal, axilary, and brachial lymph nodes were prepared. Spleen and lymph node suspensions were counted to determine the total number of cells. The total number of donor CD4 T cells was calculated from the frequency of donor CD4 T cells determined by flow cytometric analysis and the total number of cells from each organ. Cell numbers in the PBLs of recipients were estimated as follows. Samples of blood from individual animals were counted using Turk’s solution to exclude RBCs to determine total cell numbers per microliter of blood. The number of donor CD4 T cells per microliter of blood and subsets thereof were calculated from the frequency of the cell subsets determined by flow cytometric analysis and the total number of cells per microliter of blood.
Flow cytometry
mAbs used to stain cell surface molecules were: FITC-labeled
anti-V11 TCR (clone RR8-1; PharMingen), FITC-labeled
anti-CD45.1 (clone A20; PharMingen), FITC-labeled anti-Thy1.1
(clone OX-7; PharMingen), biotin-labeled anti-CD45RB (clone 16A;
PharMingen), PE-labeled Vß3 TCR (clone KJ25; PharMingen), PE-labeled
anti-L-selectin (clone Mel-14; PharMingen), PE-labeled
anti-CD44 (clone IM7; PharMingen), and Quantum Red-labeled
anti-CD4 (clone H129.19; Sigma, St. Louis, MO). Bound
biotin-labeled anti-CD45RB was detected with Texas Red-avidin
(Vector Laboratories, Burlingame, CA). RBCs were eliminated from PBLs
before staining by osmotic lysis. Cells were fixed in 1%
paraformaldehyde before analysis. Four-color analysis and FACS
sorting were performed on a FACStar system (Becton Dickinson, Mountain.
View, CA). FACS data were analyzed using LYSYS software and CellQuest
software (Becton Dickinson).
Cell cultures for cytokine assays
Fifteen weeks posttransfer, CD4 T cells were purified as
described above from pooled spleens of mice that received naive CD4 T
cells from TCR transgenic mice. The CD4 T cells were FACS sorted into
two populations: naive phenotype donor cells (Ly5.1-,
CD45RBhigh, CD44low) and memory phenotype donor
cells (Ly5.1-, CD45RBlow,
CD44high). The sorted cells were 97 and 94% pure,
respectively. APCs were syngeneic T-depleted spleen cells prepared as
described above. The purified naive and memory donor cells were each
cultured with APCs at a 1:2 ratio of T:APC in the presence of pMCC (5
µg/ml) or Con A (1 µg/ml; Pharmacia Biotech, Piscataway, NJ) in
EHAA medium supplemented with 5% FBS. To measure proliferation,
cultures were pulsed after 24 h with 1 mCi/well
[3H]TdR (ICN, Irvine, CA) in flat-bottom 96-well plates
(Costar, Cambridge, MA) and harvested at 48 h onto glass-fiber
filters. The incorporated radioactive thymidine was measured by liquid
scintillation counting. From parallel cultures, supernatants were
collected after 72 h and assayed for IL-4 and IFN- using ELISA
kits (Endogen, Woburn, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine whether cell surface phenotypes of naive CD4 T cells are stable over time, purified populations of naive CD4 T cells were adoptively transferred into adult thymectomized, lethally irradiated, bone marrow-reconstituted (ATxBM) Ly5.1+ recipients. It is known that adoptively transferred T cells can undergo considerable expansion when transferred in relatively small numbers (22, 23, 24, 25, 26), with memory T cells showing a greater potential for expansion than naive T cells (26, 27, 28). As it has been shown that the final peripheral pool size will reconstitute to the same degree regardless of the starting cell number (23, 25), we transferred a relatively large number of CD4 T cells (1.5 x 107) to our T-deficient hosts to establish stable populations with minimal expansion.
Following adoptive transfer, the surface phenotype of PBLs was
monitored over time by FACS analysis, as determined by expression of
CD45RB and L-selectin. Naive Ly5.2+ CD4 donor cells
(CD45RBhigh, L-selectinhigh) were isolated from
cytochrome c-specific TCR Tg mice. While the vast majority
(88–94% in all experiments) of donor CD4 T cells expressed the
V11/Vß3 transgenic TCR whose Ag is not present in the environment,
these mice were not crossed onto to a recombination-activating
gene-/- background; therefore, the starting population of
donor cells contains a very small proportion of CD4 T cells that does
not express the transgene, allowing us to examine the influence of
environmental Ag on these donor naive CD4 T cells. At the time of
transfer, approximately 90% of the CD4 T cells expressed high levels
of both CD45RB and L-selectin (Fig. 1
A). Thymectomy of the
recipients appeared to be complete, as no host-derived CD4 T cells
could be detected by staining of PBLs with Ab against the Ly5.1 allele
during the analysis period (data not shown). By 7 wk posttransfer,
analysis of CD4+ PBLs showed that half of the total donor
population (CD4+, Ly5.1-) had lost the naive
phenotype, becoming CD45RBlow. Of these, 26% acquired the
memory phenotype, defined here as CD45RBlow,
L-selectinlow. The remaining donor CD4 T cells had an
intermediate phenotype, perhaps indicating transitional stages. A
summary of multiple time points from individual animals in five
experiments is shown in Fig. 1B. Phenotypic changes occur as
early as 3 wk posttransfer, with a dramatic loss of both CD45RB and
L-selectin expression by 15 wk posttransfer. It should be noted that
the change appears stable, in that the naive phenotype of the donor
cells is not significantly restored over time.
|
Donor CD4 T cells that maintain expression of the V11 TCR
transgene maintain the naive phenotype following transfer
The starting population of naive CD4 donor T cells expresses the
V11/Vß3 transgenes on 88–94% of cells and can respond well to
moth or pigeon cytochrome c. However, the appropriate Ag is
not present in the environment in that few, if any, memory cells
expressing the V11/Vß3 transgenes develop in these mice (18). CD4
T cells with a memory phenotype can be isolated from TCR Tg mice;
however, they express transgenic Vß-chains paired with endogenous
V-chains, creating TCRs that may be responsive to environmental Ags
(18, 29). Therefore, it was of interest to determine whether the
accumulation of donor CD4 T cells with a memory phenotype following
transfer of naive CD4 T cells was derived from donor CD4 T cells
expressing endogenous V-chains. To determine this, we further
dissected the naive CD4 T cell donor population at each time point
posttransfer to determine whether cells displaying the memory phenotype
expressed the V11 transgene. In Fig. 2A, the expression of V11
in the starting population of naive donor CD4 T cells is compared with
V11 expression in the donor cells of an animal 12 wk posttransfer.
The majority of the starting donor CD4 population has high expression
of V11, which is greatly diminished 12 wk following transfer. There
is a large proportion of V11-negative donor CD4 T cells as well as a
population of donor CD4 T cells exhibiting an intermediate level of
expression of V11. These intermediate V11-expressing cells may
have also up-regulated expression of a second TCR that has been
documented in these transgenic mice (18, 30) (our own unpublished
observations) as well as in other transgenic systems (29, 31, 32, 33). For
this reason, the V11 intermediate cells were treated as
V11-negative CD4 T cells in examining expression of the transgene
within the total donor population in all subsequent analyses. Fig. 2B (left panel) shows that the percentage
of V11high donor CD4 T cells within the total donor
population drops very early following adoptive transfer and continues
to fall over time. This was not due to a selective and continual
expansion of V11-negative CD4 T cells within the donor population
while the V11+ population remained constant, as
quantitation of donor cell recovery at various time points showed that
the total donor population remained relatively stable and actually
slightly declined at later time points (Fig. 2B, right
panel). This suggests that the shift to the memory phenotype was a
property of donor cells expressing endogenous V-chains. However,
most interestingly, those donor cells that maintained the V11
transgene were primarily naive by phenotype. This is illustrated in
Fig. 3, which compares the numbers of
naive and memory phenotype donor cells among the V11+
and V11- donor CD4 T cells recovered from spleen and
lymph nodes of recipients at a few representative time points after
transfer. There is a decrease in the number of V11+
donor cells expressing the naive phenotype (Fig. 3A, solid
bars). Most of these cells are not entering the V11+
memory pool in that only a few V11+ donor cells (<7%
of the total donor population) have converted to the memory phenotype
(Fig. 3A, hatched bars). Furthermore, there is an increase
in the number of V11- cells expressing the memory
phenotype (Fig. 3B, hatched bars), while a similar increase
in memory phenotype is not seen in the V11+ donor cells
(Fig. 3A, hatched bars). This suggests that in the absence
of Ag, naive CD4 T cells will maintain the naive phenotype, and only
those donor cells that have the capability of responding to
environmental Ags will take on and maintain the memory phenotype. At
this time, it is unclear whether the V11- cells that
have the memory phenotype arose from naive donor cells that were
stimulated and changed to the memory phenotype or were a small
population of memory phenotype cells that accompanied the original
naive cell innoculum, which subsequently expanded.
|
|
11
transgene. The vast majority of V11+ donor cells
maintains the naive phenotype in the absence of the nominal Ag. This
suggests that the phenotypic change from naive to memory is due to
antigenic challenge present in the environment, and maintenance of the
naive phenotype is mediated by maintenance of the transgenic TCR whose
Ag is not present in the environment. Furthermore, the
transferred cells that displayed the memory phenotype retained this
phenotype over time without evidence of reversion. Naive donor CD4 T cells that convert to a memory phenotype function as memory CD4 T cells, while those that retain the naive phenotype still function as naive CD4 T cells
Because a change of phenotype was seen in the total donor
population of cells transferred from a TCR Tg donor, it was important
to know whether this also reflected a change in the functional
capabilities of these cells. This was a particularly interesting point,
since there was a small proportion V11+ donor cells that
changed to a memory phenotype in the presumed absence of Ag. To do
this, CD4 T cells were recovered from the spleens of animals that
received naive CD4 T cells from a TCR Tg donor at 15 wk following
transfer. At this time point, it was possible to recover splenic memory
cells that were comprised of a significant enough proportion of
V11+ memory cells to assess whether they were capable of
functioning as memory CD4 T cells in response to specific Ag,
cytochrome c peptide (pMCC). The donor CD4 T cells were
sorted into two populations: one displaying a naive phenotype and a
second displaying a memory phenotype. The sorted donor cell populations
were then cultured in vitro with APCs and either pMCC to stimulate
V11+ donor cells or Con A to stimulate the total CD4 T
cell population. The functional status of the donor cells was assessed
by measuring the production of IL-4 and IFN-, cytokines
characteristic of memory CD4 T cells (reviewed in 34 . Both sorted
donor populations proliferated in response to pMCC and Con A (Fig. 4). When stimulated with pMCC, the sorted
naive CD4 T cells did not produce significant levels of the memory
cytokines IL-4 and IFN-, although they responded to pMCC and Con A
by proliferation (Fig. 4). The memory phenotype donor cells produced
high levels of IL-4 and IFN- following stimulation by both pMCC and
Con A. These data indicate that CD45RBhigh
L-selectinhigh CD4 T cells respond like naive CD4 T cells,
producing IL-2 without producing IL-4 and IFN-. When a change in
phenotype to CD45RBlow L-selectinlow takes
place, the CD4 T cells produce IL-4 and IFN-, characteristic of
becoming a memory CD4 T cell.
|
In the previous experiments, populations of donor CD4 T cells with
a memory phenotype gradually appeared among donor cells that did not
express the V11/Vß3 transgenic TCR pair and appeared to maintain
the memory phenotype. However, this population may have a limited range
of specificities due to pairing of the Tg Vß3 chain with endogenous
V-chains, making it difficult to predict whether they are capable of
responding to environmental Ags. Thus we were interested in examining
whether memory phenotype is maintained in polyclonal populations of
memory CD4 T cells with diverse specificities. In experiments parallel
to the naive TCR Tg donor cell transfers, memory CD4 T cells
(CD45RBlow, L-selectinlow) were donated from
nontransgenic mice (Ly5.2+) to ATxBM recipients
(Ly5.1+), and maintenance of the memory phenotype in PBLs
was examined over time. As seen in Fig. 5A, approximately 85% of the
donor cells were of the memory phenotype (CD45RBlow,
L-selectinlow). In contrast to the changes seen in the
naive donor CD4 T cells, the memory donor CD4 T cells did not change
their phenotypic profile (Fig. 5, A and B).
|
were from a nontransgenic
donor and were, therefore, a polyclonal population of memory CD4 T
cells capable of responding to environmental Ags, it was possible that
the memory phenotype was retained by periodic stimulation of the memory
CD4 T cells. To address the question of whether changes in the
phenotype of memory CD4 T cells might take place in the absence of Ag,
it was necessary to obtain a population of memory CD4 T cells
(CD45RBlow, L-selectinlow) from the cytochrome
c-specific TCR Tg mice. It is known that immunization of TCR
Tg mice by conventional methods such as administration of peptide with
an appropriate adjuvant does not give rise to a pool of resting memory
cells within the TCR Tg mouse (35) (our unpublished observations). In
addition, TCR Tg CD4 T cells stimulated in vitro do not appear to
acquire all the characteristics of memory CD4 T cells (7) (our
unpublished observations). Therefore, we adapted a previously published
method to generate a pool of resting memory CD4 T cells in vivo (36) to
transfer into our T-deficient hosts. Briefly, CD4 T cells from TCR Tg
mice were stimulated in vitro to produce effector CD4 T cells. These
cells were then parked in sublethally irradiated, euthymic,
intermediate hosts congenic for the Thy1.1 allele to allow the
activated Tg+ cells to become resting memory CD4 T cells.
Four to six weeks following the intermediate transfer, when sufficient
numbers of Tg+ CD4 T cells had become resting
CD45RBlow memory cells as judged by FACS analysis of PBLs,
memory CD4 T cells were isolated from the spleens of the intermediate
hosts as CD45RBlow CD4 T cells. Approximately 30% of the
donor memory CD4 T cells in these preparations were transgene positive,
and the remaining memory cells were derived from the intermediate host.
All the memory CD4 T cells in these preparations were resting cells as
judged by forward and side light scatter on the FACS with a
CD45RBlow, L-selectinlow, CD44high
phenotype. These memory CD4 T cells were then transferred to their
final ATxBM recipients for the phenotypic analysis over time. For these
analyses, memory phenotype was assessed in terms of CD45RB phenotype
alone due to the need to include markers to distinguish both donor
types (Ly5.2, Thy1.2 Tg CD4 T cells and Ly5.2, Thy1.1 non-Tg CD4 T
cells) from the recipient type (Ly5.1, Thy1.2). Fig. 6A shows that similar to the
polyclonal TCR Tg-negative memory CD4 T cells, the TCR Tg+
memory CD4 T cell population can retain the memory phenotype through
the first 10 wk posttransfer. It should be noted in Fig. 6A
that two mice do not fully retain the CD45RBlow memory
phenotype. However, this does not reflect a permanent change, as it is
only seen for each animal at one time point and not at subsequent time
points. This is illustrated for one such recipient in Fig. 6B, where the reappearance of the naive phenotype appears be
an outgrowth of a group of Tg+ cells. These cells may have
somehow become activated (perhaps through a second TCR) and may
represent effector cells, which, like naive cells, can express high
levels of CD45RB (7, 17) (our unpublished observations).
|
B). The Tg+ donor
CD4 T cells decrease in frequency among the total donor population with
time and can barely be detected 10 wk posttransfer. This may be due to
the outgrowth of the polyclonal memory cells diluting out the
Tg+ donor cells, thus placing them below the limits of
detection in our assay. Another possibility is that the Tg+
memory cells do not persist in the absence of Ag and can no longer be
found in our recipients. However, without further study it is difficult
to make any strong conclusions regarding this observation. The
important point that can be made by these experiments is that memory
CD4 T cells are capable of retaining the memory phenotype over time,
independent of the presence or absence of Ag, without evidence of
permanent reversions. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
11/Vß3 transgenic TCR. Those donor cells that express
high levels of the transgenic TCR maintain the naive phenotype. The
second point is that conversion to the memory phenotype is associated
with a change in the production of cytokines upon activation,
reflective of memory function. In contrast to naive CD4 T cells, the
memory phenotype of CD4 T cells (CD45RBlow
L-selectinlow) is stable over time after adoptive transfer,
with no reaccumulation of the naive phenotype (CD45RBhigh
L-selectinhigh).
A significant proportion of naive CD4 T cells from a TCR Tg donor
converted to a memory phenotype. However, further dissection of this
donor population revealed that the change could largely be accounted
for by donor cells that did not express high levels of the V11
transgene, and those that maintained high expression of the transgene
also maintained the naive phenotype. Interestingly, those donor cells
with high expression of the V11 transgene gradually became a very
small proportion of the total donor population. This was not due to a
large expansion of the transgene-negative donor cells resulting in a
lower frequency of transgene-positive cells, since the number of total
donor CD4 T cells did not gradually increase with time. One possible
explanation for this finding is that the donor cells that expressed
high levels of the transgene simply died over time, while the few
V11- cells initially present in the starting population
responded to environmental Ags and in doing so were allowed to
proliferate and survive. It is unlikely that the expansion of the
Tg- donor cells directly caused the disappearance of the
Tg+ donor cells because the Tg- cells were of
a memory phenotype, and it has been suggested that the expansion
potentials of naive phenotype and memory phenotype CD4 T cells appear
to be independent of one another (Fig. 3C) (27). A more
likely explanation is that the donor cells with high expression of the
transgene were under environmental pressure to down-modulate the V11
TCR while up-regulating other TCR -chains to respond to
environmental Ags. This possibility is supported by the observation
that unlike the starting population, a large proportion of donor CD4 T
cells show lower, but not negative, expression of V11 over time. It
has been shown in the AND transgenic mice as well as in other
transgenic systems that a second TCR can be expressed on a small number
of TCR Tg T cells (18, 29, 30, 31, 32, 33) (our unpublished observations). In one
such system, TCR Tg CD4 T cells with a memory phenotype and displaying
memory function were observed, but when the mice were crossed to a
recombination-activating gene knockout background, these memory
phenotype CD4 T cells could no longer be found, suggesting that the
switch to a memory phenotype was caused by a second TCR responding to
Ag (29). This could also explain the observation in this study that
there is a small population of V11+ donor cells
displaying a memory phenotype and also exhibiting memory function by
production of IL-4 and IFN-. In fact, it has been shown that CD4 T
cells can be cloned out of TCR Tg mice that are capable of responding
to an Ag other than the specific Ag for the transgenic TCR (31). The
possibility of cross-reactive Ags cannot be ruled out as a cause of the
appearance of Tg+ memory cells. However, we believe that
this is unlikely, as this population did not increase with time and in
all but one animal represented <6% of the total donor population.
The appearance of a large number of memory phenotype Tg-
donor CD4 T cells may be due to environmental pressures causing
expansion of a small contaminating population of memory phenotype
Tg- CD4 T cells in the starting innoculum or may be due to
exposure to Ag, driving the change in phenotype accompanied by
expansion of these activated cells. Indeed, it has been suggested that
homeostatic mechanisms exist to maintain a particular ratio of naive
and memory CD4 T cells within an animal (14), and therefore, in
situations where almost pure populations of naive CD4 T cells are
transferred, there may be a need to generate a population of memory
cells, whether driven by Ag or not. However, we favor the argument that
the expansion of the memory phenotype cells is Ag driven due to the
fact that the small population of V11+ memory phenotype
CD4 T cells does not continue to grow (Fig. 3A), while the
V11- memory population continues to expand. However,
regardless of the origin of these memory phenotype donor cells, the
fact remains that they maintain the memory phenotype over time, without
restoration of the naive phenotype.
Transferred CD4 T cells of a memory phenotype showed no significant reversion to a naive phenotype as determined in three types of experiments. First, a large proportion of the naive CD4 T cells converted to a memory phenotype without gradual reversion to the naive phenotype. Second, transfers of polyclonal memory cells led to a stable donor population in which no reversion or reaccumulation of naive phenotype was seen. This finding is in contrast to previous studies showing that as many as one-third of polyclonal memory phenotype CD4 T cells in rats appeared to reexpress the naive phenotype (14). One possible explanation is that CD4 T cells expressing the naive phenotype defined by CD45R expression are, in fact, only effector CD4 T cells, as effector cells can share this phenotype with naive CD4 T cells. What was not determined was whether the CD4 T cells also expressed other markers of previously activated cells. In our studies memory phenotype was assessed by more than one marker. Furthermore, the mice used in this study were maintained in filtered cages to minimize the presence of significant numbers of activated effector cells at any one time. Third, Ag-specific memory CD4 T cells derived from AND transgenic mice retained the CD45RBlow memory phenotype in most recipients. Two animals showed a transient re-expression of the naive CD45RBhigh phenotype for one time point each. This re-expression of CD45RB may be due to an environmental Ag stimulating a second TCR on these donor cells, thus producing effector cells that express CD45RB. Still another possibility is that the change in phenotype is not Ag driven and is an effect of bystander activation, which has been reported to occur in human CD4 T cells (37) and murine CD8 T cells (38). Importantly, it should be noted that this re-expression does not reflect a permanent reversion to a naive phenotype. Taken together, these studies indicate that memory CD4 T cells retain a memory phenotype independent of Ag.
These data support the idea that naive and memory CD4 T cells can be reliably identified by their expression of a particular combination of cell surface molecules. Because expression of some molecules is shared by naive and effector or by memory and effector CD4 T cells, it is important to examine more than one of these markers to effectively characterize the stage of differentiation (naive or memory) or the state of activation (effector) of CD4 T cells. Ag appears to be required to recruit naive CD4 T cells into the memory compartment, leading to a change in cell surface phenotype, but is not necessary to maintain the memory phenotype.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Kim Bottomly, LH404 Section of Immunobiology, Yale University School of Medicine, 310 Cedar Street, New Haven, CT 06511. E-mail address:
3 Abbreviations used in this paper: Tg, transgenic; ATxBM, adult-thymectomized, lethally irradiated, bone marrow reconstituted; PE, phycoerythrin; pMCC, moth cytochrome c peptide.
Received for publication June 1, 1998. Accepted for publication August 31, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4ß7 and Eß7 integrins on thymocytes, intestinal epithelial lymphocytes and peripheral lymphocytes. Eur. J. Immunol. 26:897.[Medline]
production. J. Immunol. 140:1401.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  S. Jennrich, B. A. Ratsch, A. Hamann, and U. Syrbe Long-Term Commitment to Inflammation-Seeking Homing in CD4+ Effector Cells J. Immunol., June 15, 2007; 178(12): 8073 - 8080. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mostbock, M. Catalfamo, Y. Tagaya, J. Schlom, and H. Sabzevari Acquisition of antigen presentasome (APS), an MHC/costimulatory complex, is a checkpoint of memory T-cell homeostasis Blood, March 15, 2007; 109(6): 2488 - 2495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Zielinski, S. N. Jacob, F. Bouzahzah, B. E. Ehrlich, and J. Craft Naive CD4+ T Cells from Lupus-Prone Fas-Intact MRL Mice Display TCR-Mediated Hyperproliferation Due to Intrinsic Threshold Defects in Activation J. Immunol., April 15, 2005; 174(8): 5100 - 5109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Qi, T. L. Denning, and L. Soong Differential Induction of Interleukin-10 and Interleukin-12 in Dendritic Cells by Microbial Toll-Like Receptor Activators and Skewing of T-Cell Cytokine Profiles Infect. Immun., June 1, 2003; 71(6): 3337 - 3342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mirshahidi, C.-T. Huang, and S. Sadegh-Nasseri Anergy in Peripheral Memory Cd4+ T Cells Induced by Low Avidity Engagement of T Cell Receptor J. Exp. Med., September 17, 2001; 194(6): 719 - 732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Vivien, C. Benoist, and D. Mathis T lymphocytes need IL-7 but not IL-4 or IL-6 to survive in vivo Int. Immunol., June 1, 2001; 13(6): 763 - 768. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Vratsanos, S. Jung, Y.-M. Park, and J. Craft Cd4+ T Cells from Lupus-Prone Mice Are Hyperresponsive to T Cell Receptor Engagement with Low and High Affinity Peptide Antigens: A Model to Explain Spontaneous T Cell Activation in Lupus J. Exp. Med., February 5, 2001; 193(3): 329 - 338. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
