Vol.2, No.1, 25-39 (2012) Open Journal of Immunology
http://dx.doi.org/10.4236/oji.2012.21004
Innate-like CD4 T cells selected by thymocytes
suppress adaptive immune responses against
bacterial infections
Yu Qiao1, Brian M. Gray1, Mohammed H. Sofi1, Laura D. Bauler1, Kathryn A. Eaton2,
Mary X. D. O’Riordan1, Cheong-Hee Chang1*
1Department of Microbiology and Immunology, University of Michigan, Ann Arbor, USA; *Corresponding Author:
2Unit for Lab Animal Medicine, Medical School, University of Michigan, Ann Arbor, USA
Received 21 September 2011; revised 16 December 2011; accepted 30 December 2011
ABSTRACT
We have reported a new innate-like CD4 T cell
population that expr esses cell su rface makers of
effector/memory cells and produce Th1 and Th2
cytokines immediately upon activation. Unlike
conventional CD4 T cells that are selected by
thymic epithelial cells, these CD4 T cells, named
T-CD4 T cells, are selected by MHC class II ex-
pressing thymocytes. Previously, we showed
that the presence of T-CD4 T cells protected mice
from airway inflammation suggesting an immune
regulatory role of T-CD4 T cells. To further un-
derstand the function of T-CD4 T cells, we in-
vestigated immune responses mediated by T-
CD4 T cells during bacterial infection because
the generation of antigen specific CD4 T cells
contributes to clearance of infection and for the
development of immune memory. The current
study shows a suppressive effect of T-CD4 T
cells on both CD8 and CD4 T cell-mediated im-
mune responses during Listeria and Helicobac-
ter infections. In the mouse model of Listeria
monocytogenes infection, T-CD4 T cells resulted
in decreasedfre quency of Listeria -specific CD8 T
cells and the killing activity of them. Further-
more, mice with T-CD4 T cells developed poor
immune memory, demonstrated by reduced ex-
pansion of antigen-specific T cells and high
bacterial burden upon re-infection. Similarly, the
presence of T-CD4 T cells suppressed the gen-
eration of antigen-specific CD4 T cells in Heli-
cobacter pylori infected mice. Thus, our studies
reveal a novel function of T-CD4 T cells in sup-
pressing anti-bacterial immunity.
Keywords: Bacterial Inf e ction; Innate-Like CD4 T
Cells; Immune Suppression
1. INTRODUCTION
It is well established that conventional murine CD4 T
cells are selected on thymic epithelial cells (TEC) that
express MHC class II [1]. In humans, studies have shown
that CD4 T cells can be generated by a pathway that is
independent of TEC-expressed MHC class II [2-4]. The
alternative cell types supporting CD4 T cell development
seem to be hematopoietic cells, particularly thymocytes
[5-7]. Using mouse models, we and others have demon-
strated that indeed MHC class II-expressing thymocytes
successfully mediate CD4 T cell selection independent of
TEC-expressed MHC class II [8,9]. CD4 T cells selected
by MHC class II on thymocytes are called thymocyte-
selected CD4 (T-CD4) T cells, as distinguished from
conventional epithelial cell-selected CD4 (E-CD4) T
cells.
E-CD4 T cells are well studied and known to modulate
adaptive immunity by differentiating into helper cell
subsets and producing cytokines according to environ-
mental signals. Th1 cells produce the pro-inflammatory
Th1 cytokine IFN-γ and facilitate CD8 T cell-mediated
cellular immunity against intracellular pathogens [10],
whereas Th2 cells produce cytokines IL-4, IL-5 and
IL-13 that are critical for B cell differentiation and anti-
body-mediated humoral immunity [10,11]. More recently,
IL-17-producing Th17 cells have been discovered and
studied in various contexts [12-14]. In addition to regu-
lating on-going immune responses, E-CD4 T cells are
required for the memory development of CD8 T cell and
B cell immunity in various infection models [15-20].
Unlike E-CD4 T cells, T-CD4 T cells rapidly produce
Th1 and Th2 cytokines upon TCR stimulation in vitro
and in vivo [21], resembling invariant natural killer T
(iNKT) cells in their innate-like functional characteristics
[22-25]. Moreover, T-CD4 T cells maintain Th2 cytokine
production under Th1-skewing conditions [21] but are
poor IL-17 producers under Th17-skewing conditions in
vitro [26]. Therefore, T-CD4 T cells seem to be potent

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
26
effector cells. Unexpectedly, however, mice possessing
T-CD4 T cells are protected from allergen-induced air-
way inflammation [21] and development of experimental
autoimmune encephalomyelitis (EAE) was reduced in
the presence of T-CD4 T cells [27], which indicated that
T- and E-CD4 T cells function differently in physiologi-
cal contexts. However, the role of T-CD4 T cells in im-
mune responses against bacterial infections has not been
addressed.
Hosts protect themselves from infection by mounting
appropriate innate and adaptive immune responses tai-
lored toward pathogens. Intracellular pathogens such as
Listeria monocytogenes are taken up primarily by phago-
cytes and can be destroyed in the phagosomes of antigen
presenting cells (APC) upon infection. However, they can
escape into the cytosol through listeriolysin O (LLO)-
dependent mechanisms [28], and thence are processed
and presented through the MHC class I pathway, induc-
ing robust cellular immunity [29,30]. Listeria-specific
conventional CD4 and CD8 T cells exhibit similar re-
sponding kinetics of activation, expansion and contrac-
tion [31], robustly producing the Th1 cytokine IFN-γ,
which is critical for the anti-microbial activity of macro-
phages [32] and the up-regulation of MHC expression on
APC [33]. In addition, CD8 T cells directly lyse infected
cells as an important mechanism of bacterial clearance
[31,34,35]. Immunological memory protects host organ-
isms by clearing recurrent infections with enhanced ra-
pidity and effectiveness. Studies have reported the essen-
tial role of CD4 T cells in the establishment and devel-
opment of memory immunity against L. monocytogenes
infection [16,17,31]. In the absence of CD4 T cells, al-
though mice are able to mount a primary immune re-
sponse to eliminate the bacteria with a similar efficiency
to CD4-sufficient hosts, they suffer from a defective
memory immune response upon re-infection and suc-
cumb from high bacterial loads [18,19,36-38].
Helicobacter pylori infection causes gastritis due to
IFN-γ production by Helicobacter-specific Th1 cells.
In H. pylori-infected mice, CD4 T cells are necessary
and sufficient for induction of gastritis [39]. In both mice
[40] and humans [41], IFN-γ is elevated in association
with gastritis due to Helicobacter, and antigen-specific
Th1 cells are present in inflamed mucosa [41]. IFN-γ-
deficient mice [42] and T-bet knockout mice [43] fail to
develop gastritis in response to H. pylori. Thus, the evi-
dence strongly supports the hypothesis that in both mice
and humans gastritis due to H. pylori is a Th1- and
IFN-γ-dependent disease. In spite of this evidence, sev-
eral published studies have suggested that gastritis due to
H. pylori gastritis is not absolutely dependent on the Th1
response. We and others showed that while knockout
mice deficient in either IFN-γ or T-bet fail to develop
gastritis in response to H. pylori [42,43], immunodefi-
cient recipients of IFN-γ or T-bet knockout CD4 T cells
do develop gastritis, albeit of less severity than recipients
of CD4 T cells from C57BL/6 mice [39,43]. Thus, in
some situations, H. pylori gastritis can be Th1-indepen-
dent.
In the current study, we show that T-CD4 T cells play
an immunosuppressive role during infection by two dif-
ferent bacterial pathogens, implicating a possible regula-
tory function for these cells during microbial challenge.
2. MATERIALS AND METHODS
2.1. Mice
CIITATg (Tg) and WT littermates were bred and kept
under specificpathogen-free conditions in the animal
facility at the University of Michigan Medical School.
C57BL/6 mice (males and females) at 7 - 8 wk of age
were purchased from Jackson or NCI. CD45.1+ B6 mice
(B6.SJL-Ptprca/BoyAiTac) and CD45.1+ A-/- mice
(B6.SJL-Ptprca/BoyAiTac H2-Ab1tm1Gru) (7 - 8 wk of age)
were purchased from Taconic. Helicobacter-freespecific-
pathogen-free female C57BL/6J-Prkdcscid (severe, com-
bined, immunodeficient, SCID) mice were obtained from
Jackson laboratories. All mice used were 6 - 12 wk of
age. No known mouse pathogens are present in the
mouse colony as determined by routine periodic screen-
ing of sentinel mice. Mice were maintained in static mi-
croisolator cages and offered non-supplemented com-
mercial mouse chow and water ad libitum. All experi-
mental procedures andprotocols were approved by the
University Committee on Use and Care of Animals.
2.2. L. monocytogenes Infection
The recombinant strain of L. monocytogenes express-
ing a secreted form of ovalbumin (rLM-OVA) wasprevi-
ously described [44] rLM-OVAwas grown in brain heart
infusion broth (Difco) to mid-exponential phase followed
by washing with PBS prior to injection i.v. into mice. 5 ×
105 rLM-OVA are equivalent to 0.5 LD50 for infection.
2.3. T- and E-CD4 T Cell Generation
T- and E-CD4 T cells were generated by transferring
BM cells from CIITATg (Tg) or WT to lethally irradiated
A 
-/- (MHC class II-deficient) or WT hosts, respectively.
Eight weeks after the transfer, the hosts were sacrificed
and CD4 T cells were enriched from total splenocytes
using MACS anti-mouse CD4 microbeads (Miltenyi
Biotec).
2.4. Adoptive Transfer Model
For Listeria infection using the adoptive transfer model,
the recipients were sub-lethally irradiated (500 rad). Three

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39 27
days later, each of them received ~3 × 107 CD4 T cell-
depleted splenocytes from WT mice mixed with 107 E-
CD4 or T-CD4 T cells obtained from BM chimeric mice
using anti-mouse CD4 microbeads (Miltenyi Biotec). The
mice were rested overnight and then inoculated with
rLM-OVA as indicated in each experiment via the i.v.
route. For Helicobacter infections, groups of both unin-
fected and H. pylori-infected SCID mice were received
100 μl of CD4 enriched cells from WT or Tg mice via
intraperitoneal injection, for a final dose of ~1 × 106 CD4
T cells per mouse.
2.5. Stimulation of rLM-OVA-Specific T Cell
Population ex Vivo and Cytokine
Intracellular Staining
Splenocytes from infected and naïve mice were stimu-
lated with SIINFEKL peptide (1ug/ml, Biomatik Corpo-
ration) to detect rLM-OVA-specific CD8 T cells or with
LLO (listeriolysin O) 190-201 to detect rLM-OVA-spe-
cific CD4 T cells. Splenocytes were incubated with 1
µg/ml peptide for five hours, and monensin was added
before the last three hours. After five hours, cells were
washed and stained with anti-CD4 and anti-CD8 anti-
body. Next, cells were fixed in 2% paraformaldehyde for
30 min at room temperature, permeabilized with 0.2%
saponin (Sigma), and stained with anti-IFN-g (XMG1.2)
for flow cytometry.
2.6. In Vivo Killing Assay
Splenocytes from naive mice, depleted of red blood
cells, were split into two portions. One was labeled with
a high concentration of CFSE (5.0 nM, 2 × 107 cells/ml)
and pulsed with OVA 257 - 263 peptides as the target
population. The control was labeled with a low concen-
tration of CFSE (0.5 nM, 2 × 107 cells/ml) without pep-
tides. Cells were washed and then the two populations
were mixed at 1:1 ratio (4 - 5 × 106 cells each). Cells
were injected into rLM-OVA infected or PBS-treated
mice. Mice were euthanized at indicated time points, and
single-cell suspensions of spleenswere analyzed by flow
cytometry.The killing efficiency was calculated as fol-
lows: 100 – ([(% peptide pulsed in infected/% unpulsed
in infected)/(%peptide pulsed in uninfected/% unpulsed
in uninfected)] × 100).
2.7. Quantification of Listeria Load
Livers were removed and put into 14-ml tubes con-
taining 10 ml PBS with 0.2% NP40. The tissues were
homogenized by using a homogenizer (The Lab Depot,
Inc.) at maximum speed for 30 seconds. Tissue homoge-
nates were subjected to 10-fold serial dilutions and then
plated onto Luria broth agar plates. The number of colo-
nies formed were counted after 24 hour of incubation at
37˚C.
2.8. H. pyloriinfection
Overnight broth cultures of H. pylori strain SS1 in 10
ml of Brucella broth with 10% fetal calf serum were cen-
trifuged, washed, resuspended in sterile phosphate-buff-
ered saline (PBS), counted on a hemacytometer, and di-
luted to a final concentration of 1 × 108 bacteria/ml.
SCID mice were given 100 μl of sterile 0.5 M Na2CO3
via gastric feeding tube followed by100 ml of the bacte-
rial suspension giving a total dose of 1 × 107 of H. pylori
SS1 per mouse. This procedure was repeated on the fol-
lowing day, for a total of two inoculations.
2.9. Cytokine Analyses of Helicobacter
Infected Samples
T cell depleted splenocytes from uninfected C57Bl/6J
mice were irradiated with 3000 rads and used as APC.
Splenic CD4 T cells from adoptively transferred mice
were enriched, pooled, and then split into two sets. One
set was co-cultured with APCs (1:2 ratio) that had been
loaded overnight with H. pylori bacterial lysate (50 mg/
ml) for 72 hrs. The other set of cells were stimulated with
5 mg/ml plate-bound anti-CD3e (145-2C11), 1 mg/ml
anti-CD28 (37.51), and 50 U of IL-2 (Roche, Indianapo-
lis, IN) at a concentration of 1 × 106 CD4 T cells/ml for
72 hrs. The supernatants from both stimulation condi-
tions were collected, and cytokine production was meas-
ured by ELISA.
2.10. Histology
1 mm wide strips from the greater curvature were
emersion-fixed in 10% neutral buffered formalin, em-
bedded in paraffin, cut in 5 µm sections, and stained with
hematoxylin and eosin (HE). Extent of gastritis was
scored as previously described. Briefly, adjacent 200×
microscopic fields were examined for the presence of
gastric infiltrate severe enough to displace glands, pres-
ence of neutrophilic inflammation, and/or presence of
gastric epithelial metaplasia. Two longitudinal sections of
gastric fundus were scored in their entirety, and the per-
centage of positive fields in all three categories was
added together to calculate the total score. All sections
were scored blind, without prior knowledge of their
source.
2.11. Flow Cytometry
Antibodies specific for CD4 (GK1.5), CD8 (53-6.7),
CD45.1 (A20), CD45.2 (104), TCR
 (H57-597), NK1.1
(PK136), Ly-6G (Gr-1), CD11c (HL3), CD11b (M1/70),
F4/80 (6F2), I-Ab (AF6-120.1), H-2Kb (AF6-88.5) were
from PharMingen, BD Bioscience (Mountain View, CA).

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
28
The antibody against Foxp3 (FJK-16s; eBioscience) was
used according to the staining protocol provided by the
company. Samples were analyzed using a FACS Canto
flow cytometer (Becton Dickinson). Data were analyzed
using FlowJo software (Tree Star).
2.12. RNA Analyses
Total RNA of the splenocytes from infected with Lis-
teria or Helicobacter and control mice were extracted
using TRIzol (Invitrogen). The PCR reactions were per-
formed and analyzed using the iCycleriQTM (BioRad).
Conditions: 95˚C for 5min, followed by 40 cycles of
95˚C for 30 sec, 55˚C for 30 sec, and 72˚C for 40 sec.
Primers: GAPDH Forward (F): ctccactcacggcaaattca, Re-
verse (R): cgctcctggaagatggtgat. IL-1 F: caaccaacaagtgatat-
tctccatg, R: gatccacactctccagctgca. IL-10 F: ggttgccaagcct-
tatcgga, R: acctgctccactgccttgct. IFN-g F: tcaagtggcata-
gatgtggaagaa, R: tggctctgcaggattttcatg. TNF-α F: ccccaa-
agggatgagaagtt, R: cacttggtggtttgctacga. To measure cy-
tokine expression from Helicobacter infected samples,
mice were euthanized and stomachs were harvested eight
weeks after adoptive transfer. mRNA was isolated as
above, and cDNA was synthesized and analyzed on cus-
tom SuperArrays (CAPM-0752A) using SA Biotech’s
SYBR green master mix. The arrays contained probes for
IL-1β, IL-6, TNFα, and GAPDH. IL-4 mRNA expression
levels were determined separately by performing qPCR
with IL-4 specific primers of our design (F: aacgtcctca-
cagcaacgaa, R: tgcagctccatgagaacact).
2.13. Statistical Analysis
The statistical analysis was done using Prism software.
A two-tailed t-test was used for statistical analysis. p
values of ≤0.05 were considered significant, and p values
> 0.05 were not indicated and were considered statisti-
cally insignificant.
3. RESULTS
3.1. CD8 T Cell-Mediated Immune Response
Was Decreased in Mice with T-CD4 T
Cells
To ascertain the role of T-CD4 T cells during bacterial
infection, we compared the immune responses against L.
monocytogenes infection between two groups of mice:
CIITATg (Tg) and wild type (WT) mice. Tg mice express
MHC class II on both thymocytes and TEC due to the
expression of the CIITA transgene directed by the CD4
promoter [9]. Therefore, thymocytes can be selected by
thymocytes and TEC generating a mixture of T- and
E-CD4 T cells. An indirect measurement suggests that
approximately 10% - 20% of peripheral CD4 T cells are
T-CD4 T cells [45]. By contrast, WT mice possess E- but
not T-CD4 T cells because they express MHC class II
only on TEC in the thymus. Therefore, the differences in
the immune responses between Tg and WT mice are
likely attributable to the presence of T-CD4 T cells in
addition to E-CD4 T cells in Tg mice.
To facilitate the detection of Listeria-specific Tcell
populations, we used a recombinant strainof L. monocy-
togenesexpressing a secreted form of chicken ovalbumin
(rLM-OVA). The anti-OVA response was used as an in-
dicator of anti-Listerial responses of CD8 T cells [44].
Tg and WT mice were inoculated intravenously with
rLM-OVA or PBS as a control. Tissues were collected
and analyze on day 7 after infection. Infected WT and Tg
mice had bacterial counts that were below the limit of
detection at the time of analysis suggesting efficient
clearance of bacteria in both groups of mice. As shown in
Figure 1(a), total as well as CD4 and CD8 T cell num-
bers were comparable between Tg and WT groups. To
measure the Listeria-specific T cell response, CD4 and
CD8 splenocytes were stimulated separately ex vivo with
rLM-OVA-specific peptides and T cells recognizing the
cognate peptides were detected by intracellular IFN-γ
staining [18]. Significantly fewer CD8 T cells from Tg
mice expressed IFN-γ than in WT mice upon peptide
stimulation (Figure 1(b)), whereas IFN-γ+ CD4 T cells in
response to rLM-OVA were slightly increased in Tg
compared to WT mice (Figure 1(c)). Therefore, the
presence of T-CD4 T cells partially suppressed immune
responses of CD8 T cells during bacterial infection.
Induction of cytotoxicity is another important indica-
tor of CD8 T cell function, in addition to IFN-γ expres-
sion [33-35]. Therefore, an in vivo killing assay was em-
ployed to measure rLM-OVA-specific CD8 T cell killing
efficiency on day 7 [17,46]. As expected, in naïve mice,
the ratio of CFSEhi to CFSElo populations remained 1:1
(Figure 1(d)). However, the CFSEhi population was de-
creased in both groups of infected mice indicating that it
was recognized and eliminated by rLM-OVA-specific
CD8 T cells (Figure 1(d)), and the difference in the kill-
ing efficiency between the two infected groups was not
significant. Therefore, the lower number of IFN-γ CD8
T cells stimulated by the infection, compared to the wild
type controls, did not compromise killing activity in Tg
mice with T-CD4 T cells.
3.2. Memory Immune Responses against
L. monocytogenes Are Impaired in the
Presence of T-CD4 T Cells
It has been demonstrated that E-CD4 T cells play a
critical role in the establishment of optimal immune
memory of CD8 T cells against L. monocytogenes [18,19,
36,37]. Therefore, we asked whether T-CD4 T cells per-
form the same function as E-CD4 T cells do in the course
of CD8 T cell memory generation. Tg and WT mice were

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39 29
Figure 1. Reduced anti-listerial responses in Tg mice during
primary infection. Tg and WT littermates were infected intra-
venously with rLM-OVA (5 × 104) or PBS. Mice were eutha-
nized and analyzed 7 days after infection. (a) Numbers of the
total splenocytes and of the indicated cell populations are shown;
(b) and (c) Frequencies of IFN-γ-producing rLM-OVA-specific
CD8 (b) and CD4 (c) T cells. The values in representative FACS
profiles are percentages of the total splenocyte population; the
graphs on right show the percentage of antigen specific CD8
(IFN-γ + CD8/total CD8) (b) and CD4 (IFN-γ + CD4/total CD4)
(c) T cells. The bars indicate the median value; (d) In vivo kill-
ing assay. A mixture of OVA peptide-loaded target cells (CFSEhi)
and control cells (CFSElo) were injected into recipient mice.
Mice were euthanized 3 hours later and the composition of the
injected cells in the spleenswere analyzed by flow cytometry.
The numbers above histograms indicate the percentages of
CFSEhi and CFSElo.
infected with a low dose of rLM-OVA, rested for a
month, and challenged with a high dose of rLM-OVA.
Three days after the challenge, the host memory response
was analyzed using the same parameters as above. The
numbers of total splenocytes and CD4 T cells were
equivalent in both mouse groups and CD8 T cells were
slightly reduced in Tg mice although there was no statis-
tical significance (Figure 2(a)). When antigen specific
responses of CD8 T cells were examined, rLM-OVA-
specific CD8 T cells in Tg mice were decreased 2-fold
compared to WT mice (Figure 2(b)). Therefore, Tg mice
consistently exhibited decreased CD8 T cell response
against rLM-OVA during primary infection and secon-
dary challenge. However, unlike CD4 T cell responses
during the 7 days of infection, the CD4 T cell compart-
ment during challenge showed a similar response to
rLM-OVA between Tg and WT mice rLM-OVA between
Tg and WT mice (Figure 2(c)). To compare the effi-
ciency of bacterial clearance, L. monocytogenes burdenin
Figure 2. Poor memory response against listeria in Tg mice. Tg
and WT littermates were inoculated intravenously with 4 × 103
rLM-OVA and rested for one month before challenge with 5 ×
105 rLM-OVA or PBS as a control. All the mice were eutha-
nized and analyzed 3 days after the secondary infection. (a)
Numbers represent the total splenocytes and the indicated cell
populations; (b) and (c) Frequencies of IFN-γ producing rLM-
OVA-specific CD8 (b) and CD4 (c) T cells. Experiments and
data analyses were done as described in Figure 1; (d) Numbers
of viable bacteria from liver homogenates are depicted. The
bars indicate median values. The colony forming units from
mice treated with PBS were below the detection limit of 100
CFU/mouse liver.

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
30
the liver was measured using the same group of mice. We
observed that the bacteria loads in Tg mice were almost
ten fold of that in WT mice (Figure 2(d)). These data
demonstrate that Tg mice with both E- and T-CD4 T cells
exhibit impaired immunity to L. monocytogenes infec-
tion.
with CD4 T cell-depleted splenocytes from naïve WT
mice for adoptive transfer (Figures 3(a) and (b)). C57BL/6
recipients were sub-lethally irradiated three days before
adoptive transfer to facilitate the reconstitution of the
incoming cells. The irradiation abolishes the immune
responsiveness of the remaining host T cells and thus the
T cell immune responses are primarily attributable to the
transferred cells [48,49]. To distinguish transferred cells
from recipient cells, the congenic markers CD45.1 and
CD45.2 were used.
3.3. T-CD4 T Cells Were Responsible for the
Reduced Anti-Listerial Response
The reduced efficiency of the anti-Listerial response in
Tg mice suggested that T-CD4 T cells might negatively
regulate immune function. However, it was not yet clear
whether the difference was directly due to T-CD4 T cell
function or it was secondary to other unknown differ-
ences between Tg and WT mice. Recently, we reported
that IL-4 produced by T-CD4 T cells induces the genera-
tion of CD8 T cells with the effector/memory phenotype
by expressing IFN-γ and Eomes [47]. Although the func-
tion of these innate CD8 T cells during Listeria infection
is not known, the difference in immune responses could
be due to innate CD8 T cells not directly by T-CD4 T
cells. To determine whether CD8 T cells themselves in-
fluence the outcome of infections, we established an
adoptive transfer mouse model in which different groups
only differed in CD4 T cell populations. To obtain T- and
E-CD4 T cells exclusively, we constructed bone marrow
(BM) chimeric mice (Materials and Methods). In A-/-
micethat received Tg BM cells, all CD4 T cells become
T-CD4 T cells because they have to be selected by donor
MHC class II expressing thymocytes. E-CD4 T cells
were generated by transferring WT BM to WT mice.
Total T- or E-CD4 T cells were isolated from spleens
eight weeks after BM transplantation and were mixed
The recipients were analyzed 7 days after infection
(Figure 3(a)). The number of CD4 T cells in T-CD4 T
cell recipients was lower than that in E-CD4 T cell re-
cipients, although transferred CD8 T cells were similar in
number between the two groups (Figure 3(c)). To ad-
dress whether the decreased T-CD4 T cell number was
caused by infection, we compared the repopulation of
adoptively transferred T- and E-CD4 T cells in naïve
hosts (Figure 3(d)). In our infection experiments shown
in Figure 3(a), the infection was done one day after cell
transfer. Therefore, we examined mice at days 1 and 8
after transfer that correspond to the time points of infec-
tion and to 7 days after infection at which time we ana-
lyzed the immune responses. In addition, we examined cells
at day 3 (equivalent to day 2 of infection) when innate
immunity peaks to assess the cell numbers.The numbers
of transferred T-CD4 T cells declined initially but re-
mained fairly constant afterward. At day 8, the difference
in cell numbers of T-CD4 T cells compared to that of
E-CD4 T cells was similar to what was observed on day
7 in infected hosts. Therefore, the decline of the T-CD4 T
cell population in recipients was unlikely due to the in-
fection or immune responses.
OPEN A CCESS

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39 31
Figure 3. Effector CD8 T cell generation is compromised in the presence of T-CD4 cells. (a) A
scheme of the experimental protocol. The infection dose was 2 × 104 rLM-OVA; (b) The com-
position of the CD45.1+ input cell population prior to the transfer; (c) Numbers of the indicated
cells originating from the donor on day 7; (d) T- and E-CD4 ratio after co-transfer into naïve
mice. Cell populations were tracked using congenic markers CD45.1 and CD45.2 on the indi-
cated days and the T/E-CD4 ratios were normalized to the input ratio on day 0. N = 3. (e) and (f)
Frequencies of IFN-γ-producing rLM-OVA-specific CD8 (e) and CD4 (f) T cells. The FACS
data shown were gated on donor populations. The values in representative FACS profiles are
percentages of total donor splenocytes; the graphs on right represent percentages of total donor
CD8 (e) or CD4 (f) T cells that produced IFN-γ. The bars indicate the median value; (g) In vivo
killing assay. Experiments were performed as described in Figure 1(d), except that mice were
sacrificed 24 hours after cell transfer.

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
32
When antigen specific responses were examined, rLM-
OVA-specific CD8 T cells were significantly reduced in
T-CD4 T cell recipients (Figure 3(e)). In contrast, the
percentage of rLM-OVA-specific CD4 T cells was com-
parable between the two groups despite the low T-CD4
Tcell number (Figure 3(f)). We next tested the killing
efficiency as in Figure 1(d) except that we sacrificed
mice after overnight instead of 3 hours after cell transfer,
because both rLM-OVA-specific CD4 and CD8 T cell
numbers were much lower in the adoptive transfer model.
Consistent with the reduction of IFN-γ expressing CD8 T
cells in T-CD4 T cell recipients, T-CD4 recipients showed
significantly decreased cytotoxicity (Figure 3(g)). There-
fore, unlike E-CD4 T cells, T-CD4 T cells do not support
efficient development of rLM-OVA-specific CD8 T cells
upon infection, and the decreased anti-Listerial immune
responses can be directly attributed to the T-CD4 T cells
themselves
3.4. T-CD4 T Cells Do Not Support
Anti-Listerial Memory Immunity
After observing the effect of T-CD4 T cells on CD8 T
cells during primary immune responses, we then tested
whether the mice with transferred T-CD4 T cells were
also defective in memory immunity. The adoptive trans-
fer was performed as described in Figure 3(a) with
modifications (Figure 4(a)), and a group of recipients
receiving donor cells that did not contain CD4 T cells
was included as an additional control. Three groups of
mice receiving different populations of cells (Figure 4(b))
were infected and rested for a month before the challenge.
These mice were analyzed three days after challenge
with a high dose ofrLM-OVA. The number of total spleno-
cytes in T-CD4 T cell recipients was slightly lower than
in E-CD4 recipients although the difference was not sig-
nificant (Figure 4(c)). Although total CD8 T cell num-
bers were comparable between T- and E-CD4 recipients,
rLM-OVA-specific CD8 populations decreased in per-
centages and thus in cell numbers in T-CD4 recipients
(Figure 4(d)), consistent with the observation made in
Tg and WT hosts. In fact, CD8 T cell responses in T-CD4
recipients were similar to the mice that did not receive
CD4 T cells suggesting lack of help by T-CD4 T cells.
The transferred CD4 T cells, by comparison, decreased
in both total and rLM-OVA-specific populations in T-CD4
recipients (Figures 4(b) and (e)). In agreement with poor
CD8 T cell responses, bacterial burdens were increased
in T-CD4 recipients comparable to those of the mice that
did not receive CD4 T cells although the differences did
not reach statistical significance (Figure 4(f)). Therefore,
the results suggest that, during long-term immune re-
sponses, antigen specific T-CD4 T cells do not provide a
sufficient help to generate a robust memory CD8 T cell
response.
Figure 4. T-CD4 cells inhibit development of memory CD8
effector cells. (a) A scheme of the experimental protocol. Mice
were infected with 2 × 103 and 5 × 105 rLM-OVA for primary
and secondary infection, respectively, and then sacrificed three
days after the second infection; (b) The composition of the
CD45.1+ input cell populations prior to the transfer; (c) The
numbers represent total splenocytes and the indicated cells
originating from the donor; (d) and (e) Frequencies of IFN-γ-
producing rLM-OVA-specific CD8 (d) and CD4 (e) T cells (N =
5). Experimental design and data analysis were performed as
described in Figures 3(e) and (f); (f) Numbers represent viable
bacteria isolated from liver homogenates. The bars indicate
median values. The CFU counts of PBS-treated mice were be-
low the detection limit.
3.5. The Presence of T-CD4 T Cells Neither
Alters Innate Immunity nor the Treg
Population
Having observed the suppressive effect exerted by
T-CD4 T cells during bacterial infection, we asked whe-
ther T-CD4 T cells would change innate immunity be-
cause of their innate-like phenotype. However, Tg and
WT groups showed increased splenic Gr1+ F4/80+ popu-
OPEN A CCESS

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39 33
lations to a comparable level after infection (Figure 5(a))
and mRNA expression of IFN-γ, TNF-α and IL-1β from
the spleens were comparable between the two groups
(Figure 5(b)). Therefore, innate immune responses were
not directly regulated by T-CD4 T cells. Next, we tested
whether immune suppression was due to a change in
Treg populations by the presence of T-CD4 T cells. We
previously showed that natural Treg (nTreg) develop-
ment can be supported by MHC class II+ thymocytes and
they mediated comparable suppression [9]. To test whe-
ther the generation of nTreg would be different in BM
transplanted mice, we compared Foxp3+ cell populations
in BM chimeras. As shown in Figure 5(c), the percent-
ages of Foxp3+ cells of T-CD4 T cells were similar to
Figure 5. T-CD4 T cells does not change Treg populations ((a)
and (b)) Mice were infected intravenously with 100,000 rLM-
OVA and sacrificed on day 3 to assess cell populations (a) and
to measure cytokine expression (b) from the spleen of Tg and
WT mice. The RNA expression of the indicated cytokine genes
were measured by real time PCR after reverse transcription.
Relative expression of each cytokine was normalized to GAPDH.
The data are representative of 4 mice in each group. (c) Repre-
sentative profiles of Foxp3 expression in E- and T-CD4 T cells
from [TgA

-/-] and [WTWT] chimeric mice. (d) and (e)
WT and Tg mice were infected as in Figure 1. Seven days after
infection, freshly isolated CD4 T cells were stained for Foxp3
expression (d) or stimulated in the presence of rLM-OVA pep-
tides for 5 hours followed by staining of Foxp3 and IFN-γ (e).
that of E-CD4 T cells. We then asked if T-CD4 T cells
could induce the generation of Foxp3+ E-CD4 T cells
upon infection, which would in turn suppress effector
E-CD4 T cells. To test this, we infected WT and Tg mice
with rLM-OVA and examined Foxp3+ cells together with
CD8 effector T cell generation by IFN-γ staining after
primary infection as in Figure 1. Freshly isolated spleno-
cytes from infected mice showed comparable percentages
of Foxp3+ cells (Figure 5(d)). When CD4 T cells from
infected mice were stimulated with rLM-OVA antigens
in vitro, IFN-γ expressing CD4 T cells were lower in the
culture of CD4 T cells from Tg mice but Foxp3+ cells
were at a similar level between the two groups (Figure
5(e)). Therefore, suppression mediated by T-CD4 T cells
is unlikely to be due to alteration of Treg populations.
3.6. Lack of Antigen-Specific CD4 T Cell
Responses upon Helicobacter Infection
in the Presence of T-CD4 T Cells
To determine if the suppressive activity of T-CD4 T
cells is specific on CD8 T cells, we examined immune
responses mediated by CD4 T cells using an established
model of infection by Helicobacter pylori. We have
shown that adoptive transfer of CD4 T cells to H. py-
lori-infected immunodeficient (SCID) mice is necessary
and sufficient to induce severe gastritis [40]. This model
allows us to measure the effect of T-CD4 T cells on
E-CD4 T cells. To evaluate the role of T-CD4 cells in
infection by H. pylori, we transferred either E-CD4 or a
mixture of E- and T-CD4 T cells from WT and Tg mice,
respectively, to SCID mice that were either uninfected or
infected with H. pylori (Figure 6(A)). Eight weeks after
transfer, we examined CD4 T cells in the spleen. The
number of CD4 T cells from infected mice was similar in
mice given Tg CD4 T cells and those given WT CD4 T
cells, and CD4 T cells were expanded to the similar de-
gree in the two recipient groups (Figure 6(B)). CD4 T
cells were then stimulated by plate-bound anti-CD3 to-
gether with soluble anti-CD28 or with antigen loaded
APC that were pulsed with H. pylori antigen. WT and Tg
CD4 T cells produced equivalent amounts of IFN-γ upon
anti-CD3 stimulation (Figure 6(C), top left panel). In
contrast, cytokine responses to H. pylori antigen differed
depending on donor cells. As expected, CD4 T cells re-
covered from H. pylori-infected WT CD4 recipients
produced IFN-γ in response to H. pylori antigen-pulsed
antigen presenting cells as expected, whereas Tg CD4
cells recovered from infected recipient mice failed to
produce IFN-γ in response to H. pylori antigen stimula-
tion (Figure 6(C), top right panel). Thus, the presence of
T-CD4 in these mice not only failed to induce cytokine
expression by the T-CD4 T cells themselves, but also
suppressed H. pylori- specific IFN-γ responses of E-CD4
T cells. Also as expected, CD4 cells from Tg recipients

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
34
Figure 6. The presence of T-CD4 T cells inhibits the generation of Helicobacter-specific effector CD4 T
cells. (A) A scheme of the experimental protocol; (B) Recovery of T cells from Helicobacter infected mice.
Flow cytometric analysis of cells recovered from the spleens of SCID recipients of adoptive transfers of
CD4 T cells; (C) CD4 T cells were isolated from mice and then stimulated with anti-CD3/CD28 or with H.
pylori lysate. Supernatants were used forELISA. Bars are the average of 3 replicates of each condition, and
each replicate used CD4 T cells pooled from 5 mice; (D) Hematoxylin and eosin-stained sections of gastric
mucosa from H. pylori-infected SCID recipient mice. (a) WT CD4 recipient. Inflammatory infiltrate (ar-
rowheads) consists of neutrophils, lymphocytes, and macrophages. Bracket indicates metaplastic glands. (b)
Tg CD4 recipient. Arrow indicates a gland abscess. Bars = 50 m; (E) Relative mRNA levels of cytokine
genes from the gastric mucosa of uninfected and infected SCID mice that were adoptively transferred with
either WT or Tg CD4 cells. Expression of cytokine genes normalized to GAPDH expression, and multi-
plied by 10,000 for ease of visualization.
but not from WT recipients produced IL-4 in response to
anti-CD3 stimulation (Figure 6(C), bottom left panel)
compatible with previous studies demonstrating the abil-
ity of T-CD4 cells to produce Th2 as well as Th1 cyto-
kines after pan-TCR stimulation [21]. But IL-4 levels
upon H. pylori antigen stimulation showed little differ-
ence between uninfected and infected WT and Tg mice
indicating the lack of IL-4 producing CD4 T cells upon
infection (Figure 6(C), bottom right panel).
Interestingly, although H. pylori-specific CD4 T cell
IFN-γ production was different between recipients of Tg
and WT CD4 cells, gastritis was extensive in both groups
without statistically significant differences (Figure 6(D)).
In addition, expression of proinflammatory cytokine
mRNA in the gastric mucosa was elevated in both re-
cipient groups in response to infection by H. pylori. The
gastric mRNA levels of IL-1β, TNF-α, and IL-6 were
significantly elevated in infected mice compared to un-
infected mice regardless of the CD4 T cell donor (Figure
6(E)). Expression of anti-inflammatory cytokines TGF-β
OPEN A CCESS

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39 35
and IL-10 did not differ between groups regardless of
infection or donor cell type. These data suggest that T-CD4
suppress E-CD4 antigen-specific IFN-γ responses, and
indicate that in this model, gastritis and proinflammatory
cytokine production seems to be independent of IFN-γ
production by antigen-specific CD4 T cells.
4. DISCUSSION
In the present study, we provide evidence for a novel
suppressive function of T-CD4 T cells during adaptive
immune responses against two bacterial pathogens, Lis-
teria monocytogenes and Helicobacter pylori. The sup-
pressive effect mediated by T-CD4 T cells appears to be
broad because the presence of T-CD4 T cells inhibited
the generation of both CD4 and CD8 effector T cells in
response to bacterial infections. These findings are con-
sistent with our previous work that airway inflammation
was diminished in mice that have both E- and T-CD4 T
cells [21] and EAE development was suppressed by
having T-CD4 T cells together with E-CD4 T cells [50].
Taken together, our data support a suppressive role for
T-CD4 T cells in regulating immunity in many different
immune contexts.
Suppression by T-CD4 T cells is not likely due to an
alteration in Treg populations, because T-CD4 T cells
used in the assay did not contain more Treg than E-CD4
T cells based on Foxp3 staining. In addition, Treg from
T-CD4 T cell recipients are functionally equivalent to
those from E-CD4 T cell recipients when their suppres-
sion was tested in vitro [9]. Despite having the suppres-
sive function, T-CD4 T cells are distinctive from Treg
since T-CD4 T cells do not express Foxp3 (Figure 5).
Treg cells neither have pre-made IL-4 mRNAs nor do
they release Th1 and Th2 cytokines immediately after
TCR stimulation as T-CD4 T cells do. Treg cells, in con-
trast, produce large amounts of suppressive cytokines
TGF-  and IL-10 after differentiation, cytokines that are
not produced in significant quantities by either T-CD4 or
E-CD4 T cells (Chang and Chang, unpublished data). It
is yet unclear how T-CD4 T cells exert a negative regu-
latory effect on other T cells. As reported, IL-4 produced
by either T-CD4 or iNKT cells induce the generation of
innate effector CD8 T cells [49,51,52]. Therefore, it is
possible that these innate CD8 T cells exert a negative
effect during Listeria infection. However, the suppressive
effect on Helicobacter-specific CD4 T cells does not
fully support this mechanism. Perhaps, similar to Treg,
T-CD4 T cells contact the target cells directly and cause
suppression, or T-CD4 T cells may act indirectly by in-
fluencing the local environment, e.g., augmenting the
activity of Treg in vivo, which may lead to enhanced
immune suppression. Further investigations into mo-
lecular mechanisms for T-CD4 T cell-mediated immune
suppression is warranted.
T-CD4 T cells differ from E-CD4 T cells in several
ways. T-CD4 T cells appear to be born as effector cells,
expressing pre-formed mRNAs of effector T cell cyto-
kines prior to activation [21]. This allows them to pro-
duce Th1 and Th2 cytokines shortly after TCR stimula-
tion [21]. However, as we have shown in the current
study, T-CD4 T cells suppress CD8 and CD4 T cell func-
tions instead of providing help as E-CD4 T cells do.
E-CD4 T cells are required for mounting an effective
secondary immune response against bacterial infection.
In the presence of E-CD4 T cells, antigen-specific CD8 T
cells expand during the primary immune response against
Listeria and a fraction of them become memory cells. As
a result, the hosts are able to clear Listeria efficiently
upon a secondary challenge. In contrast, T-CD4 T cells
adversely affected primary and memory immunity against
Listeria. They appear to be suppressive, as evidenced by
the low number of effector CD8 T cells in the presence
of T-CD4 T cells. In fact, the CD8 response in the pres-
ence of T-CD4 T cells was similar to mice that did not
receive CD4 T cells further supporting that T-CD4 T
cells do not function as helper cells although they do pro-
duce effector cytokines.
T-CD4 T cells resulted in suppression of memory as
well as primary immune responses against Listeria infec-
tion. Perhaps memory T-CD4 T cells are generated and
they actively suppressed the response of CD8 memory
cells. It is equally possible that the memory CD8 T cells
were generated in the presence of T-CD4 T cells but they
are functionally defective. The latter seems to be the case
because the host already demonstrated reduced rLM-
OVA-specific CD8 T cell population during the primary
response. Testing two possibilities requires a model of
secondary adoptive transfer to separate the function of
T-CD4 T cells in primary response from that in memory
immune response. Nevertheless, the current data clearly
showed that, under physiological conditions, the pres-
ence of T-CD4 T cells resulted in deficient CD8 T cell
responses at both primary and memory phases.
Like in infection by L. monocytogenes, T-CD4 cells
suppressed antigen-specific responses of E-CD4 T cells
in mice infected with H. pylori. Thus, the presence of
T-CD4 cells appeared to interfere with H. pylori-specific
IFN-γ production by E-CD4 cells. Surprisingly, however,
in spite of the absence of an antigen-specific Th1 re-
sponse, the extent of gastritis in mice with E- and T-CD4
cells were not different from recipients of E-CD4 cells
(Figure 6). Although there is strong published evidence
that gastritis due to H. pylori is associated with IFN-γ
producing Th1 T cells, gastritis in mice can occur in the
absence of IFN-γ producing CD4 cells [39,43] and others
have also suggested that gastritis due to H. pylori is not
absolutely dependent on a Th1 response [53,54]. In the
current study, the presence of T-CD4 cells did affect host

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
36
H. pylori-specific T cell response, but did not appear to
alter the severity of gastric inflammation. This finding
could provide a clue to an important difference between
murine and human disease due to H. pylori. In mice, the
extent of gastritis is inversely proportional to bacterial
colonization density, and in some models in which in-
flammatory response is severe, bacterial colonization is
eventually eliminated, and the gastric mucosa returns to
normal morphology [55,56]. In humans, in contrast, al-
though there have been scattered reports of spontaneous
clearance of infection [57], most people remain infected
for life, regardless of disease severity and there is no
association between the severity of gastritis and the level
of bacterial colonization [58]. It is possible that the fail-
ure of humans with gastritis to clear infection is attribut-
able to suppressive T-CD4 cells. These cells may suffi-
ciently downregulate the immune response to result in
failure of eradication but not sufficiently to affect the
level of gastric inflammation and disease.
We have observed that T-CD4 T cells had lower cell
recovery than E-CD4 T cells after Listeria infection when
equal numbers were adoptively transferred. This was also
observed in naïve recipients, and the ratio between
co-transferred T- and E-CD4 T cells became stable after
the initial reduction (Figure 3). Therefore, the decrease
in total T-CD4 T cell number seems to be due to the re-
duced reconstitution efficiency rather than loss of T-CD4
T cells by the infection. Although underlying mecha-
nisms for a deficit in T-CD4 T cell reconstitution are not
yet understood, our data suggest that the reduced anti-
gen-specific responses against both bacteria were not
likely the consequence of a low T-CD4 T cell number in
adoptively transferred mice. During a primary immune
response when CD8 T cells responded equally well with
or without E-CD4 T cells [18,19,36,37], the presence of
T-CD4 T cells resulted in decreased host response. In ad-
dition, Tg mice that possess sufficient numbers of E-
CD4 T cells still showed compromised CD8 T cell re-
sponse during infection. Finally, E-CD4 T cells alone
were able to mount Th1 responses against Helicobacter
infection but this response was lost when T-CD4 T cells
were present. Together, the data support that poor im-
mune responses during bacterial infection is a result of
suppression mediated by T-CD4 T cells.
The role of T-CD4 T cells in humans is unknown.
Human T-CD4 T cells appear during gestation and de-
crease after birth [6], which coincides with MHC class II
expression in thymocytes [59]. Moreover, fetal stem cells
possess greater potential to differentiate to Treg than
adult stem cells [60]. Therefore, a large number of CD4
T cells in infants seem to have the suppressor function.
Interestingly, infants are known to be highly susceptible
to infection [61], which we propose may be due in part to
the presence of these suppressor cells at a high level.
Currently, little is known the role of T-CD4 T cells in
infants during the course of an immune response. But it
is tempting to speculate that having a T cell compartment
comprised of cells with negative function dampens adap-
tive immunity in infants upon infection or vaccination.
Patients who have received BM transplantations can also
develop T-CD4 T cells [62-65]. Studies with DiGeorge
patients strongly support the development of T-CD4 T
cells with suppressive function in the patients [65]. Fur-
ther investigations to understand the potential role of
T-CD4 T cells during the human immune response will
be important to improve our treatment of infectious and
autoimmune diseases.
5. ACKNOWLEDGEMENTS
We are thankful for the help of Dr. Jihoon Chang and Tim Voorhees
with the comparison of cytokine levels between T- and E-CD4 culture
supernatant and Drs. Phil King and Derek Sant’Angelo for critical
reading of the manuscript. We are also grateful for Dr. Hao Shen who
generously provided us with the recombinant Listeria strain used in our
study. This study was supported by the NIH grant AI064540 to M.
O’Riordan, AI043643 to K. Eaton, and AI073677 to C.-H. Chang.
REFERENCES
[1] Germain, R.N. (2002) T-cell development and the CD4-
CD8 lineage decision. Nature Reviews Immunology, 2,
309-322. doi:10.1038/nri798
[2] Bonduel, M., Pozo, A., Zelazko, M., Raslawski, E., Del-
fino, S., Rossi, J., Figueroa, C. and Sackmann, M.F. (1999)
Successful related umbilical cord blood transplantation
for graft failure following T cell-depleted non-identical
bone marrow transplantation in a child with major histo-
compatibility complex class II deficiency. Bone Marrow
Transplant, 24, 437-440.
doi:10.1038/sj.bmt.1701915
[3] Godthelp, B.C., van Eggermond, M.C., Peijnenburg, A.,
Tezcan, I., van Lierde, S., van Tol, M.J., Vossen, J.M. and
van den Elsen, P.J. (1999) Incomplete T-cell immune re-
constitution in two major histocompatibility complex class
II-deficiency/bare lymphocyte syndrome patients after
HLA-identical sibling bone marrow transplantation. Blood,
94, 348-358.
[4] Klein, C., Cavazzana-Calvo, M., Le Deist, F., Jabado, N.,
Benkerrou, M., Blanche, S., Lisowska-Grospierre, B.,
Griscelli, C. and Fischer, A. (1995) Bone marrow trans-
plantation in major histocompatibility complex class II
deficiency: A single-center study of 19 patients. Blood, 85,
580-587.
[5] De Smedt, M., Hoebeke, I. and Plum, J. (2004) Human
bone marrow CD34+ progenitor cells mature to T cells on
OP9-DL1 stromal cell line without thymus microenvi-
ronment. Blood Cells, Molecules, and Diseases, 33, 227-
232. doi:10.1016/j.bcmd.2004.08.007
[6] Lee, Y.J., Jeon, Y.K., Kang, B.H., Chung, D.H., Park,
C.G., Shin, H.Y., Jung, K.C. and Park, S.H. (2010) Gen-

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39 37
eration of PLZF+ CD4+ T cells via MHC class II-de-
pendent thymocyte-thymocyte interaction is a physio-
logical process in humans. The Journal of Experimental
Medicine, 207, 237-246
[7] Van Coppernolle, S., Verstichel, G., Timmermans, F., Vel-
ghe, I., Vermijlen, D., De Smedt, M., Leclercq, G., Plum,
J., Taghon, T., Vandekerckhove, B. and Kerre, T. (2009)
Functionally mature CD4 and CD8 TCRalphabeta cells
are generated in OP9-DL1 cultures from human CD34+
hematopoietic cells. Journal of Immunology, 183, 4859-
4870. doi:10.4049/jimmunol.0900714
[8] Choi, E.Y., Jung, K.C., Park, H.J., Chung, D.H., Song,
J.S., Yang, S.D., Simpson, E. and Park, S.H. (2005) Thy-
mocyte-thymocyte interaction for efficient positive selec-
tion and maturation of CD4 T cells. Immunity, 23, 387-
396. doi:10.1016/j.immuni.2005.09.005
[9] Li, W., Kim, M.G., Gourley, T.S., McCarthy, B.P., Sant’
Angelo, D.B. and Chang, C.H. (2005) An alternate path-
way for CD4 T cell development: Thymocyte-expressed
MHC class II selects a distinct T cell population. Immu-
nity, 23, 375-86. doi:10.1016/j.immuni.2005.09.002
[10] Zhu, J. and Paul, W.E. (2010) Heterogeneity and plastic-
ity of T helper cells. Cell Research, 20, 4-12.
doi:10.1038/cr.2009.138
[11] Elgueta, R., de Vries, V.C. and Noelle, R.J. (2010) The
immortality of humoral immunity. The Immunological
Reviews, 236, 139-150.
doi:10.1111/j.1600-065X.2010.00924.x
[12] Annunziato, F., Cosmi, L., Santarlasci, V., Maggi, L., Li-
otta, F., Mazzinghi, B., Parente, E., Fili, L., Ferri, S., Fro-
sali, F., Giudici, F., Romagnani, P., Parronchi, P., Tonelli,
F., Maggi, E. and Romagnani, S. (2007) Phenotypic and
functional features of human Th17 cells. The Jouranl of
Experimental Medicine, 204, 1849-1861.
doi:10.1084/jem.20070663
[13] Bettelli, E., Carrier, Y., Gao, W., Korn, T., Strom, T.B.,
Oukka, M., Weiner, H.L. and Kuchroo, V.K. (2006) Re-
ciprocal developmental pathways for the generation of
pathogenic effector TH17 and regulatory T cells. Nature,
441, 235-238. doi:10.1038/nature04753
[14] Vokaer, B., Van Rompaey, N., Lemaitre, P.H., Lhomme,
F., Kubjak, C., Benghiat, F.S., Iwakura, Y., Petein, M.,
Field, K.A., Goldman, M., Le Moine, A. and Charbonnier,
L.M. (2010) Critical role of regulatory T cells in Th17-
mediated minor antigen-disparate rejection. Journal of
Immunology, 185, 3417-3425.
doi:10.4049/jimmunol.0903961
[15] Crotty, S., Kersh, E.N., Cannons, J., Schwartzberg, P.L.
and Ahmed, R. (2003) SAP is required for generating
long-term humoral immunity. Nature, 421, 282-287.
doi:10.1038/nature01318
[16] Janssen, E.M., Droin NM, Lemmens, E.E., Pinkoski, M.J.,
Bensinger, S.J., Ehst, B.D., Griffith, T.S., Green, D.R. and
Schoenberger, S.P. (2005) CD4+ T-cell help controls
CD8+ T-cell memory via TRAIL-mediated activation-
induced cell death. Nature, 434, 88-93.
doi:10.1038/nature03337
[17] Janssen, E.M., Lemmens, E.E., Wolfe, T., Christen, U.,
von Herrath, M.G. and Schoenberger, S.P. (2003) CD4+ T
cells are required for secondary expansion and memory in
CD8+ T lymphocytes. Nature, 421, 852-856.
doi:10.1038/nature01441
[18] Shedlock, D.J. and Shen, H. (2003) Requirement for CD4
T cell help in generating functional CD8 T cell memory.
Science, 300, 337-339. doi:10.1126/science.1082305
[19] Sun, J.C. and Bevan, M.J. (2003) Defective CD8 T cell
memory following acute infection without CD4 T cell
help. Science, 300, 339-342. doi :10.1126/science.1083317
[20] Vieira, P. and Rajewsky, K. (1990) Persistence of memory
B cells in mice deprived of T cell help. International Im-
munology, 2, 487-494. doi:10.1093/intimm/2.6.487
[21] Li, W., Sofi, M.H., Yeh, N., Sehra, S., McCarthy, B.P.,
Patel, D.R., Brutkiewicz, R.R., Kaplan, M.H. and Chang,
C.H. (2007) Thymic selection pathway regulates the ef-
fector function of CD4 T cells. The Jouranl of Experi-
mental Medicine, 204, 2145-57. doi:10.1084/je m.20070 321
[22] Bendelac, A., Killeen, N., Littman, D.R. andSchwartz,
R.H. (1994) A subset of CD4+ thymocytes selected by
MHC class I molecules. Science, 263, 1774-1778.
doi:10.1126/science.7907820
[23] Budd, R.C., Miescher, G.C., Howe, R.C., Lees, R.K.,
Bron, C. and MacDonald, H.R. (1987) Developmentally
regulated expression of T cell receptor beta chain variable
domains in immature thymocytes. The Jouranl of Experi-
mental Medicine, 166, 577-582.
doi:10.1084/jem.166.2.577
[24] Fowlkes, B.J., Kruisbeek, A.M., Ton-That, H., Weston,
M.A., Coligan, J.E., Schwartz, R.H. and Pardoll, D.M.
(1987) A novel population of T-cell receptor alpha beta-
bearing thymocytes which predominantly expresses a sin-
gle V beta gene family. Nature, 329, 251-254.
doi:10.1038/329251a0
[25] Gapin, L., Matsuda, J.L., Surh, C.D. and Kronenberg, M.
(2001) NKT cells derive from double-positive thymo-
cytes that are positively selected by CD1d. Nature Im-
munology, 2, 971-978. doi:10.1038/ni710
[26] Sofi, M.H., Liu, Z., Zhu, L., Yu, Q., Kaplan, M.H. and-
Chang, C.H. (2010) Regulation of IL-17 expression by
the developmental pathway of CD4 T cells in the thymus.
Molecular Immunology, 47, 1262-1268.
doi:10.1016/j.molimm.2009.12.010
[27] Park, J.H., Chang, S.H., Kim, M.C., Shin, S.H., Youn,
H.J., Kim, J.K., Jang, Y.S. and Kim, C.W. (1998) Up-re-
gulation of the expression of major histocompatibility
complex class I antigens by plasmid DNA transfection in
non-hematopoietic cells. FEBS Letters, 436, 55-60.
doi:10.1016/S0014-5793(98)01097-7
[28] Portnoy, D.A. and Jones, S. (1994) The cell biology of
Listeria monocytogenes infection (escape from a vacuole).
Annals of the New York Academy of Science, 730, 15-25.
doi:10.1111/j.1749-6632.1994.tb44235.x
[29] Portnoy, D.A., Auerbuch, V. and Glomski, I.J. (2002) The
cell biology of Listeria monocytogenes infection: the in-
tersection of bacterial pathogenesis and cell-mediated im-
munity. The Journal of Cell Biology, 158, 409-414.
doi:10.1083/jcb.200205009
[30] Zwickey, H.L. and Potter, T.A. (1999) Antigen secreted

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
38
from noncytosolic Listeria monocytogenes is processed
by the classical MHC class I processing pathway. Journal
of Immunology, 162, 6341-6350.
[31] Harty, J.T. and White, D. (1999) A knockout approach to
understanding CD8+ cell effector mechanisms in adaptive
immunity to Listeria monocytogenes. Immunobiology, 201,
196-204.
[32] Buchmeier, N.A. and Schreiber, R.D. (1985) Requirement
of endogenous interferon-gamma production for resolu-
tion of Listeria monocytogenes infection. Proceedings of
the National Academy of Sciences USA, 82, 7404-7408.
doi:10.1073/pnas.82.21.7404
[33] Basham, T., Smith, W., Lanier, L., Morhenn, V. and Meri-
gan, T. (1984) Regulation of expression of class II major
histocompatibility antigens on human peripheral blood
monocytes and Langerhans cells by interferon. Human
Immunology, 10, 83-93. doi:10.1016/0198-8859(84)90075-2
[34] Finelli, A., Kerksiek, K.M., Allen, S.E., Marshall, N.,
Mercado, R., Pilip, I., Busch, D.H. and Pamer, E.G. (1999)
MHC class I restricted T cell responses to Listeria mono-
cytogenes, an intracellular bacterial pathogen. Immuno-
logic Research, 19, 211-223. doi:10.1007/BF02786489
[35] Kagi, D., Ledermann, B., Burki, K., Hengartner, H. and
Zinkernagel, R.M. (1994) CD8+ T cell-mediated protect-
tion against an intracellular bacterium by perforin-de-
pendent cytotoxicity. European Journal of Immunology,
24, 3068-3072. doi:10.1002/eji.1830241223
[36] Smith, C.M., Wilson, N.S., Waithman, J., Villadangos, J.A.,
Carbone, F.R., Heath, W.R. and Belz, G.T. (2004) Cog-
nate CD4(+) T cell licensing of dendritic cells in CD8(+)
T cell immunity. Nature Immunology, 5, 1143-1148.
doi:10.1038/ni1129
[37] Sun, J.C., Williams, M.A. and Bevan, M.J. (2004) CD4+
T cells are required for the maintenance, not program-
ming, of memory CD8+ T cells after acute infection. Na-
ture Immunology, 5, 927-933. doi:10.1038/ni1105
[38] Shedlock, D.J., Whitmire, J.K., Tan, J., MacDonald, A.S.,
Ahmed, R. and Shen, H. (2003) Role of CD4 T cell help
and costimulation in CD8 T cell responses during Listeria
monocytogenes infection. Journal of Immunology, 170,
2053-2063.
[39] Eaton, K.A., Mefford, M. and Thevenot, T. (2001) The
role of T cell subsets and cytokines in the pathogenesis of
helicobacter pylori gastritis in mice. Journal of Immuno-
logy, 166, 7456-7461.
[40] Eaton, K.A., Ringler, S.R. and Danon, S.J. (1999) Murine
splenocytes induce severe gastritis and delayed-type hy-
persensitivity and suppress bacterial colonization in Heli-
cobacter pylori-infected SCID mice. Infection and Immu-
nity, 67, 4594-4602.
[41] D’Elios, M.M., Manghetti, M., De Carli, M., Costa, F.,
Baldari, C.T., Burroni, D., Telford, J.L., Romagnani, S.
and Del Prete, G. (1997) T helper 1 effector cells specific
for Helicobacter pylori in the gastric antrum of patients
with peptic ulcer disease. Journal of Immunology, 158,
962-967.
[42] Smythies, L.E., Waites, K.B., Lindsey, J.R., Harris, P.R.,
Ghiara, P. and Smith, P.D. (2000) Helicobacter pylori-in-
duced mucosal inflammation is Th1 mediated and exac-
erbated in IL-4, but not IFN-gamma, gene-deficient mice.
Journal of Immunology, 165, 1022-1029.
[43] Eaton, K., Benson, L., Haeger, J., Gray, B. (2006) Role of
transcription factor T-bet expression by CD4+ cells in
gastritis due to Helicobacter pylori in mice. Infection and
Immunity, 74, 4673-4684. doi:10.1128/IAI.01887-05
[44] Pope, C., Kim, S.K., Marzo, A., Masopust, D., Williams,
K., Jiang, J., Shen, H. and Lefrancois, L. (2001) Organ-
specific regulation of the CD8 T cell response to Listeria
monocytogenes infection. Journal of Immunology, 166,
3402-3409.
[45] Sofi, M.H., Qiao, Y., Ansel, K.M., Kubo, M. and Chang,
C.H. (2011) Induction and maintenance of IL-4 expres-
sion are regulated differently by the 3' enhancer in CD4 T
cells. Journal of Immunology, 186, 2792-2799.
doi:10.4049/jimmunol.1003353
[46] Barber, D.L., Wherry, E.J. and Ahmed, R. (2003) Cutting
edge: Rapid in vivo killing by memory CD8 T cells.
Journal of Immunology, 171, 27-31.
[47] Beek, B. (1981) Cell proliferation and chromosomal dam-
age in human leukocytes: Dicentrics and premature
chromosome condensations in first, second, and third
mitoses after X-irradiation. Human Genetics, 57, 75-77.
doi:10.1007/BF00271172
[48] Roos, W.P. and Kaina, B. (2006) DNA damage-induced
cell death by apoptosis. Trends in Molecular Medicine, 12,
440-450. doi:10.1016/j.molmed.2006.07.007
[49] Min, H.S., Lee, Y.J., Jeon, Y.K., Kim, E.J., Kang, B.H.,
Jung, K.C., Chang, C.H. and Park, S.H. (2011) MHC
Class II-restricted interaction between thymocytes plays
an essential role in the production of innate CD8 T cells.
Journal of Immunology, 186, 5749-5757.
doi:10.4049/jimmunol.1002825
[50] Park, W.S., Bae, Y., Chung, D.H., Choi, Y.L., Kim, B.K.,
Sung, Y.C., Choi, E.Y., Park, S.H. and Jung, K.C. (2004)
T cell expression of CIITA represses Th1 immunity. In-
ternational Immunology, 16, 1355-1364.
doi:10.1093/intimm/dxh132
[51] Weinreich, M.A., Odumade, O.A., Jameson, S.C. and Hog-
quist, K.A. (2010) T cells expressing the transcription
factor PLZF regulate the development of memory-like
CD8+ T cells. Nature Immunology, 11, 709-716.
doi:10.1038/ni.1898
[52] Verykokakis, M., Boos, M.D., Bendelac, A. and Kee, B.L.
(2010) SAP protein-dependent natural killer T-like cells
regulate the development of CD8(+) T cells with innate
lymphocyte characteristics. Immunity, 33, 203-215.
doi:10.1016/j.immuni.2010.07.013
[53] Caruso, R., Fina, D., Paoluzi, O.A., Del Vecchio Blanco,
G., Stolfi, C., Rizzo, A., Caprioli, F., Sarra, M., Andrei, F.,
Fantini, M.C., MacDonald, T.T., Pallone, F. and Monte-
leone, G. (2008) IL-23-mediated regulation of IL-17 pro-
duction in Helicobacter pylori-infected gastric mucosa.
European Journal of Immunology, 38, 470-478.
doi:10.1002/eji.200737635
[54] Mizuno, T., Ando, T., Nobata, K., Tsuzuki, T., Maeda, O.,
Watanabe, O., Minami, M., Ina, K., Kusugami, K., Peek,
R.M. and Goto, H. (2005) Interleukin-17 levels in Helico-
bacter pylori-infected gastric mucosa and pathologic se-

Y. Qiao et al. / Open Journal of Immunology 2 (2012) 25-39
39
quelae of colonization. World Journal of Gastroenterol-
ogy, 11, 6305-6311.
[55] Eaton, K.A., Ringler, S.R. and Danon, S.J. (1999) Murine
splenocytes induce severe gastritis and delayed-type hy-
persensitivity and suppress bacterial colonization in Heli-
cobacter pylori-infected SCID mice. Infection and Immu-
nity, 67, 4594-4602.
[56] Matsumoto, Y., Blanchard, T.G., Drakes, M.L., Basu, M.,
Redline, R.W., Levine, A.D. and Czinn, S.J. (2005) Era-
dication of Helicobacter pylori and resolution of gastritis
in the gastric mucosa of IL-10-deficient mice. Helico-
bacter, 10, 407-415.
doi:10.1111/j.1523-5378.2005.00349.x
[57] Marshall, B.J. (1995) Helicobacter pylori in peptic ulcer:
Have Koch’s postulates been fulfilled? Annals of Medi-
cine, 27, 565-568. doi:10.3109/07853899509002470
[58] Wilson, K.T. and Crabtree, J.E. (2007) Immunology of
Helicobacter pylori: Insights into the failure of the immune
response and perspectives on vaccine studies. Gastroen-
terology , 133, 288-308. doi:10.1053/j.gastro.2007.05.008
[59] Park, S.H., Bae, Y.M., Kim, T.J., Ha, I.S., Kim, S.T., Chi,
J.G. and Lee, S.K. (1992) HLA-DR expression in human
fetal thymocytes. Human Immunology, 33, 294-298.
doi:10.1016/0198-8859(92)90338-N
[60] Mold, J.E., Venkatasubrahmanyam, S., Burt, T.D., Micha-
ëlsson, J., Rivera, J.M., Galkina, S.A., Weinberg, K.,
Stoddart, C.A. and McCune, J.M. (2010) Fetal and adult
hematopoietic stem cells give rise to distinct T cell line-
ages in humans. Science, 330, 1695-1699.
[61] PrabhuDas, M., Adkins, B., Gans, H., King, C., Levy, O.,
Ramilo, O. and Siegrist, C.A. (2011) Challenges in in-
fant immunity: Implications for responses to infection
and vaccines. Nature Immunology, 12, 189-194.
doi:10.1038/ni0311-189
[62] Markert, M., Devlin, B., Alexieff, M., Li, J., McCarthy,
E., Gupton, S., Chinn, I., Hale, L., Kepler, T., He, M.,
Sarzotti, M., Skinner, M., Rice, H. and Hoehner, J. (2007)
Review of 54 patients with complete DiGeorge anomaly
enrolled in protocols for thymus transplantation: outcome
of 44 consecutive transplants. Blood, 109, 4539-4547.
doi:10.1182/blood-2006-10-048652
[63] Markert, M.L., Alexieff, M.J., Li, J., Sarzotti, M., Ozaki,
D.A., Devlin, B.H., Sedlak, D.A., Sempowski, G.D., Hale,
L.P., Rice, H.E., Mahaffey, S.M. and Skinner, M.A. (2004)
Postnatal thymus transplantation with immunosuppres-
sion as treatment for DiGeorge syndrome. Blood, 104,
2574-2581. doi:10.1182/blood-2003-08-2984
[64] Markert, M.L., Boeck, A., Hale, L.P., Kloster, A.L., Mc-
Laughlin, T.M., Batchvarova, M.N., Douek, D.C., Koup,
R.A., Kostyu, D.D., Ward, F.E., Rice, H.E., Mahaffey,
S.M., Schiff, S.E., Buckley, R.H. and Haynes, B.F. (1999)
Transplantation of thymus tissue in complete DiGeorge
syndrome. The New England Journal of Medicine, 341,
1180-1189. doi:10.1056/NEJM199910143411603
[65] Markert, M.L., Sarzotti, M., Ozaki, D.A., Sempowski,
G.D., Rhein, M.E., Hale, L.P., Le Deist, F., Alexieff, M.J.,
Li, J., Hauser, E.R., Haynes, B.F., Rice, H.E., Skinner,
M.A., Mahaffey, S.M., Jaggers, J., Stein, L.D. and Mill,
M.R. (2003) Thymus transplantation in complete DiGeorge
syndrome: immunologic and safety evaluations in 12 pa-
tients. Blood, 102, 1121-1130.
doi:10.1182/blood-2002-08-2545